Transient Mark Mode on by default

Transient Mark Mode on by default

[Chong Yidong]

Thanks for all those who contributed to the recent discussion on the
list.  I've discussed this some more with Stefan, and the result is
that Transient Mark Mode is now turned on by default, in the trunk.

I have updated etc/NEWS accordingly.  The documentation now needs to
be updated accordingly.  I am already partway through updating the
Emacs manual, and will check in these changes over the next few days.
When I'm done, I'll inform the list and welcome additional
corrections.

Would anyone like to check the tutorial and the Lisp manual?

Re: Transient Mark Mode on by default

[Tassilo Horn]

Chong Yidong <cyd@stupidchicken.com> writes:

> Would anyone like to check the tutorial

IMO there's nothing wrong in the tutorial.  The only thing I would add
it to mention that the region between mark and point is highlighted.

--8<---------------cut here---------------start------------->8---
Index: etc/tutorials/TUTORIAL
===================================================================
RCS file: /sources/emacs/emacs/etc/tutorials/TUTORIAL,v
retrieving revision 1.3
diff -u -r1.3 TUTORIAL
--- etc/tutorials/TUTORIAL	12 Mar 2008 13:40:40 -0000	1.3
+++ etc/tutorials/TUTORIAL	24 Mar 2008 10:14:03 -0000
@@ -374,7 +374,8 @@
 >> Type C-<SPC>.  Emacs should display a message "Mark set"
    at the bottom of the screen.
 >> Move the cursor to the n in "end", on the second line of the
-   paragraph.
+   paragraph.  The region between the Y and point should be highlighted
+   when doing that.
 >> Type C-w.  This will kill the text starting from the Y,
    and ending just before the n.
 
--8<---------------cut here---------------end--------------->8---

Should I check it in?  If yes, I'll adjust the German tutorial
accordingly.

Bye,
Tassilo

Re: Transient Mark Mode on by default

[Alan Mackenzie]

Hi, Yidong and Emacs!

On Sun, Mar 23, 2008 at 07:00:10PM -0400, Chong Yidong wrote:
> Thanks for all those who contributed to the recent discussion on the
> list.  I've discussed this some more with Stefan, and the result is
> that Transient Mark Mode is now turned on by default, in the trunk.

I feel I must protest here as strongly as I can.  Unless I've missed
something (not unlikely, given the volume and heat of recent posts on the
topic) this change has NOT been discussed properly, and a consensus has
NOT been reached on it.

The recent discussions have been almost entirely about the mechanisms and
the practicalities of Transient Mark Mode, about key sequences, about
which of several variations is better.  NOT about whether it is a Good
Thing as a default.  I feel that this change, making Transient Mark Mode
on by default, is being slipped through almost surreptitiously, at a time
when we're too weary (from the recent discussions) to object.

I say, yet again, Transient Mark Mode is NOT a good default.  It violates
the philosophy of Emacs in several ways:
(i) It's very complicated, certainly when compared with the elegant
  simplicity of the classical Emacs mark.
(ii) It introduces "modal" behaviour (as in vi's insert/command mode)
  into Emacs - many commands behave differently when the mark is active.
(iii) It's obtrusive; it's "in your face"; it will provoke the angry
  reaction "how do I get rid of this #@!!ing thing!!!!!".

Objectively, the mode is inferior to the traditional Emacs mark mechanism
(although I admit "objective" is less important than "subjective" here).

There have been few people indeed who have posted "I think transient mark
mode, as it now is, should be made the default in Emacs".  Dan has,
Stefan has.  Anybody else?  I have opined strongly that Transient Mark
Mode should NOT become default, and I think David Kastrup may have done
the same.  Of "ousiders", Evans Winner says TMM should be off by default,
Jason Earl says it should be on.

I think the vast majority of Emacs hackers have failed to state a firm
position on this, which for such a fundamental change is surely a bad
thing.  May I suggest that you hold a poll of people's views?  And that
if there isn't overwhelming support (whatever that might mean) for this
change of default, that it be reversed?

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Transient Mark Mode on by default

[Juanma Barranquero]

On Mon, Mar 24, 2008 at 12:55 PM, Alan Mackenzie <acm@muc.de> wrote:

>  There have been few people indeed who have posted "I think transient mark
>  mode, as it now is, should be made the default in Emacs".  Dan has,
>  Stefan has.  Anybody else?

FWIW, I think it should be made the default, but still this:

> I've discussed this some more with Stefan, and the result is
> that Transient Mark Mode is now turned on by default, in the trunk.

made me uneasy as well.

 Juanma

Re: Transient Mark Mode on by default

[Tassilo Horn]

Alan Mackenzie <acm@muc.de> writes:

Hi Alan,

> I say, yet again, Transient Mark Mode is NOT a good default.  It violates
> the philosophy of Emacs in several ways:
> (i) It's very complicated, certainly when compared with the elegant
>   simplicity of the classical Emacs mark.
> (ii) It introduces "modal" behaviour (as in vi's insert/command mode)
>   into Emacs - many commands behave differently when the mark is active.
> (iii) It's obtrusive; it's "in your face"; it will provoke the angry
>   reaction "how do I get rid of this #@!!ing thing!!!!!".

For (ii) and (iii) you can set the mark and turing off TMM temporally
with C-SPC C-SPC.  So it's the exact opposite behavior of disabled TMM,
where C-SPC C-SPC enables TMM temporally.

I really like the new default, because most of the time I use marking to
apply commands to the region, so the highlighting is a good thing.  Only
when I set a mark for navigation purposes, I set it with C-SPC C-SPC, so
that the region highlighting doesn't distract me.

Bye,
Tassilo

Re: Transient Mark Mode on by default

[Richard Stallman]

I have not seen a basis for this radical change.

Re: Transient Mark Mode on by default

[Chong Yidong]

Alan Mackenzie <acm@muc.de> writes:

> I feel I must protest here as strongly as I can.  Unless I've missed
> something (not unlikely, given the volume and heat of recent posts on the
> topic) this change has NOT been discussed properly, and a consensus has
> NOT been reached on it.
>
> The recent discussions have been almost entirely about the mechanisms and
> the practicalities of Transient Mark Mode, about key sequences, about
> which of several variations is better.  NOT about whether it is a Good
> Thing as a default.  I feel that this change, making Transient Mark Mode
> on by default, is being slipped through almost surreptitiously, at a time
> when we're too weary (from the recent discussions) to object.

I'm sorry you think that; it was not the intention to "sneak in" the
feature.  I've read the entire discussion on emacs-devel, as well as
some older discussions in help-gnu-emacs and elsewhere.  People on
both sides have already put forth good arguments about why tmm
should/should not be made default, and my impression was that the
discussion has run its course.  So, after some more discussion with
Stefan, I went ahead and changed the default.  If you want a fresh
discussion, let's have that.

For the moment, let's leave tmm turned on in the trunk, at least for a
couple of weeks.  If nothing else, this change might provoke more
people to join the discussion, who may have good points to make.

> I say, yet again, Transient Mark Mode is NOT a good default.  It violates
> the philosophy of Emacs in several ways:
> (i) It's very complicated, certainly when compared with the elegant
>   simplicity of the classical Emacs mark.
> (ii) It introduces "modal" behaviour (as in vi's insert/command mode)
>   into Emacs - many commands behave differently when the mark is active.
> (iii) It's obtrusive; it's "in your face"; it will provoke the angry
>   reaction "how do I get rid of this #@!!ing thing!!!!!".

I am not sure what you mean by (i); could you elaborate?  My
impression is that tmm is simpler than the invisible mark.  Because
the region is highlighted, the user doesn't have to memorize where the
mark is at each point of time.

(ii) Modal behavior has been in default Emacs for a long time now,
e.g. C-s.  I've read a few posts in which people say they find tmm
works pretty seamlessly, which is the important thing.  Tthis is my
experience also.

(iii) This has been answered in a couple of ways.  First, it's easy to
turn tmm off; you don't even have to write Lisp code, you can simply
use the menu bar, under Options->Active Region Highlighting.  Second,
the old invisible mark behavior is available with C-SPC C-SPC
(previously, this was the main sticking point, since the mark is also
useful for buffer navigation.)

> There have been few people indeed who have posted "I think transient mark
> mode, as it now is, should be made the default in Emacs".  Dan has,
> Stefan has.  Anybody else?  I have opined strongly that Transient Mark
> Mode should NOT become default, and I think David Kastrup may have done
> the same.  Of "ousiders", Evans Winner says TMM should be off by default,
> Jason Earl says it should be on.

I don't think head-counting is indicative, since what matters is the
points people make.  But, as far as I can tell, Dan, Stefan, myself,
Drew, and Lennart have spoken in favor of making tmm the default;
Mathias says he finds tmm OK; and yourself, Sascha, and David Kastrup
are against making it the default.  Those who haven't expressed a
strong opinion either way include Eli, Juanma, Miles (who said he
thinks tmm works pretty well) and Kim (who wrote CUA mode and
presumably uses it).  Please correct me if I misrepresented anyone.

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

Chong Yidong wrote:
> (ii) Modal behavior has been in default Emacs for a long time now,
> e.g. C-s.  I've read a few posts in which people say they find tmm
> works pretty seamlessly, which is the important thing.  Tthis is my
> experience also.

And of course the mark.

> I don't think head-counting is indicative, since what matters is the
> points people make.  But, as far as I can tell, Dan, Stefan, myself,
> Drew, and Lennart have spoken in favor of making tmm the default;
> Mathias says he finds tmm OK; and yourself, Sascha, and David Kastrup
> are against making it the default.  Those who haven't expressed a
> strong opinion either way include Eli, Juanma, Miles (who said he
> thinks tmm works pretty well) and Kim (who wrote CUA mode and
> presumably uses it).  Please correct me if I misrepresented anyone.

My only little objection is that cua-selection-mode might have been a 
better choice to start with and then making that as close to tmm as 
possible without breaking cua of course.

Long time goal: make tmm an alias for cua-selection-mode. But starting 
this way you may perhaps reach making cua-selection an alias for tmm ;-)

And the real point is keeping cua compatibility where it is important + 
enhancements. Getting rid of unnecessary complexity.

Re: Transient Mark Mode on by default

[Stefan Monnier]

> I feel I must protest here as strongly as I can.

Just add

  (transient-mark-mode -1)

to your .emacs, just like I and many other people have had to add

  (transient-mark-mode 1)

in theirs in the past.  This has already discussed more than enough, no
consensus was reached and no consensus will ever be reached.


        Stefan

Re: Transient Mark Mode on by default

[Sascha Wilde]

Alan Mackenzie <acm@muc.de> wrote:
> On Sun, Mar 23, 2008 at 07:00:10PM -0400, Chong Yidong wrote:
>> Thanks for all those who contributed to the recent discussion on the
>> list.  I've discussed this some more with Stefan, and the result is
>> that Transient Mark Mode is now turned on by default, in the trunk.
[...]
> I say, yet again, Transient Mark Mode is NOT a good default.  It violates
> the philosophy of Emacs in several ways:
[...]

FWIW: Full Ack.

As I wrote on this topic before:  I was used to put forms into my .emacs
to turn on fancy features that I consider worthwhile -- nowadays I find
my self more and more often fiddling with .emacs to turn annoying
features of.

I liked the philosophy behind the old way better: make the defaults
simple and clean and leave it to the user to turn on all the bells,
whistles and trautonii he likes.

cheers
sascha
-- 
Sascha Wilde
Real programmers don't want "what you see is what you get", they want
"you asked for it, you got it".  They want editors that are terse,
powerful, cryptic, and unforgiving.  In a word, Teco.

Re: Transient Mark Mode on by default

[Chong Yidong]

Sascha Wilde <wilde@sha-bang.de> writes:

> I liked the philosophy behind the old way better: make the defaults
> simple and clean and leave it to the user to turn on all the bells,
> whistles and trautonii he likes.

Only within reason.  For example, we turn Font Lock mode on by
default.  Similarly, Transient Mark mode works very well (due to years
of Emacs hackers turning it on in their own init files), and provides
a real benefit to new users (direct visual feedback).  If there is any
subset of Transient Mark mode behavior that people feel is too
intrusive to be the default, such as commands activate the mark
unnecesarily, we can always tweak that.

Don't worry---we aren't gonna go down the path of XEmacs.

Re: Transient Mark Mode on by default

[paul r]

2008/3/24, Stefan Monnier <monnier@iro.umontreal.ca>:

> Just add
>   (transient-mark-mode -1)
>  to your .emacs, just like I and many other people have had to add (...)

Thanks, Stefan. I hope Chong and you will keep observing this common
sens principle.
   Emacs is fully customizable, but only advanced users *really can*
customize. So please, always make defaults suit what beginners expect,
and leave gurus change them if they want.

1 - Tmm suits better what any beginner expect than default emacs marks.
2 - It can be turned off by advanced users
--> Good choice. Thanks


Paul

Re: Transient Mark Mode on by default

[Alan Mackenzie]

Hi, again!

On Mon, Mar 24, 2008 at 11:05:44AM -0400, Chong Yidong wrote:
> Alan Mackenzie <acm@muc.de> writes:

> > I feel I must protest here as strongly as I can.  Unless I've missed
> > something (not unlikely, given the volume and heat of recent posts on
> > the topic) this change has NOT been discussed properly, and a
> > consensus has NOT been reached on it.

> > The recent discussions have been almost entirely about the mechanisms
> > and the practicalities of Transient Mark Mode, about key sequences,
> > about which of several variations is better.  NOT about whether it is
> > a Good Thing as a default.  I feel that this change, making Transient
> > Mark Mode on by default, is being slipped through almost
> > surreptitiously, at a time when we're too weary (from the recent
> > discussions) to object.

> I'm sorry you think that; it was not the intention to "sneak in" the
> feature.  I've read the entire discussion on emacs-devel, as well as
> some older discussions in help-gnu-emacs and elsewhere.  People on both
> sides have already put forth good arguments about why tmm should/should
> not be made default, and my impression was that the discussion has run
> its course.  So, after some more discussion with Stefan, I went ahead
> and changed the default.  If you want a fresh discussion, let's have
> that.

My impression is that people have said a lot about why the feature is good or
bad, and how it can be improved.  There's a massive difference between
praising/using/loving a feature and advocating it's enablement by default.
For example, I use Hi Lock Mode extensively, think it's one of Emacs's best
features, and would be almost lost without it.  Yet I don't think it should be
enabled by default.

[ .... ]

> > I say, yet again, Transient Mark Mode is NOT a good default.  It violates
> > the philosophy of Emacs in several ways:
> > (i) It's very complicated, certainly when compared with the elegant
> >   simplicity of the classical Emacs mark.
> > (ii) It introduces "modal" behaviour (as in vi's insert/command mode)
> >   into Emacs - many commands behave differently when the mark is active.
> > (iii) It's obtrusive; it's "in your face"; it will provoke the angry
> >   reaction "how do I get rid of this #@!!ing thing!!!!!".

> I am not sure what you mean by (i); could you elaborate?

You can describe the classical mark and region in a single sentence: The
@dfn{region} is the text between point and mark; there are many commands which
operate on it.  Further elaboration is hardly needed.

The corresponding description in Transient Mark Mode is a @bullet list with 8
@items in it (page "Transient Mark" in the Emacs manual), and occupies 41
lines of 72 characters lines (admittedly including paragraph spacing and a
left margin).  This description is well written, containing little redundancy.

That's a factor of between 10 and 100 times more complexity - complexity whose
opposite is conceptual simplicity.

> My impression is that tmm is simpler than the invisible mark.  Because the
> region is highlighted, the user doesn't have to memorize where the mark is
> at each point of time.

The user only needs to know where the mark is occasionally.  A lot of these
times, she will have explicitly set it just before.

Evans Winner, in his post from 2008-02-20 (Message-ID:
<86lk5f4fjb.fsf@timbral.net>) tells us how his unthinking enablement of TMM
prevented him from understanding the versatility of the mark.  It seems to me
(though I may be wrong, of course) that it was the (somewhat arbitrary)
complexity imposed by TMM which disguised the ingenious simplicity of the
mark.

> (ii) Modal behavior has been in default Emacs for a long time now,
> e.g. C-s.

With respect, that is a straw man.  But even if one regards isearch-mode as in
some sense "modal", searching is performed in the minibuffer, and that
"modality" occurs due to switching buffers.  (OK, that was sheer rubbish, but
it was plausible, wasn't it?  ;-)

> I've read a few posts in which people say they find tmm works pretty
> seamlessly, which is the important thing.  This is my experience also.

I don't doubt it works seamlessly, just as vim also does.  However, it's a
psychological jar to find either as default within Emacs, an editor which was
consciously designed to be NON-modal.  It's a bit like having a couple of
Rembrandts in an exhibition of Andy Warhol paintings - no matter how splendid
these Rembrandts may be, they would destroy the conceptual integrity of the
exhibition.

> (iii) This has been answered in a couple of ways.  First, it's easy to
> turn tmm off; you don't even have to write Lisp code, you can simply
> use the menu bar, under Options->Active Region Highlighting.

We're talking about the DEFAULT setup here - the impression a newby will get.
Whether he'll ever "get it" about the mark and region.

> Second, the old invisible mark behavior is available with C-SPC C-SPC
> (previously, this was the main sticking point, since the mark is also useful
> for buffer navigation.)

Again, of course there are workarounds to problems, once people have become
angry enough to be pestered into finding them - a bit like telling a recipient
of spam "you can _just_ delete it - where's the problem?".  It's not that the
"old" behaviour has vanished.  It's that garish distractions are being imposed
on people, many of whom will find it objectionable, without asking them first
if they want it.

> > There have been few people indeed who have posted "I think transient mark
> > mode, as it now is, should be made the default in Emacs".  Dan has, Stefan
> > has.  Anybody else?  I have opined strongly that Transient Mark Mode
> > should NOT become default, and I think David Kastrup may have done the
> > same.  Of "ousiders", Evans Winner says TMM should be off by default,
> > Jason Earl says it should be on.

> I don't think head-counting is indicative, since what matters is the points
> people make.  But, as far as I can tell, Dan, Stefan, myself, Drew, and
> Lennart have spoken in favor of making tmm the default; Mathias says he
> finds tmm OK; and yourself, Sascha, and David Kastrup are against making it
> the default.  Those who haven't expressed a strong opinion either way
> include Eli, Juanma, Miles (who said he thinks tmm works pretty well) and
> Kim (who wrote CUA mode and presumably uses it).  Please correct me if I
> misrepresented anyone.

Of course TMM works pretty well - it's part of Emacs.  Whether it should be on
by default is a completely different matter.

I've grepped my copy of emacs-devel for posts which (i) have "[Tt]ransient" in
the Subject: and (ii) have "default" somewhere in the text.  This doesn't find
your own personal advocacy of TMM as default, though if you assure me you've
said so I'll believe you, of course.  I think Drew and Lennart have _almost_
advocated TMM as default, but not quite.

Of course, I've gently argued against TMM becoming a default, as have Sascha
and David K.

It disturbs me that we are making such an important change when so few Emacs
hackers have unequivocally endorsed it.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Transient Mark Mode on by default

[Sascha Wilde]

Chong Yidong <cyd@stupidchicken.com> wrote:
> Sascha Wilde <wilde@sha-bang.de> writes:
>
>> I liked the philosophy behind the old way better: make the defaults
>> simple and clean and leave it to the user to turn on all the bells,
>> whistles and trautonii he likes.
>
> Only within reason. 

Agreed.

> For example, we turn Font Lock mode on by
> default.

This is an interesting example, as Font Lock mode is actually one of the
features I my self like very much and I used to turn it on in my .emacs
long before it was made default.  But I'm still not perfectly sure, that
making it default was trttd.

> Similarly, Transient Mark mode works very well

One might argue, that Transient Mark mode is way more intrusive than
Font Lock.

> (due to years of Emacs hackers turning it on in their own init files),
> and provides a real benefit to new users (direct visual feedback).

I have to state that I'm very skeptical about that "new users"
argument.  Granted: it will help people coming from usual gui-driven
editors to "feel at home" -- but on the other hand I suspect that many
of this "oh I know this" features prevent new users from learning the
"Emacs way" of things, which IMNSHO is at least worth knowing it.

Further more, I recall from my newbie times, that being forced to look
at the manual for some rather basic features I expected from an editor
like Emacs to have, actually helped me, as it made me come across some
of the less obvious features early.

In short: a user who quickly believes to know how an elaborated tool
works is likely to miss something important.

[...]
> Don't worry---we aren't gonna go down the path of XEmacs.

Sooo good to read this... :-)

cheers
sascha
-- 
Sascha Wilde  :  "I heard that if you play the Windows CD backward, you
              :  get a satanic message. But that's nothing compared to
              :  when you play it forward: It installs Windows...." 
              :  -- G. R. Gaudreau

RE: Transient Mark Mode on by default

[Drew Adams]

> I think Drew and Lennart have _almost_
> advocated TMM as default, but not quite.

Not that it matters much, but I have _explicitly_ advocated TMM as the default
more than once.

To be 100% clear: (1) I am in favor of transient-mark mode being on by default.
(2) I am even in favor of delete-selection mode being on by default. (3) But I
am not in favor of CUA mode or CUA selection mode or PC selection mode being on
by default.

Again, FWIW. Just one opinion. If you want the reasons, see previous posts or
email me off list.

RE: Transient Mark Mode on by default

[Drew Adams]

> Emacs is fully customizable, but only advanced users *really can*
> customize.

If that is true, then we need to improve the customization process. It is OK for
some customizing to be only for advanced users, but it is not good for all
customizing to be only for advanced users.

Can you point out some specific customization problems for a beginner (in a new
thread?) The first step to fixing them is to identify them.

Re: Transient Mark Mode on by default

[paul r]

2008/3/24, Sascha Wilde <wilde@sha-bang.de>:

>  Further more, I recall from my newbie times, that being forced to look
>  at the manual for some rather basic features I expected from an editor
>  like Emacs to have, actually helped me, as it made me come across some
>  of the less obvious features early.

I really agree, because I had the same experience. On the other hand,
5 years ago, I got really frustrated beginning on emacs, so much that
I gave up. I came back to it 2 years later, because someone pushed me,
gave me its .emacs, helped me to get hands on.

I think this, really, is a fundamental question : "On what criterion
should we choose defaults ?"

Here are some of my thoughts :
 -  pushing educational ambition in default settings, really, is a big
mistake. You want to educate, so improve tutorial, improve
documentation, make more interactive tutorials. But do *not*
deliberately harden the way for beginners.
  -  It should *never* be a target make the default set so that gurus
here can minimize the size of their .emacs, ideally having a void
.emacs. Beginners should be comfortable using emacs, even with a void
.emacs. Not advanced users.
  - never compromise the wonderfull ability of emacs to be fully
customizable ( I do not worry too much about this one ;)

What are yours ?

Regards,

-- Paul

RE: Transient Mark Mode on by default

[Drew Adams]

> I have to state that I'm very skeptical about that "new users"
> argument.

I agree. Such an argument can only go so far, and it needs context.

> Granted: it will help people coming from usual gui-driven
> editors to "feel at home" -- but on the other hand I suspect that many
> of this "oh I know this" features prevent new users from learning the
> "Emacs way" of things, which IMNSHO is at least worth knowing it.
> 
> Further more, I recall from my newbie times, that being forced to look
> at the manual for some rather basic features I expected from an editor
> like Emacs to have, actually helped me, as it made me come across some
> of the less obvious features early.
> 
> In short: a user who quickly believes to know how an elaborated tool
> works is likely to miss something important.

I agree.

Personally, I'm in favor of TMM mode or delete-selection mode as the default
because I feel they are (1) close to what newbies are used to, but also (2) good
for many (most?) experienced users as well. 

It is #2 that makes me not be in favor of CUA selection mode or PC selection
mode as the default: my impression is that TMM and delete-selection mode fit the
rest of Emacs better. (I am no expert on the various selection modes, however.)

We should hold out a hand to newbies, but we should also nudge them along the
path to some of the best ways of doing things with Emacs. We should not just
emulate inferior environments that newbies might be used to. We can disagree
about some of those "best ways" to use Emacs, but we might also agree about
some.

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

paul r wrote:
> I really agree, because I had the same experience. On the other hand,
> 5 years ago, I got really frustrated beginning on emacs, so much that
> I gave up. I came back to it 2 years later, because someone pushed me,
> gave me its .emacs, helped me to get hands on.

It is a very, very different situation if you are alone or if you know 
someone who have used Emacs. The complexity is much higher if you are 
all alone.

I think the complexity we meat is not a linear function of the number of 
details. Rather it is an exponential function. Our working memory is 
quite limited and perhaps one can think of this as like our mind have to 
swap between different sets of details if they get too many.

The way our mind works we need to reduce complexity. That is IMO really 
the magic behind mathematics and logics. It reduces complexity so even 
normal human can (more or less) understand things that are complex. 
(Paper and pencil visualisation is another marvelous tool to reduce 
complexity which I tend to use when I get stuck ;-) .)


> I think this, really, is a fundamental question : "On what criterion
> should we choose defaults ?"
> 
> Here are some of my thoughts :
>  -  pushing educational ambition in default settings, really, is a big
> mistake. You want to educate, so improve tutorial, improve
> documentation, make more interactive tutorials. But do *not*
> deliberately harden the way for beginners.
>   -  It should *never* be a target make the default set so that gurus
> here can minimize the size of their .emacs, ideally having a void
> .emacs. Beginners should be comfortable using emacs, even with a void
> .emacs. Not advanced users.

(global-set-key [(control ?c) ?e] (lambda () (interactive) (find-file 
"~/.emacs")))

>   - never compromise the wonderfull ability of emacs to be fully
> customizable ( I do not worry too much about this one ;)
> 
> What are yours ?

I agree with you and with RMS when he wrote:

"In Eclipse, these features are very visible and they "just work".
It would be good if that were true in Emacs also."


> Regards,
> 
> -- Paul
> 
> 
> 

Re: Transient Mark Mode on by default

[paul r]

2008/3/24, Drew Adams <drew.adams@oracle.com>:
> > Emacs is fully customizable, but only advanced users *really can*
>  > customize.
>
> If that is true, then we need to improve the customization process. It is OK for
>  some customizing to be only for advanced users, but it is not good for all
>  customizing to be only for advanced users.
>
>  Can you point out some specific customization problems for a beginner (in a new
>  thread?) The first step to fixing them is to identify them.

Most average developpers I know will simply never tweak any of the
piece of software they use. You are all very smart here, so I'm sure
you are not concerned at all, but really I think a lot of people are.
Some developpers I know will start to tweak their tools after many
monthes of happy usage with default settings.
Almost none of them will start to tweak it *before* really using it.

I think emacs customization process is already fairly good. But many
beginners will simply give up *before* even discovering M-x customize
... sad, but true at least for people I know.

Re: Transient Mark Mode on by default

[Alan Mackenzie]

Hi, Drew!

On Mon, Mar 24, 2008 at 01:32:38PM -0700, Drew Adams wrote:
> > I think Drew and Lennart have _almost_
> > advocated TMM as default, but not quite.

> Not that it matters much, but I have _explicitly_ advocated TMM as the
> default more than once.

Thanks for clarifying, and sorry I missed it earlier.

> To be 100% clear: (1) I am in favor of transient-mark mode being on by default.
> (2) I am even in favor of delete-selection mode being on by default. (3) But I
> am not in favor of CUA mode or CUA selection mode or PC selection mode being on
> by default.

> Again, FWIW. Just one opinion. If you want the reasons, see previous posts or
> email me off list.

-- 
Alan.

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

paul r wrote:
> I think emacs customization process is already fairly good. But many
> beginners will simply give up *before* even discovering M-x customize
> ... sad, but true at least for people I know.

It you are on w32 chances are you may give up quite a lot of times since 
it is IMO more complicated to get a useful Emacs there. That is one 
reason I wrote the installation package for EmacsW32. I still think 
Emacs on w32 should be distributed as an installation package for those 
who wants a precompiled Emacs with the needed extra binaries (grep etc).

RE: Transient Mark Mode on by default

[Drew Adams]

> > > Emacs is fully customizable, but only advanced users *really can*
> > > customize.
> >
> > If that is true, then we need to improve the customization 
> > process. It is OK for some customizing to be only for
> > advanced users, but it is not good for all
> > customizing to be only for advanced users.
> >
> > Can you point out some specific customization problems for 
> > a beginner (in a new
> > thread?) The first step to fixing them is to identify them.
> 
> Most average developpers I know will simply never tweak any of the
> piece of software they use. You are all very smart here, so I'm sure
> you are not concerned at all, but really I think a lot of people are.
> Some developpers I know will start to tweak their tools after many
> monthes of happy usage with default settings.
> Almost none of them will start to tweak it *before* really using it.
> 
> I think emacs customization process is already fairly good. But many
> beginners will simply give up *before* even discovering M-x customize
> ... sad, but true at least for people I know.

In that case, what you meant was not that "only advanced users *really can*
customize" but that "only advanced users *will try to* customize". That is a big
difference.

And I doubt that it is true. Emacs is an "extensible" editor, and sooner or
later users discover this and take advantage of it. Some new users might take
longer to discover that, but I doubt that *only advanced* users customize at
all.

But yes, if a newbie Emacs user is in the habit of *never* changing app
preferences, then s?he might need to discover that Emacs is especially
customizable, and that s?he really might want to customize it in some ways.

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

Lennart Borgman (gmail) wrote:
> paul r wrote:
>> I think emacs customization process is already fairly good. But many
>> beginners will simply give up *before* even discovering M-x customize
>> ... sad, but true at least for people I know.
> 
> It you are on w32 chances are you may give up quite a lot of times since 
> it is IMO more complicated to get a useful Emacs there. That is one 
> reason I wrote the installation package for EmacsW32. I still think 
> Emacs on w32 should be distributed as an installation package for those 
> who wants a precompiled Emacs with the needed extra binaries (grep etc).

Here is a fresh, just some minutes old, example of the difficulties from 
the emacs on windows list. I have seen many like this.

 >>>>>>>>>>>>>>>>>>>>>>>>>
"I am trying to print from Emacs on windows XP. I am using Emacs 22.1.1. 
I am trying to use postscript print through ghostscript. The ghostscript 
version is 8.61.  My .emacs is:

;; Setup Ghostscript 8.61 as the postscript printer

(setenv "GS_LIB" "e:\\Tools\\gs\\gs8.61;e:\\Tools\\gs\\fonts")

(setq ps-lpr-command "e:/Tools/gs/gs8.61/bin/gswin32c")

(setq ps-lpr-switches '("-q" "-dNOPAUSE" "-dBATCH" "-sDEVICE=mswinpr2"))

(setq ps-printer-name t)



When I attempt to print there is no error message and no print – 
nothing. Ghostscript does seem to work – at least through ghostview. It 
is probably a simple erro, but I don’t see it.

thanx"

<<<<<<<<<<<<<<<<<<<<<<<<<<<

Re: Transient Mark Mode on by default

[Alan Mackenzie]

Hi, Yidong!

On Mon, Mar 24, 2008 at 03:09:48PM -0400, Chong Yidong wrote:
> Sascha Wilde <wilde@sha-bang.de> writes:

> > I liked the philosophy behind the old way better: make the defaults
> > simple and clean and leave it to the user to turn on all the bells,
> > whistles and trautonii he likes.

> Only within reason.  For example, we turn Font Lock mode on by
> default.  Similarly, Transient Mark mode works very well (due to years
> of Emacs hackers turning it on in their own init files), and provides
> a real benefit to new users (direct visual feedback).  If there is any
> subset of Transient Mark mode behavior that people feel is too
> intrusive to be the default, such as commands activate the mark
> unnecesarily, we can always tweak that.

Well, this probably isn't a very constructive thing to say, but it is the
highlighting of the region which most disturbs me.  It is brash, it jumps
out from where you least expect it and shouts "GOT YOU!!!  HAH!!!".  It
is impolite, inconsiderate and self-important.  It is violent.  I cannot
work effectively whilst constantly flinching in fear of this anticipated
copper sulphate coloured attack.

How could it be better for people like me?  Well, highlighting the single
character at the mark rather than the whole region would be a massive
improvement (is this perhaps already available as an option?).
show-paren-mode has this as an option, so why not the highlighting in
transient mark mode too?

The other big irritation is that region-face almost, but not quite,
completely obliterates the font-locking in the region - all that remains
visible is whether the font-lock face was bold or not, the foreground
being forced to white, the background to deep blue.  I don't see there is
much that can be done about this on a 16-colour tty.  But on a 2^24
colour GUI system, it would be nice to have a "mute" attribute in a face,
something that would reduce the saturation of the colour it's applied to.
So that red would become a bit pink, pure blue would move towards sky
blue, white would become light grey, black would become dark grey.  I
don't think such a face attribute exists at the moment.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Transient Mark Mode on by default

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 4745 bytes --]

On Mon, 24 Mar 2008 19:40:37 +0100
Sascha Wilde <wilde@sha-bang.de> wrote:

> Alan Mackenzie <acm@muc.de> wrote:
> > On Sun, Mar 23, 2008 at 07:00:10PM -0400, Chong Yidong wrote:
> >> Thanks for all those who contributed to the recent discussion on
> >> the list.  I've discussed this some more with Stefan, and the
> >> result is that Transient Mark Mode is now turned on by default, in
> >> the trunk.
> [...]
> > I say, yet again, Transient Mark Mode is NOT a good default.  It
> > violates the philosophy of Emacs in several ways:
> [...]
> 
> FWIW: Full Ack.

Ditto.

One vital aspect missing from the discussion was the reason why windows/mac 
highlighting works the way it does. Those systems are entirely different:
for starters not having a persistent mark.

Other systems are not necessarily better, any of us have probably seen annoyed users
selecting a region over and over again, because simple inputs kill the 
highlighting. It's not possible to move one of the bounds of the region
while retaining the opposite bounds like you can by swapping the point and mark.

The visual feedback argument also does not sound right to me. Feedback is
useful yes, but the reason you need a region highlighted on garden variety
UI's stems from the fact that most inputs kill the active region. The highlighting
shows the user that they haven't cancelled the region inadvertently with
a command.

Another point is implementation. If it is a transient mark, then it
shouldn't ever interact with a persistent mark ring. That is remarkably
obvious. The fact that the two have become conflated is either poor design
or a implementation shortcut to get the existing body of code working
with active regions.

If Emacs wants to attract new users there is one sure-fire way. Show
them something *better* than what they use right now. Better can
mean: copying (compat), cherry-picking, and innovating. 

When contemplating gifts from the greeks (like active region), prudence
demands that you actually look closely at the gift before bringing it
inside the city.

A person who was serious about finding the Right Thing would have done
a considerable amount of research, contacted the people who invented
the active region (xerox parc ?), wrote comparative analysis sensitive
to historical context/accident, and experimented on live humans.

This TMM looks like blind emulation, poor impedance matching with Emacs,
and as a result many use cases have proliferated modality and other
obvious warning signs. I don't want a full blown FSA diagram to figure
out when my sorta mark, becomes a real mark. especially since authors
of commands who haven't studied this issue closely will likely shim
all sorts of surprising behavior into their commands. Complexity begets
complexity.

I do not oppose the idea of a active region entirely, but I prefer to
read the kind of articulation posted by Thomas Lord, with tables,
analysis, and clear thought. No articulation is perfect, but at least
there was method and a body of thought upon which people could judge
and revise.

The "everyone else is doing it" argument is bogus, but it seems that
must be stated yet again. "Purging" or blood-letting was accepted
medical practice, and many patients survived, but that didn't
magically make it a cure, just common practice. The Right Thing
endures the test of time, common practice becomes the ridicule
of the next generation.

So many programmers clearly think themselves to be unusually rational,
a step above the general population, but the half baked arguments
put forth from both directions validate that clear thought takes
hard work and rigor ; virtues contrary to our nature.

I for one lament that blog quality rhetoric exhausts the discussion,
at which point someone actually does real analysis out of sheer
frustration to assassinate the thread - but by then people's opinions
unreachable by reason from sheer fatigue.

If you share that view of the discussion so far, then the discussion
has just started, as people now begin to realize that their arguments
need weight of reason, and that a quick bum-rush of pontification
has failed to convince the skeptical and carry the day with unanimity.

> As I wrote on this topic before:  I was used to put forms into
> my .emacs to turn on fancy features that I consider worthwhile --
> nowadays I find my self more and more often fiddling with .emacs to
> turn annoying features of.

I have noticed that as well.
 
> I liked the philosophy behind the old way better: make the defaults
> simple and clean and leave it to the user to turn on all the bells,
> whistles and trautonii he likes.
> 
> cheers
> sascha

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Honoring traditional defaults [was: Transient Mark Mode on by default]

[Stephen J. Turnbull]

The only reason not to make Windows/Mac-like behavior the default that
makes sense to me is if we think that traditional Emacs behavior is
*clearly* superior for *most* beginners, enough to make it worth a
short period of confusion and annoyance while they learn to use the
Emacs behaviors.  For something as controversial (and deservedly so)
as t-m-m, I think trying the change at this stage in the release cycle
is a good idea.

On the other hand, the "everything I need to know about Emacs I
learned in kindergarten" crowd *should* have a "revert to tradition"
customization available.  Something like an alist of prior defaults
for customizable variables, having the form ((VAR (VERSION
PRIOR-DEFAULT) ...) ...), where VAR is the symbol naming a
customizable variable, VERSION is a version string identifying a point
of change, and PRIOR-DEFAULT the previous default value.

Then there would be a command `custom-set-all-to-prior-defaults' or
so, which would get a version from the user, defaulting to the prior
public release.  Next, map over the alist of defaults accumulating the
most recent default prior to the user-specified version, if any.  Call
this the "prior defaults alist".  Now the command maps over the prior
defaults alist.  If a variable appears as a key in the prior defaults
alist, and the user has a customization, we ignore it, and continue
with the next variable.  If the user has no customization for the
variable, then we create one, setting the user's customization to the
prior default.

Finally, it emits a warning telling the user which variables it
customized.

If desired, there could also be a customizable variable for
determining how far back to turn the clock, something like
`emacs-version-for-prior-defaults'.  Presumably Alan would set this to
"18.59" or so<wink>.  This would be used instead of the "most recent
public release" as the default for `custom-set-all-to-prior-defaults'.

IMO this handles changes in defaults with a minimum of annoyance to
those with a classical education while making it possible to change
defaults to something more friendly to the GUI generation.

To be honest, I'm not interested in implementing this scheme at this
time, but if and when I get around to it, I'll post here.  If somebody
decides to grab the ball and run with it, I'd appreciate the courtesy
of an email, though.

Re: Transient Mark Mode on by default

[Stephen J. Turnbull]

Chong Yidong writes:

 > Don't worry---we aren't gonna go down the path of XEmacs.

That's a relief!

Re: Transient Mark Mode on by default

[Sascha Wilde]

"paul r" <paul.r.ml@gmail.com> wrote:
> Most average developpers I know will simply never tweak any of the
> piece of software they use.
[...]
> I think emacs customization process is already fairly good. But many
> beginners will simply give up *before* even discovering M-x customize

To be honest: I don't care to much about that kind of people.  There is
a fair amount of software out there to suit there (limited?) needs --
please don't try to make Emacs just another one of these.

I like Emacs for being driven by "doing the right thing" and by that
aiming to become the perfect tool from hackers for hackers[0].

By all means: educate those who are willing to learn:  don't make it
unnecessarily hard for them and point them in the right direction,
e.g. by pointing out that customization is crucial to make Emacs
"yours".

But _pretty please_ don't choose the defaults on what joe average
programmers would expect.

cheers
sascha

[0] btw. I think that a significant number of those folks you are
    talking about will not know the meaning of the word "hacker" in the
    sense I'm using it here.  Coincidence?  I don't think so... ;-)
-- 
Sascha Wilde  :  "The PROPER way to handle HTML postings is to cancel
the article, then hire a hitman to kill the poster, his wife and kids,
and fuck his dog and smash his computer into little bits. Anything
more is just extremism."  -- Paul Tomblin

RE: Transient Mark Mode on by default

[Drew Adams]

> To be 100% clear: (1) I am in favor of transient-mark mode 
> being on by default. (2) I am even in favor of
> delete-selection mode being on by default. (3) But I
> am not in favor of CUA mode or CUA selection mode or PC 
> selection mode being on by default.

I should have also mentioned that I'm OK with temporary TMM as the default also.

RE: Honoring traditional defaults [was: Transient Mark Mode on bydefault]

[Drew Adams]

> On the other hand, the "everything I need to know about Emacs I
> learned in kindergarten" crowd *should* have a "revert to tradition"
> customization available.

Interesting. I first (mis)read that to mean newbies who just wanted to use Emacs
(at least at first) in a way they were used to. IOW, I read the opposite of what
you meant. By "revert to tradition", I at first thought you meant a newbie's
idea of tradition, i.e. what s?he was used to.

Then I read on, and saw what you really meant.

It's interesting (and a bit ironic), because the same approach you propose for
Emacs traditionalists to turn back the clock to a prior Emacs release (Please
just make it work like before!) could also be used to provide alternative
out-of-the-box experiences for Emacs newbies (and others).

That is, provide one or more predefined sets of preference settings. Instead of
the only customization possibility being to dive into the tangled swamp of
Emacs's myriad options, users could choose a suitable macro-level option: a set
of option settings.

The individual low-level options already exist, and a suitable way (e.g. Options
submenu) for users to choose a set of settings could easily be designed. The
hard part might be agreeing on which such sets to provide. But assuming we could
do that without too much trouble, I think it would be a good idea.

Users could then choose among a few predefined Emacs "skins" (though it's more
than skin deep) in, say, the Options menu. Each skin would make a bunch of
settings, such as CUA selection mode, show/hide menus, toolbars, tooltips,...,
whatever. Things that we think newbies might appreciate. And oldbies: Sets that
correspond to the default Emacs behavior for previous releases (what you
described) could be included.

With the possibility of providing more than one skin, just which settings to use
for each skin would be less of a big deal (fight). One of the available skins
would be chosen as the default Emacs behavior. For now, at least, that default
would have the default settings that Emacs already has.

This would provide users with an easy rough cut, to let them quickly get
something more or less suitable. Later, they could fine-tune preferences, like
we all do. This could (1) give users a coarse-grain way to customize, (2)
substitute for some of the here-newbie-start-with-my-dot-emacs that goes around,
and perhaps (3) reduce some of the haggling here over what is TRT to start out
with.

The first task would be to create the infrastructure - something along the lines
of what Stephen proposed. The second task would be to create the UI for choosing
such a set (skin). The third task would be to decide on which sets of which
settings to predefine. The tasks could be done in parallel, if we were sure to
do all three.

It would of course be possible for Lispy users and programs to extend the set of
skins. Color themes are analogous (and a color theme could be a skin component).

WDOT?

> Then there would be a command `custom-set-all-to-prior-defaults' or
> so, which would get a version from the user, defaulting to the prior
> public release.  Next, map over the alist of defaults accumulating the
> most recent default prior to the user-specified version, if any.  Call
> this the "prior defaults alist".  Now the command maps over the prior
> defaults alist.  If a variable appears as a key in the prior defaults
> alist, and the user has a customization, we ignore it, and continue
> with the next variable.  If the user has no customization for the
> variable, then we create one, setting the user's customization to the
> prior default.
> 
> Finally, it emits a warning telling the user which variables it
> customized.
> 
> If desired, there could also be a customizable variable for
> determining how far back to turn the clock, something like
> `emacs-version-for-prior-defaults'.  Presumably Alan would set this to
> "18.59" or so<wink>.  This would be used instead of the "most recent
> public release" as the default for `custom-set-all-to-prior-defaults'.
> 
> IMO this handles changes in defaults with a minimum of annoyance to
> those with a classical education while making it possible to change
> defaults to something more friendly to the GUI generation.
> 
> To be honest, I'm not interested in implementing this scheme at this
> time, but if and when I get around to it, I'll post here.  If somebody
> decides to grab the ball and run with it, I'd appreciate the courtesy
> of an email, though.

Re: Transient Mark Mode on by default

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 1885 bytes --]

On Mon, 24 Mar 2008 23:28:16 +0100
Sascha Wilde <wilde@sha-bang.de> wrote:

> "paul r" <paul.r.ml@gmail.com> wrote:
> > Most average developpers I know will simply never tweak any of the
> > piece of software they use.
> [...]
> > I think emacs customization process is already fairly good. But many
> > beginners will simply give up *before* even discovering M-x
> > customize
> 
> To be honest: I don't care to much about that kind of people.  There
> is a fair amount of software out there to suit there (limited?) needs
> -- please don't try to make Emacs just another one of these.
> 
> I like Emacs for being driven by "doing the right thing" and by that
> aiming to become the perfect tool from hackers for hackers[0].
> 
> By all means: educate those who are willing to learn:  don't make it
> unnecessarily hard for them and point them in the right direction,
> e.g. by pointing out that customization is crucial to make Emacs
> "yours".
> 
> But _pretty please_ don't choose the defaults on what joe average
> programmers would expect.

I second this as well. I also find the entire concept of a "average joe"
to be profoundly wrong. Some people may lack the initiative to search
for a better way, but that does not mean that they do not have the capacity
for it. Let Emacs be a refuge for progress, average is not a legacy.

"
Give me your tired, your poor,
Your huddled masses yearning to breathe free,
The wretched refuse of your teeming shore.
Send these, the homeless, tempest-tossed, to me:
I lift my lamp beside the golden door.
" - Emma Lazarus http://www.bartleby.com/59/6/givemeyourti.html

> cheers
> sascha
> 
> [0] btw. I think that a significant number of those folks you are
>     talking about will not know the meaning of the word "hacker" in
> the sense I'm using it here.  Coincidence?  I don't think so... ;-)

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Transient Mark Mode on by default

[paul r]

2008/3/24, Sascha Wilde <wilde@sha-bang.de>:

> To be honest: I don't care to much about that kind of people.  There is
>  a fair amount of software out there to suit there (limited?) needs --
>  please don't try to make Emacs just another one of these.
>
>  I like Emacs for being driven by "doing the right thing" and by that
>  aiming to become the perfect tool from hackers for hackers[0].
>
>  By all means: educate those who are willing to learn:  don't make it
>  unnecessarily hard for them and point them in the right direction,
>  e.g. by pointing out that customization is crucial to make Emacs
>  "yours".
>
>  But _pretty please_ don't choose the defaults on what joe average
>  programmers would expect.

My point of view is that newbies are precious resource in any project,
and neglecting them is dangerous. For many reasons, emacs would
benefit from having a greater number of beginners and casual users.
They are the connection between inside hackers and the outside world
[1]. A lot of things happen outside emacs world, RMS had an
interesting post about that recently.
Another point is that newbies tend to spend a lot of time expressing
what they like or not like on any project, and more generally
discussing about choices of software. That way, they maintain a
dynamic activity visible from people that do not know emacs yet.
Finally, their point of view might be biased by common software
practices, but at least it is not biased by many many years of
emacs-only usage :)

I do not see any serious threat to have emacs turned into a notepad,
an eclipse, a gedit or a kate, just because its default settings make
it more accessible to the mere mortal.

-- Paul

[1] btw, in this world, people start counting from 1, not from 0 ;)

Re: Honoring traditional defaults

[Sascha Wilde]

"Stephen J. Turnbull" <stephen@xemacs.org> wrote:

> On the other hand, the "everything I need to know about Emacs I
> learned in kindergarten" crowd *should* have a "revert to tradition"
> customization available.

While I actually had fun reading your mail I think your "solution"
points in the wrong direction.  It seems you are assuming the problem is
about a bunch of old farts not willing to change there overcome habits.

But what the problem is really about is some poor youngsters, grown up
in the believe, that operating computers is all about hunting a little
box around your desk and the only detail they disagree about is the
number of buttons that are supposed to be on that little box.

Maybe I'm mistaken, but afaik there is already a special mode for those
spoiled rodent lovers: CUA mode.  And even more: there is already a wide
consensus not to make CUA default.

So, the defaults can be kept the way they are: a useful but non
intrusive starting point for every users customization.

That does not mean that the defaults should never change, it only means
that the main target of the defaults shouldn't be what newbies might
_expect_.  (Note, how that is different from what might be really
_useful_ to newbies.)

cheers
sascha
-- 
Sascha Wilde   -.-. ..- .-. .. --- ... .. - -.-- 
               -.- .. .-.. .-.. . -.. 
               - .... .
               -.-. .- -

Re: Honoring traditional defaults [was: Transient Mark Mode on bydefault]

[paul r]

2008/3/24, Drew Adams <drew.adams@oracle.com>:

>  WDOT?

To me, any "oldie" irritated by a change in default settings can go in
its .emacs and revert to previous behaviour in a matter of seconds.
I'm again under the impression that this idea provides convenience to
experienced users that do not really need it. Newbies need default
settings that encourage them to carry on, this is the real challenge I
think.

RE: Honoring traditional defaults [was: Transient Mark Mode on bydefault]

[Drew Adams]

> >  WDOT?
> 
> To me, any "oldie" irritated by a change in default settings can go in
> its .emacs and revert to previous behaviour in a matter of seconds.
> I'm again under the impression that this idea provides convenience to
> experienced users that do not really need it. Newbies need default
> settings that encourage them to carry on, this is the real challenge I
> think.

Sounds like you didn't read what I wrote. Try again.

Re: Transient Mark Mode on by default

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 2792 bytes --]

On Tue, 25 Mar 2008 00:11:40 +0100
"paul r" <paul.r.ml@gmail.com> wrote:

> 2008/3/24, Sascha Wilde <wilde@sha-bang.de>:
> 
> > To be honest: I don't care to much about that kind of people.
> > There is a fair amount of software out there to suit there
> > (limited?) needs -- please don't try to make Emacs just another one
> > of these.
> >
> >  I like Emacs for being driven by "doing the right thing" and by
> > that aiming to become the perfect tool from hackers for hackers[0].
> >
> >  By all means: educate those who are willing to learn:  don't make
> > it unnecessarily hard for them and point them in the right
> > direction, e.g. by pointing out that customization is crucial to
> > make Emacs "yours".
> >
> >  But _pretty please_ don't choose the defaults on what joe average
> >  programmers would expect.
> 
> My point of view is that newbies are precious resource in any project,
> and neglecting them is dangerous. For many reasons, emacs would
> benefit from having a greater number of beginners and casual users.
> They are the connection between inside hackers and the outside world
> [1]. A lot of things happen outside emacs world, RMS had an
> interesting post about that recently.
> Another point is that newbies tend to spend a lot of time expressing
> what they like or not like on any project, and more generally
> discussing about choices of software. That way, they maintain a
> dynamic activity visible from people that do not know emacs yet.
> Finally, their point of view might be biased by common software
> practices, but at least it is not biased by many many years of
> emacs-only usage :)
> 
> I do not see any serious threat to have emacs turned into a notepad,
> an eclipse, a gedit or a kate, just because its default settings make
> it more accessible to the mere mortal.

1. Emacs use does not confer immortality or superiority.

2. If I want to use eclipse, gedit, or kate, I can start them. I don't
   see how emulation makes Emacs better. In fact I use Eclispe in
   conjunction with Emacs already, solely for RAD tools and source
   analysis that Emacs lacks.

   Making Emacs highlight like eclipse doesn't mean a dingle, Making
   Emacs analyze C, C++, and Java, like other tools do, does.

   Those are the kind of features that compel adoption no matter what
   your opinion of their aesthetics. Solving hard problems is a sure
   way to a larger user base.

3. Newbies aren't a precious resource until donation is mandatory. 
   people who contribute patches are a precious resource.
   Your statement is sound though, IMHO, because contributors all started
   as newbies.

> -- Paul
> 
> [1] btw, in this world, people start counting from 1, not from 0 ;)

irrelevant.
 
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Honoring traditional defaults

[Lennart Borgman (gmail)]

Sascha Wilde wrote:
> Maybe I'm mistaken, but afaik there is already a special mode for those
> spoiled rodent lovers: CUA mode.  And even more: there is already a wide
> consensus not to make CUA default.

The consensus is that it is very difficult to make it the default. Not 
that everyone thinks that it is a bad idea.

> That does not mean that the defaults should never change, it only means
> that the main target of the defaults shouldn't be what newbies might
> _expect_.  (Note, how that is different from what might be really
> _useful_ to newbies.)

You can only think that way if you assume that you know more in a 
certain case.

Re: Transient Mark Mode on by default

[Jason Rumney]

Mike Mattie wrote:
> One vital aspect missing from the discussion was the reason why windows/mac 
> highlighting works the way it does.

Calling it windows/mac highlighting seems to me to be a dishonest way of 
denigrating it. Pretty much every X application other than Emacs 
highlights the region like this.

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

Mike Mattie wrote:
>    Making Emacs highlight like eclipse doesn't mean a dingle, Making
>    Emacs analyze C, C++, and Java, like other tools do, does.


I would be surprised if not a lot of the development of those capacities 
toke place with editors that use CUA key conventions.

Re: Transient Mark Mode on by default

[paul r]

2008/3/25, Mike Mattie <codermattie@gmail.com>:
>  > [1] btw, in this world, people start counting from 1, not from 0 ;)
> irrelevant.

Relevant to anyone reading what I was replying to, I think. But no
offense, that was -simply- a joke.

Re: Transient Mark Mode on by default

[Chong Yidong]

"Stephen J. Turnbull" <stephen@xemacs.org> writes:

> Chong Yidong writes:
>
>  > Don't worry---we aren't gonna go down the path of XEmacs.
>
> That's a relief!

Lest this be misconstrued, what I mean is this: the XEmacs project has
generally been more adventurous about incorporating new packages and
"turning on the bells and whistles", and XEmacs users presumably
appreciate this adventurousness.  In contrast, we've been more careful
about making changes, and this obviously isn't going to change.

Re: Honoring traditional defaults [was: Transient Mark Mode on by default]

[Lennart Borgman (gmail)]

[-- Attachment #1: Type: text/plain, Size: 1627 bytes --]

Stephen J. Turnbull wrote:
> Then there would be a command `custom-set-all-to-prior-defaults' or
> so, which would get a version from the user, defaulting to the prior
> public release.  Next, map over the alist of defaults accumulating the
> most recent default prior to the user-specified version, if any.  Call
> this the "prior defaults alist".  Now the command maps over the prior
> defaults alist.  If a variable appears as a key in the prior defaults
> alist, and the user has a customization, we ignore it, and continue
> with the next variable.  If the user has no customization for the
> variable, then we create one, setting the user's customization to the
> prior default.
> 
> Finally, it emits a warning telling the user which variables it
> customized.
> 
> If desired, there could also be a customizable variable for
> determining how far back to turn the clock, something like
> `emacs-version-for-prior-defaults'.  Presumably Alan would set this to
> "18.59" or so<wink>.  This would be used instead of the "most recent
> public release" as the default for `custom-set-all-to-prior-defaults'.
> 
> IMO this handles changes in defaults with a minimum of annoyance to
> those with a classical education while making it possible to change
> defaults to something more friendly to the GUI generation.
> 
> To be honest, I'm not interested in implementing this scheme at this
> time, but if and when I get around to it, I'll post here.  If somebody
> decides to grab the ball and run with it, I'd appreciate the courtesy
> of an email, though.

I think implementing the scheme is easy. I have attached a starter.


[-- Attachment #2: custsets.el --]
[-- Type: text/plain, Size: 2545 bytes --]

;;; custsets.el --- Sets of named customizations
;;
;; Author: Lennart Borgman (lennart O borgman A gmail O com)
;; Created: 2008-03-25T00:17:06+0100 Mon
;; Version:
;; Last-Updated:
;; URL:
;; Keywords:
;; Compatibility:
;;
;; Features that might be required by this library:
;;
;;   None
;;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;
;;; Commentary:
;;
;; After an idea expressed by among other Stephen Turnbull on the
;; emacs devel list.
;;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;
;;; Change log:
;;
;;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;
;; This program is free software; you can redistribute it and/or
;; modify it under the terms of the GNU General Public License as
;; published by the Free Software Foundation; either version 2, or
;; (at your option) any later version.
;;
;; This program is distributed in the hope that it will be useful,
;; but WITHOUT ANY WARRANTY; without even the implied warranty of
;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
;; General Public License for more details.
;;
;; You should have received a copy of the GNU General Public License
;; along with this program; see the file COPYING.  If not, write to
;; the Free Software Foundation, Inc., 51 Franklin Street, Fifth
;; Floor, Boston, MA 02110-1301, USA.
;;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;
;;; Code:

(defcustom custsets-sets
  '(
    ("Windows"
     (cua-mode t)
     )
    )
  "Sets of customizations."
  :group 'custsets)

(defun custsets-turn-on (set-name)
  (interactive "sCustomization set: ")
  (let ((set (assoc-string set-name custsets-sets t)))
    (unless set
      (error "Can't find customization set %s" set-name))
    (dolist (opt-rec (cdr set))
      (let* ((opt (car opt-rec))
             (val (cdr opt-rec))
             (saved-opt (get opt 'saved-value))
             (saved-val saved-opt) ;; fix-me
             (ask (if saved-opt
                      (format "You have currently customized %s to %s. Change this to %s? "
                              opt saved-opt val)
                    (format "Customize %s to %s? " opt val)))
             )
        (when (y-or-n-p ask)
          (customize-set-variable opt val)
          (customize-set-value opt val)
          (customize-mark-to-save opt))
        )
      )
    (custom-save-all)))


(provide 'custsets)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;; custsets.el ends here

Re: Transient Mark Mode on by default

[Chong Yidong]

Alan Mackenzie <acm@muc.de> writes:

> The other big irritation is that region-face almost, but not quite,
> completely obliterates the font-locking in the region - all that remains
> visible is whether the font-lock face was bold or not, the foreground
> being forced to white, the background to deep blue.  I don't see there is
> much that can be done about this on a 16-colour tty.  But on a 2^24
> colour GUI system, it would be nice to have a "mute" attribute in a face,
> something that would reduce the saturation of the colour it's applied to.
> So that red would become a bit pink, pure blue would move towards sky
> blue, white would become light grey, black would become dark grey.  I
> don't think such a face attribute exists at the moment.

This is a good idea, but difficult to implement.  The face code is
already pretty darn complicated.

You're the only one who's complained about the region obliterating
font-lock, so I don't know how widespread this feeling is.  Maybe
others who are against tmm can chime in about whether they feel
strongly about this?

If this is a "big" objection, here's an idea: make the region face
specify only the background.  Try this (colors chosen as example
only):

(set-face-attribute 'region nil :foreground 'unspecified :background "#333366")
;; for dark backgrounds

(set-face-attribute 'region nil :foreground 'unspecified :background "#DDDDFF")
;; for bright backgrounds

Re: Transient Mark Mode on by default

[Chong Yidong]

Chong Yidong <cyd@stupidchicken.com> writes:

> If this is a "big" objection, here's an idea: make the region face
> specify only the background.

Whoops, ignore that.  Mind wandering.

(This is, of course, the default behavior of the region highlighting.)

Re: Transient Mark Mode on by default

[Thomas Lord]

Jason Rumney wrote:
> Mike Mattie wrote:
>> One vital aspect missing from the discussion was the reason why 
>> windows/mac highlighting works the way it does.
>
> Calling it windows/mac highlighting seems to me to be a dishonest way 
> of denigrating it. Pretty much every X application other than Emacs 
> highlights the region like this.
>
>

That's not the problem. 

The problem is that calling it "windows/mac highlighting"
suggests that it works that way for users -- that it actually
has a "look and feel" that matches those other systems.
It doesn't match those other systems so that's misleading.

-t

Re: Transient Mark Mode on by default

[Jason Rumney]

Thomas Lord wrote:
> That's not the problem.
> The problem is that calling it "windows/mac highlighting"
> suggests that it works that way for users -- that it actually
> has a "look and feel" that matches those other systems.
> It doesn't match those other systems so that's misleading.

Yes, that is a problem. If the reason for enabling TMM by default is to 
make life easier for new users, then I think it is not enough of an 
improvement without also enabling selection with Shift-movement keys. If 
enabling TMM by default upsets many long-time users, then perhaps it is 
better to only enable temporary-TMM when the Shift-movement keys are 
used to select, and leave the behaviour of traditional Emacs keys alone.

Re: Transient Mark Mode on by default

[Bastien]

Chong Yidong <cyd@stupidchicken.com> writes:

> But, as far as I can tell, Dan, Stefan, myself, Drew, and Lennart have
> spoken in favor of making tmm the default; Mathias says he finds tmm
> OK; and yourself, Sascha, and David Kastrup are against making it the
> default.  Those who haven't expressed a strong opinion either way
> include Eli, Juanma, Miles (who said he thinks tmm works pretty well)
> and Kim (who wrote CUA mode and presumably uses it).  Please correct
> me if I misrepresented anyone.

FWIW I don't think Transient Mark Mode should be enabled by default.

-- 
Bastien

Re: Transient Mark Mode on by default

[Bastien]

Alan Mackenzie <acm@muc.de> writes:

> Evans Winner, in his post from 2008-02-20 (Message-ID:
> <86lk5f4fjb.fsf@timbral.net>) tells us how his unthinking enablement
> of TMM prevented him from understanding the versatility of the mark.
> It seems to me (though I may be wrong, of course) that it was the
> (somewhat arbitrary) complexity imposed by TMM which disguised the
> ingenious simplicity of the mark.

Maybe people overlooked this argument so far, but this is a strong one. 

In the same stream of thoughts, if Transient Mark Mode is enabled by
default then beginners (or those who never customize their Emacs) will
always live in this world:

  no highlighted region = no active mark

No sure they will gain something from that.  And C-SPC C-SPC is really
not *that* far, is it?

-- 
Bastien

Re: Transient Mark Mode on by default

[Richard Stallman]

      Similarly, Transient Mark mode works very well (due to years
    of Emacs hackers turning it on in their own init files), and provides
    a real benefit to new users (direct visual feedback).

It is a total pain in the neck unless mark-even-if-inactive is t.

If you are considering a drasticly incompatble change, discussing it
on this list is not enough.  You should poll the users first.  Polling
the users means making an announcement on info-gnu-emacs and
help-gnu-emacs, asking people to try the feature and say what they
think.

Re: Transient Mark Mode on by default

[Bastien]

"paul r" <paul.r.ml@gmail.com> writes:

> 1 - Tmm suits better what any beginner expect than default emacs marks.
> 2 - It can be turned off by advanced users
> --> Good choice. Thanks

I don't follow this reasoning.  More than that: I think _nobody_ really
does -- yes, even you, Paul :)

Remember when you first learned Emacs?  Did it behave like you would
expect a text editor to behave?  At least for me the answer is no, it
behave in a radically new way.  And today I am quite happy Emacs did not
behave like I would have expected it to behave, otherwise it would just
be yet-another-editor.

This is not about gurus/beginners, this is about (learning) paths: is
there a good path from Transient Mark mode to the simple mark?  Is there
a path from the simple mark to Transient Mark mode?  

-- 
Bastien

Re: Honoring traditional defaults

[Bastien]

"Drew Adams" <drew.adams@oracle.com> writes:

> That is, provide one or more predefined sets of preference settings. Instead of
> the only customization possibility being to dive into the tangled swamp of
> Emacs's myriad options, users could choose a suitable macro-level option: a set
> of option settings.

I think this would be useful.  

Especially because it would force us in terms of consistent sets of
options, rather than "defaults for n00bs" vs "defaults for gurus."

If you have a "skin" or a set of options called "RMS-Emacs" I guess 
many newbies will first try this one.  

PS: As for myself, I woud try Alan-Emacs :)

-- 
Bastien

Re: Transient Mark Mode on by default

[Bastien]

Sascha Wilde <wilde@sha-bang.de> writes:

> FWIW: Full Ack.

(defun rotten13 (string)
  (concat (substring string 0 2)
	  (substring string -2)
	  (substring string 4 6)	  
	  (substring string 2 4)))

(rotten13 "Full Ack")

-- 
Baeniest

Re: Honoring traditional defaults

[Bastien]

(Sorry, typo.)

Bastien <bzg@altern.org> writes:

> "Drew Adams" <drew.adams@oracle.com> writes:
>
>> That is, provide one or more predefined sets of preference settings. Instead of
>> the only customization possibility being to dive into the tangled swamp of
>> Emacs's myriad options, users could choose a suitable macro-level option: a set
>> of option settings.
>
> I think this would be useful.  
>
> Especially because it would force us in terms of consistent sets of
                                      ^
                                 [to think]

> options, rather than "defaults for n00bs" vs "defaults for gurus."
>
> If you have a "skin" or a set of options called "RMS-Emacs" I guess 
> many newbies will first try this one.  
>
> PS: As for myself, I woud try Alan-Emacs :)

-- 
Bastien

Re: Transient Mark Mode on by default

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 299 bytes --]

On Tue, 25 Mar 2008 03:12:11 +0100
Bastien <bzg@altern.org> wrote:

> (defun rotten13 (string)
>   (concat (substring string 0 2)
> 	  (substring string -2)
> 	  (substring string 4 6)	  
> 	  (substring string 2 4)))
> 
> (rotten13 "Full Ack")

Lol! Thanks for putting the fun back in!

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Honoring traditional defaults

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 1195 bytes --]

On Tue, 25 Mar 2008 03:06:47 +0100
Bastien <bzg@altern.org> wrote:

> "Drew Adams" <drew.adams@oracle.com> writes:
> 
> > That is, provide one or more predefined sets of preference
> > settings. Instead of the only customization possibility being to
> > dive into the tangled swamp of Emacs's myriad options, users could
> > choose a suitable macro-level option: a set of option settings.
> 
> I think this would be useful.  
> 
> Especially because it would force us in terms of consistent sets of
> options, rather than "defaults for n00bs" vs "defaults for gurus."
> 
> If you have a "skin" or a set of options called "RMS-Emacs" I guess 
> many newbies will first try this one.  
> 
> PS: As for myself, I woud try Alan-Emacs :)
> 

I hacked away on something like this. There is a huge problem with how
much stuff is outside of the Emacs mainline (CEDET), and installation
difficulty.

Not everything belongs in Emacs Core, we really need something like CPAN
with how much essential functionality is floating around without deployment
cohesion.

It would also be a good ground for experiments to be verified before moving
to mainline.

Cheers,
Mike Mattie

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Transient Mark Mode on by default

[Chong Yidong]

Jason Rumney <jasonr@gnu.org> writes:

> If the reason for enabling TMM by default is to make life easier for
> new users, then I think it is not enough of an improvement without
> also enabling selection with Shift-movement keys.

The plan is to add shift selection too.

Re: Transient Mark Mode on by default

[Evans Winner]

Since my earlier post has been referenced in this thread I
thought I should clarify a couple of things:

Though I don't personally use it anymore, I do not have the
expertise to make a suggestion on this list about the
specific merits of transient-mark-mode, or whether it ought
to be enabled by default.  I merely wanted to mention a
possible issue that might be worth considering in any such
case -- that is, whether making something a default was
liable to make it more difficult for new users to discover
and learn to appreciate the specifically Emacs approach.
Obviously there are apt to be reasonable disagreements about
what the specific nature of that approach is, even among
expert users and developers.  There is nothing wrong with
that.

I think it might be relevant to point out that, though I
have no data to support the notion, there are few ``beginner
users'' of Emacs these days in the sense that there might
have been some years ago.  That is, most beginners are not
beginning to learn to use a text editor and that editor is
Emacs -- but rather most, I imagine, are refugees from
various other systems and tools and come to Emacs with a
great deal of ``baggage'' from those other contexts.  When
we are talking about making things easier for the beginner
-- which is a reasonable goal on the face of it -- it might
be important to differentiate between that, and making
things easier for those who have learned to use other
systems, and have expectations based on that.  Because if we
are talking about the later, then a question becomes: To
what degree should Emacs be made to resemble other, and
(most of us would agree) in most respects inferior systems
for the sake of those beginners who would be surprised by
the differences?--and who is really being helped by such
changes?

There are countless free text editors out there and most of
them do the wrong thing, or perhaps even could be said to do
something like the right thing for the many people for whom
their text editor is not a tool of major importance in their
life work.  But Emacs is at least one of the few that still
really caters to the people who want a powerful tool, even
at the /inevitable/ expense of having to spend some time
learning how to use it.

Emacs is, as someone pointed out, not in danger of becoming
a new version of gedit; it could be in danger of becoming a
huge, crufty, inefficient version of gedit, with gobs of
code that no one cares about or maintains that implements
features no one uses anymore because the /culture/ of the
Emacs approach no longer exists or interests those who can
happily and obliviously go on living with the inefficient
habits they've learned from notepad.exe.  If one is truly
concerned with helping new users, then the so-called
``primacy effect'' is the principle to be concerned with --
the principle that people learn most thoroughly that which
they are first exposed to; and that unlearning the wrong
thing is much harder than learning the right thing in the
first place.

Re: Transient Mark Mode on by default

[Stefan Monnier]

>       Similarly, Transient Mark mode works very well (due to years
>     of Emacs hackers turning it on in their own init files), and provides
>     a real benefit to new users (direct visual feedback).

> It is a total pain in the neck unless mark-even-if-inactive is t.

It *is* t.


        Stefan

Re: Honoring traditional defaults

[Stephen J. Turnbull]

Sascha Wilde writes:
 > "Stephen J. Turnbull" <stephen@xemacs.org> wrote:
 > 
 > > On the other hand, the "everything I need to know about Emacs I
 > > learned in kindergarten" crowd *should* have a "revert to tradition"
 > > customization available.
 > 
 > While I actually had fun reading your mail I think your "solution"
 > points in the wrong direction.  It seems you are assuming the problem is
 > about a bunch of old farts not willing to change there overcome habits.

No.  What I'm trying to address is that it is harder to implement a
custom theme engine than it is to implement `revert-to-tradition'.

Note that the idea of a custom theme is at least a decade old with
(rotten13 "Full Ack") success to show.  My idea is simple enough that
in the few hours since I posted it it has already been implemented,
and I'm willing to bet (not having looked at it) that even I will
consider it a pretty faithful implementation. ;-)

RE: Honoring traditional defaults [was: Transient Mark Mode on bydefault]

[Stephen J. Turnbull]

Drew Adams writes:

 > It's interesting (and a bit ironic), because the same approach you
 > propose for Emacs traditionalists to turn back the clock to a prior
 > Emacs release (Please just make it work like before!) could also be
 > used to provide alternative out-of-the-box experiences for Emacs
 > newbies (and others).

It is not that easy, and the proof is trivial: "Please make it work
like before" is a well-defined specification.  "Provide alternative
experiences" is not, as the moribund "custom themes" stuff shows.

RE: Transient Mark Mode on by default

[Drew Adams]

> > [TMM] is a total pain in the neck unless
> > mark-even-if-inactive is t.
> 
> It *is* t.

Really? That's not what I see with emacs -Q.
And then turning on TMM doesn't make it t either.

BTW, I should have mentioned that my vote for TMM as default comes with non-nil
mark-even-if-inactive default also. (And without any automatic activation via
Shift.)

RE: Transient Mark Mode on by default

[Drew Adams]

> > If the reason for enabling TMM by default is to make life easier for
> > new users, then I think it is not enough of an improvement without
> > also enabling selection with Shift-movement keys.
> 
> The plan is to add shift selection too.

Oh, in that case, please remove my vote for it. I want new users to learn about
the mark and C-SPC. TMM works well with Emacs and its notion of a user-set mark,
IMO. It does not dumb Emacs selection down in any way (provided
mark-even-if-inactive is non-nil).

If you want automatic mark activation via Shift by default, then it sounds like
you might want CUA selection mode by default.

I don't want CUA selection mode by default, but I wouldn't mind delete-selection
mode by default. (FWIW, delete-selection mode is what I use.)

Re: Transient Mark Mode on by default

[Jan Djärv]

Richard Stallman skrev:
>       Similarly, Transient Mark mode works very well (due to years
>     of Emacs hackers turning it on in their own init files), and provides
>     a real benefit to new users (direct visual feedback).
> 
> It is a total pain in the neck unless mark-even-if-inactive is t.

mark-even-if-inactive seems to be t by default.

> 
> If you are considering a drasticly incompatble change, discussing it
> on this list is not enough.  You should poll the users first.  Polling
> the users means making an announcement on info-gnu-emacs and
> help-gnu-emacs, asking people to try the feature and say what they
> think.
> 
> 


FWIW, I think t-m-m should be on by default.  A poll may be OK, but if turning 
on t-m-m is for new users (and the new users I've seen really want it on) a 
poll will not reach those users.  IMHO we should take a look at how other 
applications behaves GUI-wise and adjust Emacs defaults to be as similar as it 
can.  A revert-to-classic-emacs option is a good idea, but may be difficult to 
implement.  I guess the opinion on what is classical Emacs differs a lot 
between experienced users :-).


	Jan D.

RE: Honoring traditional defaults [was: Transient Mark Mode onbydefault]

[Drew Adams]

>  > It's interesting (and a bit ironic), because the same approach you
>  > propose for Emacs traditionalists to turn back the clock to a prior
>  > Emacs release (Please just make it work like before!) could also be
>  > used to provide alternative out-of-the-box experiences for Emacs
>  > newbies (and others).
> 
> It is not that easy, and the proof is trivial: "Please make it work
> like before" is a well-defined specification.  "Provide alternative
> experiences" is not, as the moribund "custom themes" stuff shows.

To define a given set of settings, we do not need a well-defined specification.
We need only (1) decide on the mechanism for implementing  sets of settings and
(2) agree on the settings to set and their values.

For options and faces, applying custom-set-variables to an options alist and
similarly for custom-set-faces could be enough for #1. Some customization can be
more complex than setting options and faces, but it's not clear yet that we need
any predefined "skins" that provide such complex customizations.

Wrt #2: Individual preference sets do not need to set the same settings. Any set
of preferences is a candidate for consideration.

RE: Transient Mark Mode on by default

[Drew Adams]

> > > [TMM] is a total pain in the neck unless
> > > mark-even-if-inactive is t.
> > 
> > It *is* t.
> 
> Really? That's not what I see with emacs -Q.
> And then turning on TMM doesn't make it t either.

I should have added:

In GNU Emacs 22.1.90.1 (i386-mingw-nt5.1.2600)
 of 2008-01-30 on PRETEST
Windowing system distributor `Microsoft Corp.', version 5.1.2600
configured using `configure --with-gcc (3.4) --cflags -Ic:/gnuwin32/include'

If this was changed since January, so much the better and sorry for the noise.

Re: Honoring traditional defaults [was: Transient Mark Mode on bydefault]

[Mathias Dahl]

>  To me, any "oldie" irritated by a change in default settings can go in
>  its .emacs and revert to previous behaviour in a matter of seconds.
>  I'm again under the impression that this idea provides convenience to
>  experienced users that do not really need it. Newbies need default
>  settings that encourage them to carry on, this is the real challenge I
>  think.

I agree completely with this, and I consider myself quite "young"
(started using Emacs 1997) compared to a lot of the people on this
list. Now that t-m-m is set as default I don't think I will change it
however, I will give it a try, it isn't like it is the end of the
world, I know I can use C-g to deactivate the region, for example.
Give the new users good defaults and let us "oldies" twiddle away. Of
course, there will always be discussions about what is a good default,
and maybe then we get into the problem with some oldies having such
strong opinions about things and think they know what is best for a
beginner. These things are not easy.

Re: Transient Mark Mode on by default

[Mathias Dahl]

>  This is a good idea

I agree, but it's more like "it would be neat" for me.

>  You're the only one who's complained about the region obliterating
>  font-lock, so I don't know how widespread this feeling is.  Maybe
>  others who are against tmm can chime in about whether they feel
>  strongly about this?

I haven't even noticed it and I guess the reason is that when I have a
highlighted region it is because I want to operate on that region so I
am focused on that rather than if font locking is preserved.

Re: Transient Mark Mode on by default

[Mathias Dahl]

>  The visual feedback argument also does not sound right to me. Feedback is
>  useful yes, but the reason you need a region highlighted on garden variety
>  UI's stems from the fact that most inputs kill the active region. The highlighting
>  shows the user that they haven't cancelled the region inadvertently with
>  a command.

For *me*, the visual feedback is the only reason to use t-m-m.

Re: Transient Mark Mode on by default

[paul r]

2008/3/25, Bastien <bzg@altern.org>:

> I don't follow this reasoning.  More than that: I think _nobody_ really
>  does -- yes, even you, Paul :)
>
>  Remember when you first learned Emacs?  Did it behave like you would
>  expect a text editor to behave?  At least for me the answer is no, it
>  behave in a radically new way.  And today I am quite happy Emacs did not
>  behave like I would have expected it to behave, otherwise it would just
>  be yet-another-editor.

I can remember precisely what I felt. And that is true I learned a lot
from emacs defaults.
But what I'm really concerned about is users giving up after a couple
of weeks because their productivity dropped below the acceptable
level. Those people will not learn *anything* from emacs, because they
simply won't use emacs at all. To put matter into context, I'm from
europe, I'm young as I just finished my studies. In my school,
everybody had to use emacs because it was the editor installed by
default. Today, less than 3% keep using it, and those people use it
because I almost enforced them to do so, and because I put hundreds of
lines in their .emacs. Emacs is full of very smart design decisions,
and I wished any newbie would have enough courage to discover them
all. Or at least I wished a lot would. But I think the learning path,
today, passes through unacceptable points for most people trying
emacs.
So back to my previous question :
--- "Should default settings have educational purpose ?"
-> If yes, then, please, make sure the learning path remains walkable
for most new comers. My opinion is that it isn't in its current state.
-> If no, then, emacs default settings should be chosen to minimize
the productivity drop encoutered when trying emacs.

I do not want default emacs to become a bloated gedit as someone
suggested. But I do not want the user base to lower in number so much
that the future would be really uncertain. A compromise must be found,
don't you think ?

Re: Honoring traditional defaults

[Sascha Wilde]

"Lennart Borgman (gmail)" <lennart.borgman@gmail.com> wrote:
> Sascha Wilde wrote:
>> That does not mean that the defaults should never change, it only means
>> that the main target of the defaults shouldn't be what newbies might
>> _expect_.  (Note, how that is different from what might be really
>> _useful_ to newbies.)
>
> You can only think that way if you assume that you know more in a
> certain case.

True.  But someone knowing both, the concepts and ideas behind "the
Emacs way" and those of common GUI applications (and maybe a bunch of
other non GUI editors) definitely _does_ know more than one who only
knows CUA GUI stuff.

cheers
sascha
-- 
Sascha Wilde 
"If you were young again, would you start writing TeX again or would
you use Microsoft Word, or another word processor?" - "I hope to die
before I *have* to use Microsoft Word."    --  Prof. Donald E. Knuth

Re: Transient Mark Mode on by default

[René Kyllingstad]

* Jan:
>  FWIW, I think t-m-m should be on by default.  A poll may be OK, but if
>  turning on t-m-m is for new users (and the new users I've seen really
>  want it on) a poll will not reach those users.  IMHO we should take a
>  look at how other applications behaves GUI-wise and adjust Emacs
>  defaults to be as similar as it can.  A revert-to-classic-emacs option
>  is a good idea, but may be difficult to implement.  I guess the opinion
>  on what is classical Emacs differs a lot between experienced users :-).

M-x use-emacs-22-defaults, use-emacs-21-defaults, etc ?


-- René

Re: Transient Mark Mode on by default

[Chong Yidong]

"Drew Adams" <drew.adams@oracle.com> writes:

>> > > [TMM] is a total pain in the neck unless
>> > > mark-even-if-inactive is t.
>> > 
>> > It *is* t.
>> 
>> Really? That's not what I see with emacs -Q.
>> And then turning on TMM doesn't make it t either.
>
> I should have added:
>
> In GNU Emacs 22.1.90.1 (i386-mingw-nt5.1.2600)
>  of 2008-01-30 on PRETEST
> Windowing system distributor `Microsoft Corp.', version 5.1.2600
> configured using `configure --with-gcc (3.4) --cflags -Ic:/gnuwin32/include'
>
> If this was changed since January, so much the better and sorry for the noise.

It was turned on in the trunk, but not in the branch.

Re: Transient Mark Mode on by default

[Stefan Monnier]

>> > [TMM] is a total pain in the neck unless
>> > mark-even-if-inactive is t.
>> 
>> It *is* t.

> Really? That's not what I see with emacs -Q.
> And then turning on TMM doesn't make it t either.

TMM *is* on as well, so you're obviously using some unrelated version
of Emacs.


        Stefan

RE: Transient Mark Mode on by default

[Drew Adams]

> TMM *is* on as well, so you're obviously using some unrelated version
> of Emacs.

Yes, the version I have is almost 2 months old. Sorry for the noise.

Re: Transient Mark Mode on by default

[Richard Stallman]

The discussion about this major incompatible change
should not be limited to the people on this list.

RE: Honoring traditional defaults [was: Transient Mark Mode onbydefault]

[Stephen J. Turnbull]

Drew Adams writes:

 > (2) agree

On emacs-devel, agreement to get a beer is NP-hard; for anything that
has to do with code, I refer you to the Loewenheim-Skolem Theorem.
There just aren't enough words ....

Revert to status quo ante is trivial, unless CVS lies.

Re: Honoring traditional defaults [was: Transient Mark Mode on by default]

[Alan Mackenzie]

Hi, Stephen!

On Tue, Mar 25, 2008 at 07:15:10AM +0900, Stephen J. Turnbull wrote:
> The only reason not to make Windows/Mac-like behavior the default that
> makes sense to me is if we think that traditional Emacs behavior is
> *clearly* superior for *most* beginners, enough to make it worth a
> short period of confusion and annoyance while they learn to use the
> Emacs behaviors.  For something as controversial (and deservedly so)
> as t-m-m, I think trying the change at this stage in the release cycle
> is a good idea.

FWIW, I think that the trad Emacs way is clearly superior.  The fact that
I use Emacs enthusiastically bears this out.

> On the other hand, the "everything I need to know about Emacs I
> learned in kindergarten" crowd *should* have a "revert to tradition"
> customization available.  Something like an alist of prior defaults
> for customizable variables, having the form ((VAR (VERSION
> PRIOR-DEFAULT) ...) ...), where VAR is the symbol naming a
> customizable variable, VERSION is a version string identifying a point
> of change, and PRIOR-DEFAULT the previous default value.

The whole point of this thread is (or, at least, should be and was) how
the _default_ Emacs looks, what "emacs -Q" gets you.  The fact that all
of us here can configure Emacs to Alpha Centauri and back again is beside
the mark.

I am not worried what TMM will do to _my_ Emacs - it'll make me add
another line to my .emacs.  I _do_ worry, and worry a lot, that newcomers
might never come to grok the delightful conceptual unity of the Emacs
mark - and that their use of it will thus forever be inefficient.

[ .... ]

> If desired, there could also be a customizable variable for
> determining how far back to turn the clock, something like
> `emacs-version-for-prior-defaults'.  Presumably Alan would set this to
> "18.59" or so<wink>.

No, I don't think so.  18.59 is a bit behind the times.  I'd probably go
for 19.34.   

[ .... ]

:-)

> IMO this handles changes in defaults with a minimum of annoyance to
> those with a classical education while making it possible to change
> defaults to something more friendly to the GUI generation.

See above.  Some of us on this list think that the GUI crowd should be
encouraged to learn some Latin and Greek.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Honoring traditional defaults [was: Transient Mark Mode on by default]

[Lennart Borgman (gmail)]

Alan Mackenzie wrote:
> I _do_ worry, and worry a lot, that newcomers
> might never come to grok the delightful conceptual unity of the Emacs
> mark - and that their use of it will thus forever be inefficient.

Why not cross the bridge and try to invent a way to learn all us coming 
from the faraway cua land that and other secrets? Or is that beyond reality?

Re: Transient Mark Mode on by default

[Richard Stallman]

    For *me*, the visual feedback is the only reason to use t-m-m.

Would you be happy using Transient Mark mode with
`mark-even-if-inactive' set to t?

Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[Alan Mackenzie]

Hi, Drew and Paul R!

On Mon, Mar 24, 2008 at 03:47:38PM -0700, Drew Adams wrote:

[ .... ]

> Users could then choose among a few predefined Emacs "skins" (though
> it's more than skin deep) in, say, the Options menu. Each skin would
> make a bunch of settings, such as CUA selection mode, show/hide menus,
> toolbars, tooltips,..., whatever. Things that we think newbies might
> appreciate. And oldbies: Sets that correspond to the default Emacs
> behavior for previous releases (what you described) could be included.

I've been thinking; what are the requirements here?  I think they're:
(i) It must be _easy_ for a newbie to start an Emacs in "lazy" mode
  (i.e., with the UI stuff from lesser applications enabled).
(ii) The said newby must be made aware that she's started a "dumbed down"
  version of Emacs, and encouraged to switch a standard setup.
(iii) The most standard way of starting emacs (i.e., the command "emacs")
  must start the standard setup.

> With the possibility of providing more than one skin, just which
> settings to use for each skin would be less of a big deal (fight). One
> of the available skins would be chosen as the default Emacs behavior.
> For now, at least, that default would have the default settings that
> Emacs already has.

I think all these things can be achieved with a simple alias:

    % alias emacs_easy='emacs --load /path/to/lisp/emacs-easy.el'

.  emacs-easy.el, besides setting up the "easy" defaults, should display
a startup screen with a message something like:

   "You are running an Emacs configuration designed to be easy to
    _learn_.  When you have become somewhat proficient in its use, you
    may wish to switch to a standard Emacs setup, which is optimised for
    ease of _use_ rather than ease of learning."

What do people think?

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[Lennart Borgman (gmail)]

Alan Mackenzie wrote:

> I've been thinking; what are the requirements here?  I think they're:
> (i) It must be _easy_ for a newbie to start an Emacs in "lazy" mode
>   (i.e., with the UI stuff from lesser applications enabled).

Yes.

> (ii) The said newby must be made aware that she's started a "dumbed down"
>   version of Emacs, and encouraged to switch a standard setup.

Yes.

> (iii) The most standard way of starting emacs (i.e., the command "emacs")
>   must start the standard setup.

No.

> I think all these things can be achieved with a simple alias:
> 
>     % alias emacs_easy='emacs --load /path/to/lisp/emacs-easy.el'

      % alias emacs_uneasy='emacs --load /path/to/lisp/emacs-uneasy.el'

>    "You are running an Emacs configuration designed to be easy to
>     _learn_.  When you have become somewhat proficient in its use, you
>     may wish to switch to a standard Emacs setup, which is optimised for
>     ease of _use_ rather than ease of learning."

"You are running an Emacs configuration designed to be easy to _learn_. 
If you think you are smart enough and want to become a real Emacs user 
then you can switch to an Emacs hero setup which is optimized for those 
who are able to learn it."

Other than that I think the idea is good, but I would prefer Stephens 
way to actually do it.

RE: Honoring traditional defaults - how to do it. [was: Transient MarkMode on bydefault]

[Drew Adams]

> I think all these things can be achieved with a simple alias:
> 
>     % alias emacs_easy='emacs --load /path/to/lisp/emacs-easy.el'
> 
> .  emacs-easy.el, besides setting up the "easy" defaults, 
> should display a startup screen with a message something like...

_None_ of the things I mentioned can be achieved that way, AFAICT.

The point was to provide one or more predefined sets of preference (e.g. option
and face) settings, and let users easily pick such a set from a menu. The
default Emacs behavior would be one such set. The choice would be persistent via
custom-file/.emacs: the chosen set (name and value) would itself be the value of
a user option.

I don't see how any of that would be realized by an alias and a message that
tells users they can customize Emacs. 

Re: Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[paul r]

2008/3/25, Alan Mackenzie <acm@muc.de>:

>  I've been thinking; what are the requirements here?  I think they're:
>  (i) It must be _easy_ for a newbie to start an Emacs in "lazy" mode
>   (i.e., with the UI stuff from lesser applications enabled).
ok
>  (ii) The said newby must be made aware that she's started a "dumbed down"
>   version of Emacs, and encouraged to switch a standard setup.

I think this is acceptable. Emacs(easy) must be made to minimize
surprise, then going from emacs(simple) to emacs(regular) should
remain a doable path, with a tutorial explaining what will change, and
why those changes will increase power. If this is well implemented, I
think it can be a working solution.

>  (iii) The most standard way of starting emacs (i.e., the command "emacs")
>   must start the standard setup.

I do not agree on that point. Again, I think power users must adapt,
not beginners.

>  I think all these things can be achieved with a simple alias:
>
>     % alias emacs_easy='emacs --load /path/to/lisp/emacs-easy.el'
>
>  .  emacs-easy.el, besides setting up the "easy" defaults, should display
>  a startup screen with a message something like:

Could you detail what is the benefit from your pov of an alias over a
"mode", as proposed before ?

>    "You are running an Emacs configuration designed to be easy to
>     _learn_.  When you have become somewhat proficient in its use, you
>     may wish to switch to a standard Emacs setup, which is optimised for
>     ease of _use_ rather than ease of learning."

I like how you formulate the difference between ease of learn and ease of use.

Re: Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[Alan Mackenzie]

Hi, Paul,

On Tue, Mar 25, 2008 at 11:28:55PM +0100, paul r wrote:
> 2008/3/25, Alan Mackenzie <acm@muc.de>:

> >  I've been thinking; what are the requirements here?  I think they're:
> >  (i) It must be _easy_ for a newbie to start an Emacs in "lazy" mode
> >   (i.e., with the UI stuff from lesser applications enabled).
> ok
> >  (ii) The said newby must be made aware that she's started a "dumbed
> >    down" version of Emacs, and encouraged to switch a standard setup.

> I think this is acceptable. Emacs(easy) must be made to minimize
> surprise, then going from emacs(simple) to emacs(regular) should remain
> a doable path, with a tutorial explaining what will change, and why
> those changes will increase power. If this is well implemented, I think
> it can be a working solution.

> >  (iii) The most standard way of starting emacs (i.e., the command
> >  "emacs") must start the standard setup.

> I do not agree on that point. Again, I think power users must adapt,
> not beginners.

I think beginners should be urged to become power users.  Although it's
easier to type "emacs" than "emacs-easy", the difference is so small that
it isn't a nag.

> >  I think all these things can be achieved with a simple alias:

> >     % alias emacs_easy='emacs --load /path/to/lisp/emacs-easy.el'

> >  .  emacs-easy.el, besides setting up the "easy" defaults, should display
> >  a startup screen with a message something like:

> Could you detail what is the benefit from your pov of an alias over a
> "mode", as proposed before ?

It's easier.  There's no configuration to be done, no options to set.
Beginners aren't going to be doing any configuration.  Yet there must be
some slight pressure to urge them eventually to convert to the "standard"
config.  If the beginner first has to insert "(transient-mark-mode -1)"
into her .emacs, whether manually or through `customize-option', she'll
never get round to doing it.

> >    "You are running an Emacs configuration designed to be easy to
> >     _learn_.  When you have become somewhat proficient in its use, you
> >     may wish to switch to a standard Emacs setup, which is optimised for
> >     ease of _use_ rather than ease of learning."

> I like how you formulate the difference between ease of learn and ease
> of use.

Thanks!

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Transient Mark Mode on by default

[Bastien]

"paul r" <paul.r.ml@gmail.com> writes:

> But I think the learning path, today, passes through unacceptable
> points for most people trying emacs.

I think this discussion turns to be a bit too general.

Let's go back to the original question: "Should Transient Mark mode be
turned on by default?"  

Is your claim that the (old) behavior prevents beginners from using
Emacs?  And that the learning curve for clicking on the Active Region
Highlighting menu option is _that_ high?

For sure we can discuss the general policy for this kind of changes.
And by doing this we certainly will reach a form of consensus, but I
doubt agreeing on abstract policies is really useful for the decision 
at stake here.

-- 
Bastien

Re: Honoring traditional defaults - how to do it. [was: Transient MarkMode on bydefault]

[Alan Mackenzie]

Hi, Drew!

On Tue, Mar 25, 2008 at 03:26:57PM -0700, Drew Adams wrote:
> > I think all these things can be achieved with a simple alias:

> >     % alias emacs_easy='emacs --load /path/to/lisp/emacs-easy.el'

> > .  emacs-easy.el, besides setting up the "easy" defaults, 
> > should display a startup screen with a message something like...

> _None_ of the things I mentioned can be achieved that way, AFAICT.

Can't they?  I think they can.

> The point was to provide one or more predefined sets of preference
> (e.g. option and face) settings, and let users easily pick such a set
> from a menu. The default Emacs behavior would be one such set. The
> choice would be persistent via custom-file/.emacs: the chosen set (name
> and value) would itself be the value of a user option.

If the users are beginners, "easily" and "pick from a menu" are mutually
exclusive.  I think my idea could supplement (rather than replace) optons
in a menu.  Extending it a little, the following aliases could all be
defined:

emacs_easy
emacs_fruit         # for an angry fruit salad set of faces
emacs_1934          # for a "classic" set of options
emacs_2001          # to resemble Emacs-21, as far as possible.
emacs_CUA           # for those who love what I hate.

> I don't see how any of that would be realized by an alias and a message that
> tells users they can customize Emacs. 

Why not?  Any option setting can go into emacs-easy.el.  The idea of the
message is not to patronise the users (they already known Emacs can be
customised), but to prod them as gently as possible into using the most
standard, sleekest, puristest, most efficient configuration.

My aim is to soften the painful dilemma we face, that of chosing our
default configuration as either "comfortable for newbies" (when they'll
never get to become power users) or "lean and mean" (when too many of the
newbies will give Emacs up before becoming proficient).

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Transient Mark Mode on by default

[Mathias Dahl]

>     For *me*, the visual feedback is the only reason to use t-m-m.
>
>  Would you be happy using Transient Mark mode with
>  `mark-even-if-inactive' set to t?

Yes, and I do use it otherwise I would go crazy. :)

Re: Honoring traditional defaults - how to do it. [was: Transient MarkMode on bydefault]

[Mathias Dahl]

>  If the users are beginners, "easily" and "pick from a menu" are mutually
>  exclusive.

Could you elaborate on this? I think menus are a good way to find
options and commands as a beginner. It is how I learn new modes in
Emacs (if I have them turned on, it depends on my mood). Are you
referring to the need to use the mouse to click?

>  Why not?  Any option setting can go into emacs-easy.el.  The idea of the
>  message is not to patronise the users (they already known Emacs can be
>  customised), but to prod them as gently as possible into using the most
>  standard, sleekest, puristest, most efficient configuration.

The question then is what is easy to learn and what is easy to use,
when are they the same and when are they not the same? When I first
started using Emacs I immediately enabled `pc-selection-mode' and used
it for many years. Then, for some reason that I cannot remember, I
wanted to try out the, in my opinion back then, more cumbersome style
of using C-SPC + cursor movement. I still like "shift select" however.
although I don't have it enabled in my Emacs because I want the
binding for moving between windows. I use shift select in every other
place I do text editing and I use it a lot and I like it and I cannot
see what harm it would do to have that enabled by default in Emacs
(apart from binding a couple of previously free keys).

Anyway, I did not really mean to discuss this feature in particular,
it is just an example where it is hard to say what is easy to learn
and what is easy to use. I guess it all depends where we come from and
what other preferences we might have.

Then there are all those other words you used, sleek, efficient, pure,
standard... Of those I think the only one we might have a chance to
agree on when it comes to different scenarios is efficient because it
can be measured quite well. Sleekness, pureness and "standardness",
those are hard...

>  My aim is to soften the painful dilemma we face, that of chosing our
>  default configuration as either "comfortable for newbies" (when they'll
>  never get to become power users) or "lean and mean" (when too many of the
>  newbies will give Emacs up before becoming proficient).

It is a good aim, but I agree with Paul that the default should be
"comfortable to newbies", although that is a very loaded expression
(hmm, does that translate from Swedish to English...). We power users
can always use that extra switch or whatever to get the lean and mean
Emacs that we want, if and when we want it. I sometime use emacs -Q to
get a lean and mean Emacs as I have quite a few bells and whistles
(iswitchb, partial completion mode etc etc etc) turned on in my
.emacs. And, before anyone brings it up, having t-m-m turned on by
default in emacs -Q would not make it less mean. Maybe a bit less lean
though...

I better stop now, good night! :)

Re: Honoring traditional defaults [was: Transient Mark Mode on by default]

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 4401 bytes --]

On Tue, 25 Mar 2008 20:53:53 +0000
Alan Mackenzie <acm@muc.de> wrote:

> Hi, Stephen!
> 
> On Tue, Mar 25, 2008 at 07:15:10AM +0900, Stephen J. Turnbull wrote:
> > The only reason not to make Windows/Mac-like behavior the default
> > that makes sense to me is if we think that traditional Emacs
> > behavior is *clearly* superior for *most* beginners, enough to make
> > it worth a short period of confusion and annoyance while they learn
> > to use the Emacs behaviors.  For something as controversial (and
> > deservedly so) as t-m-m, I think trying the change at this stage in
> > the release cycle is a good idea.
> 
> FWIW, I think that the trad Emacs way is clearly superior.  The fact
> that I use Emacs enthusiastically bears this out.
> 
[snip]
> 
> The whole point of this thread is (or, at least, should be and was)
> how the _default_ Emacs looks, what "emacs -Q" gets you.  The fact
> that all of us here can configure Emacs to Alpha Centauri and back
> again is beside the mark.
> 
> I am not worried what TMM will do to _my_ Emacs - it'll make me add
> another line to my .emacs.  I _do_ worry, and worry a lot, that
> newcomers might never come to grok the delightful conceptual unity of
> the Emacs mark - and that their use of it will thus forever be
> inefficient.
> 

Honoring tradition is a very pertinent title for the human side of the issue.
I see it as honoring a *promise* not tradition. When i began to use Emacs
it was hard, even though I was a fairly proficient programmer at the time.

I persisted with Emacs because Emacs was sold to me like this:

* It is insanely hard to learn, but once mastered you will have acquired awesome
  powers and perspective. 

* It might even help you write code faster.

What it actually turned out to be was this:

* It's really hard to overcome the reinforcement of habit in your grey matter.

* Using Emacs to it's full potential carries the price of learning elisp.

* It does make good on it's promise of the Right Thing. Just throw your .emacs into RCS
  and let the itch drive the rest.

Why it matters:

So the new user, takes this promise of the Right Thing on faith, dedicating themselves, their precious free
time to learning Emacs. They climb the mountain to discover what ? The Right Thing wrested from meticulous
thought, passionate debate, and decades of endless polish ?

I certainly hope so, as I was fortunate to receive the bounty of that
promise. As long as there is real value in Emacs - value untarnished by the mediocrity of 
consensus and tribalism (cultural habit) - value appraised from a perspective that holds better 
or worse as objective measures, refugees will find a home in Emacs.

Does adding T-M-M, CUA mode, and brethren keep that promise or forsake it ?

How terrible a thing to contemplate, a student climbing that mountain only to discover yet again thoughtless
concession to the irrational, hows without whys, and design without theme. 

The moment Emacs trades away it's most precious value, neglects it's most important promise to a user, it
becomes just another editor.

The fact that Emacs does things different, the hope that the Emacs way is better, and the celebration of
creativity in elisp is the Emacs pact with the user.

I hold up viper as an example. Emacs delivered so well on that promise that vi people built a emulator for their
editor inside Emacs. Why did they do that ? I think it's because the abstractions hit a incredible design
sweet-spot. Buffers, regions, functions, marks, the point etc.

Freedom of Choice:

Unfortunately this world is full of people who are certain they know what's better for their neighbor than their
neighbor. Emacs does not need to become theocratic. People should be free to create and add what they wish
to Emacs. The core however is *sacred*, the common trust in which we have poured our time and minds. Let us
guard it well with reason, keep the promise. We can encourage freedom, experimentation, and sharing in something
like ELPA. But keep the promise in the core.

Enabling t-m-m by default is endorsing t-m-m as the best way of editing. Do we really think this is the best way to
use emacs ? deserving of the core and privileged default status ? or is it a transition path at best ?

Now that is a question for the community.

Cheers,
Mike Mattie

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Transient Mark Mode on by default

[Richard Stallman]

    > It is a total pain in the neck unless mark-even-if-inactive is t.

    It *is* t.

I am surprised.  The NEWS item doesn't say that, so I assumed it was
nil.

This is not nearly such an incompatible change as I thought it was.
So I don't think it is necessary to poll the users about this.

Re: Honoring traditional defaults [was: Transient Mark Mode on by default]

[Stephen J. Turnbull]

Alan Mackenzie writes:

 > See above.  Some of us on this list think that the GUI crowd should be
 > encouraged to learn some Latin and Greek.

I don't disagree.  I'm just not willing to go to the enormous effort
that I believe is required to do it effectively.

RE: Transient Mark Mode on by default

[Drew Adams]

> Enabling t-m-m by default is endorsing t-m-m as the best way 
> of editing. Do we really think this is the best way to
> use emacs ? deserving of the core and privileged default 
> status ? or is it a transition path at best ?

My own answer is yes (with non-nil mark-even-if-inactive), it is the best way to
use Emacs. No, it is not only a transition path at best.

Likewise for delete-selection mode (which uses t-m-m). I believe either is
better, for both newbies and experienced users, than no notion of active region
and no highlighting. 

The same might be true for temporary t-m-m. IMO, any of these is the right thing
to teach and to learn, from the outset. I don't believe this is true for CUA
selection mode or PC selection mode.

Just one opinion.

[However, you've now moved the topic back into the thread "Transient Mark Mode
on by default". So I changed the Subject back to that.]

Re: Transient Mark Mode on by default

[paul r]

2008/3/26, Bastien <bzg@altern.org>:
>  Is your claim that the (old) behavior prevents beginners from using
>  Emacs?  And that the learning curve for clicking on the Active Region
>  Highlighting menu option is _that_ high?

*Default* is *default*. "Default" is not "easy to find in menu". Most
users will not change any single thing in defaults before having used
the tool for some time. So yes, being real default matters, I think.

Re: Transient Mark Mode on by default

[Chong Yidong]

Richard Stallman <rms@gnu.org> writes:

>     > It is a total pain in the neck unless mark-even-if-inactive is t.
>
>     It *is* t.
>
> I am surprised.  The NEWS item doesn't say that, so I assumed it was
> nil.
>
> This is not nearly such an incompatible change as I thought it was.
> So I don't think it is necessary to poll the users about this.

Thanks for reminding.  I've added a NEWS entry about
mark-even-if-inactive defaulting to t.

Re: Transient Mark Mode on by default

[paul r]

I see a problematic use-case with transient-mark-mode and rectangles.
In the following 3 text lines :
------------
foo
foobar
bar
-----------

It is problematic to cut the whole 3 lines as a rectangle, with
selection active. Precisely, one needs to insert 3 spaces after 'bar',
then mark region, then cut as rectangle.
mark-even-when-inactive allows to mark from foo to bar, then enter
spaces (at this point, region is not highlighted anymore), then C-x r
k, but selection will not be highlighted.
I just wanted to mention that, although I have nothing yet to propose.

Paul

Re: Transient Mark Mode on by default

[Jari Aalto]

* Mon 2008-03-24 Alan Mackenzie <acm@muc.de> gmane.emacs.devel
* Message-Id: 20080324115510.GA1563@muc.de
> Hi, Yidong and Emacs!
>
> On Sun, Mar 23, 2008 at 07:00:10PM -0400, Chong Yidong wrote:
>> that Transient Mark Mode is now turned on by default, in the trunk.
>
> I feel I must protest here as strongly as I can.

Just like the font-lock, the starting users expect certain features to
be default in programs.

This is good news and helps introducing Emacs to the wider audience.
The opposite can be easily be configured in ~/.emacs

Jari

Re: Transient Mark Mode on by default

[Jari Aalto]

* Mon 2008-03-24 Drew Adams <drew.adams@oracle.com> gmane.emacs.devel
* Message-Id: 002301c88df5$5b61d020$c2b22382@us.oracle.com
>> Most average developpers I know will simply never tweak any of the
>> piece of software they use.
>
> and sooner or later users discover this and take advantage of it.
> Some new users might take longer to discover that

Young have more and more difficulties in concentrating and spending
time in reading instructions.

The majority does not customize anything. If the software lacks
expected features (if not turned on by default), they move on to some
other program. Many prefer nano(1) over Emacs, because it displays
command help at the bottim -- no need to memorize anything.

The customizations possibilities are only for the very few, because it
requires certain amount of stamina; to understand by trial and error;
to learn, which possibly spans over long perid of time.

Today's generation is hard to convince to spend that time.

Jari

Re: Honoring traditional defaults [was: Transient Mark Mode on by default]

[Jari Aalto]

* Tue 2008-03-25 Stephen J. Turnbull <stephen@xemacs.org> gmane.emacs.devel
* Message-Id: 874pavg45t.fsf@uwakimon.sk.tsukuba.ac.jp
> The only reason not to make Windows/Mac-like behavior the default that
> makes sense to me is if we think that traditional Emacs behavior is
> *clearly* superior for *most* beginners, enough to make it worth a
> short period of confusion and annoyance while they learn to use the
> Emacs behaviors.

The principle "make them learn" isn't a productive starting point.
This old thinking a is sure way to degrade Emacs to a niche for only
the very enduring people.

As Paul said, we should nor neglect to consider new users as a
valuable new resource. They should be approached with care and with
listening ear.

If they don't get hooked, they will not (and don't want to) spend time
in learning. This is reality today; things might have been a little
different 10-20 years ago in C/C++ era. We're no longer there.

Jari

Re: Transient Mark Mode on by default

[M Jared Finder]

Richard Stallman wrote:
> The discussion about this major incompatible change
> should not be limited to the people on this list.

I have introduced *many* people to Emacs and within the first day, I 
always hear:

"How do I make Cut/Copy/Paste work like every other program?"
"How do I make highlighting work like every other program?"

I then tell them the answer (cua-mode), and 60% of the time they say, 
"Why isn't that the default, LIKE IN EVERY OTHER PROGRAM?"

By then, they're noticeably pissed off and already have a bad feeling 
about Emacs.  About 75% of the people I introduce Emacs to leave, and 
the only reason why is that cut/copy/paste and highlighting is very 
different.  Their experience is otherwise positive, but this awful first 
impression stays with them for the rest of their days.

If you want to keep users coming to Emacs to stay with Emacs, cua-mode 
should be on by default.

   -- MJF

Re: Transient Mark Mode on by default

[Peter Danenberg]

Quoth M Jared Finder on Prickle-Prickle, Discord 16, 3174:
> If you want to keep users coming to Emacs to stay with Emacs,
> cua-mode should be on by default.

transient-mark-mode I could see, sure; but cua-mode introduces all
these bizarre many-to-one mappings with C-c, C-v and C-x.

Re: Transient Mark Mode on by default

[paul r]

2008/3/30, M Jared Finder <jared@hpalace.com>:

>  If you want to keep users coming to Emacs to stay with Emacs, cua-mode
>  should be on by default.

Although I hope your POV is a bit extremist as of today, I think you
stated clearly the general trend. As I previously said between lines,
I'm experiencing similar behaviour of people giving up emacs after a
few days because of unacceptable (to them) drop in their
"productivity".

Re: Transient Mark Mode on by default

[David Kastrup]

M Jared Finder <jared@hpalace.com> writes:

> Richard Stallman wrote:
>> The discussion about this major incompatible change
>> should not be limited to the people on this list.
>
> I have introduced *many* people to Emacs and within the first day, I
> always hear:
>
> "How do I make Cut/Copy/Paste work like every other program?"
> "How do I make highlighting work like every other program?"
>
> I then tell them the answer (cua-mode), and 60% of the time they say,
> "Why isn't that the default, LIKE IN EVERY OTHER PROGRAM?"

Because its keybindings conflict with that of Emacs.

> By then, they're noticeably pissed off and already have a bad feeling
> about Emacs.  About 75% of the people I introduce Emacs to leave, and
> the only reason why is that cut/copy/paste and highlighting is very
> different.

Emacs is different in many other ways, so they would not have stayed for
long, anyway.

> Their experience is otherwise positive, but this awful first
> impression stays with them for the rest of their days.
>
> If you want to keep users coming to Emacs to stay with Emacs, cua-mode
> should be on by default.

Nope.  If you want to keep users coming to Emacs to stay just for a few
days, cua-mode should be on by default.  It will get into your way
eventually because it makes Emacs inconsistent with itself rather than
with other applications, and that's something that does not go away
after getting used to it.  Whereas the difference to other applications
ceases to annoy once you use Emacs for everything.

Anyway, this discussion was about transient-mark-mode, and that does
even less make Emacs behave like other applications with regard to
highlighting and cut&paste.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Transient Mark Mode on by default

[David Kastrup]

"paul r" <paul.r.ml@gmail.com> writes:

> 2008/3/30, M Jared Finder <jared@hpalace.com>:
>
>>  If you want to keep users coming to Emacs to stay with Emacs, cua-mode
>>  should be on by default.
>
> Although I hope your POV is a bit extremist as of today, I think you
> stated clearly the general trend. As I previously said between lines,
> I'm experiencing similar behaviour of people giving up emacs after a
> few days because of unacceptable (to them) drop in their
> "productivity".

cua-mode will not help them much.  Emacs will remain different.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

David Kastrup wrote:
> "paul r" <paul.r.ml@gmail.com> writes:
> 
>> 2008/3/30, M Jared Finder <jared@hpalace.com>:
>>
>>>  If you want to keep users coming to Emacs to stay with Emacs, cua-mode
>>>  should be on by default.
>> Although I hope your POV is a bit extremist as of today, I think you
>> stated clearly the general trend. As I previously said between lines,
>> I'm experiencing similar behaviour of people giving up emacs after a
>> few days because of unacceptable (to them) drop in their
>> "productivity".
> 
> cua-mode will not help them much.  Emacs will remain different.

In contrary I think cua-mode is an excellent help for many beginners. 
That Emacs is different does not mean that cua-mode does not help.

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

David Kastrup wrote:
>> By then, they're noticeably pissed off and already have a bad feeling
>> about Emacs.  About 75% of the people I introduce Emacs to leave, and
>> the only reason why is that cut/copy/paste and highlighting is very
>> different.
> 
> Emacs is different in many other ways, so they would not have stayed for
> long, anyway.

You are meating experience with your own guesses.

Re: Transient Mark Mode on by default

[David Kastrup]

"Lennart Borgman (gmail)" <lennart.borgman@gmail.com> writes:

> David Kastrup wrote:
>> "paul r" <paul.r.ml@gmail.com> writes:
>>
>>> 2008/3/30, M Jared Finder <jared@hpalace.com>:
>>>
>>>>  If you want to keep users coming to Emacs to stay with Emacs, cua-mode
>>>>  should be on by default.
>>> Although I hope your POV is a bit extremist as of today, I think you
>>> stated clearly the general trend. As I previously said between lines,
>>> I'm experiencing similar behaviour of people giving up emacs after a
>>> few days because of unacceptable (to them) drop in their
>>> "productivity".
>>
>> cua-mode will not help them much.  Emacs will remain different.
>
> In contrary I think cua-mode is an excellent help for many
> beginners.

But should we make them beginners forever?

> That Emacs is different does not mean that cua-mode does not help.

What about "much" did you not understand?

At the moment I am getting rather annoyed at the amount of strawmen I
perceive getting thrown at me on the Emacs list, on a variety of topics.
Could be just me, but I don't remember things being that bad.  It's bad
enough if people pick out just a single fragment you wrote to base their
arguments on.  But if they pick up an invention of their own, it is
quite frustrating.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

David Kastrup wrote:
> "Lennart Borgman (gmail)" <lennart.borgman@gmail.com> writes:
> 
>> David Kastrup wrote:
>>> "paul r" <paul.r.ml@gmail.com> writes:
>>>
>>>> 2008/3/30, M Jared Finder <jared@hpalace.com>:
>>>>
>>>>>  If you want to keep users coming to Emacs to stay with Emacs, cua-mode
>>>>>  should be on by default.
>>>> Although I hope your POV is a bit extremist as of today, I think you
>>>> stated clearly the general trend. As I previously said between lines,
>>>> I'm experiencing similar behaviour of people giving up emacs after a
>>>> few days because of unacceptable (to them) drop in their
>>>> "productivity".
>>> cua-mode will not help them much.  Emacs will remain different.
>> In contrary I think cua-mode is an excellent help for many
>> beginners.
> 
> But should we make them beginners forever?

You are making an inappropriate assumption.

>> That Emacs is different does not mean that cua-mode does not help.
> 
> What about "much" did you not understand?
> 
> At the moment I am getting rather annoyed at the amount of strawmen I
> perceive getting thrown at me on the Emacs list, on a variety of topics.
> Could be just me, but I don't remember things being that bad.  It's bad
> enough if people pick out just a single fragment you wrote to base their
> arguments on.  But if they pick up an invention of their own, it is
> quite frustrating.

There was nothing I did not understand.

Re: Transient Mark Mode on by default

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 952 bytes --]

On Sun, 30 Mar 2008 21:47:15 +0200
David Kastrup <dak@gnu.org> wrote:

> "paul r" <paul.r.ml@gmail.com> writes:
> 
> > 2008/3/30, M Jared Finder <jared@hpalace.com>:
> >
> >>  If you want to keep users coming to Emacs to stay with Emacs,
> >> cua-mode should be on by default.
> >
> > Although I hope your POV is a bit extremist as of today, I think you
> > stated clearly the general trend. As I previously said between
> > lines, I'm experiencing similar behaviour of people giving up emacs
> > after a few days because of unacceptable (to them) drop in their
> > "productivity".
> 
> cua-mode will not help them much.  Emacs will remain different.
> 

at least for programmers it will take more than cua mode to match productivity
offered in other environments. Navigating and analyzing code and documentation is 
far more important. I use eclipse, a horrible editor & UI simply for the analysis tools.

Cheers,
Mike Mattie

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Transient Mark Mode on by default

[M Jared Finder]

David d wrote:
> "paul r" <paul.r.ml@gmail.com> writes:
>
>   
>> 2008/3/30, M Jared Finder <jared@hpalace.com>:
>>
>>     
>>>  If you want to keep users coming to Emacs to stay with Emacs, cua-mode
>>>  should be on by default.
>>>       
>> Although I hope your POV is a bit extremist as of today, I think you
>> stated clearly the general trend. As I previously said between lines,
>> I'm experiencing similar behaviour of people giving up emacs after a
>> few days because of unacceptable (to them) drop in their
>> "productivity".
>>     
>
> cua-mode will not help them much.  Emacs will remain different.
>   

I can only speak for my own experience, and I consider myself a pretty 
experienced Emacs user.  I understand keyboard macros, Elisp extensions 
(and have even created a few of my own), the difference between C-f, 
M-f, C-M-f, and such.  I regularly navigate with M-., and enjoy 
displaying the same file in two buffers.  I have a .emacs that is over 
1000 lines.

And I use cua-mode.

You guys all say the experienced user never uses cua-mode, and that the 
newb would be confused because it makes Emacs inconsistent with itself.  
I completely disagree.  If cua-mode did not exist, I *would not be using 
Emacs today*.  Cut/copy/paste is such a fundamental concept to me, that 
learning new keys for them would have been (and STILL IS) intolerable.  
Cua-mode is a joy to use.  99% of the time, it does exactly what I want.

As a solution to this whole problem, why not place an option on the 
splash screen that chooses if cua-mode is on or not?  That way newer 
users will always see the option, right in their face, and it could 
default to being enabled if you see the splash screen.  Once you dismiss 
that screen, it can save the setting in the users .emacs.  And everyone 
here who hates cua-mode (and has already disabled the splash screen) 
will never know the difference.

  -- MJF

Re: Transient Mark Mode on by default

[Thomas Lord]

[-- Attachment #1: Type: text/plain, Size: 1631 bytes --]

David Kastrup wrote:
>
>> I then tell them the answer (cua-mode), and 60% of the time they say,
>> "Why isn't that the default, LIKE IN EVERY OTHER PROGRAM?"
>>     
>
> Because its keybindings conflict with that of Emacs.
>   


See: I am pretty sure that is not really the problem.   User's
can cope easily with something as simple as that.

The problem is the strange way that Emacs tries to mix
its native "marker" concept with the "fat cursor" concept
that is found in other systems.    It's an unnatural mix, imo.

A subtle aspect of Emacs' architecture ca. 18.x is that
there's a simple, ad hoc but nice and comfortable, "state
machine" which is the buffer data structure with markers,
points, etc. -- and then Emacs lisp "scripts" over that and
Emacs lisp programs have the same logical "perspective"
as an interactive user.

The way tmm/cua stuff is getting done, it's drifting into
becoming just a big bundle of hair.   

A tiny tweak to the original simple core is the fat cursor
concept.

A hairy work-around is all this stuff that results because
no 10 users can seem to agree about how best to hork
the original concept of a "mark" to add the concept of
"active/de-activated".

TMM/CUA Announce to users familiar with those other
systems that "Ok, now we have something that works kinda-like
fat cursors but then there's also a ring for saving half of
the information in a fat cursor and sometimes that ring is
and other times it isn't itself the fat cursor and, anyway, there's
a whole bunch of options to figure out when it should do what
which is all cool cause, you know, just fix it in yr .emacs, luser."



-t


[-- Attachment #2: Type: text/html, Size: 2227 bytes --]

Re: Transient Mark Mode on by default

[Thomas Lord]

In general, 90% of the (presumably) desired behavior
of shift-select is from a user perspective "A way to
invoke some other command in a special context."

A tiny number of commands in those other applications
are in any way aware of shift selection but, mostly,
it's just built into how text works.  (At least from a user
perspective.)   Invoke *any* command in the right dynamic
context and the selection DTRT.   So just give the user
a means to invoke that special dynamic context.

The code should reflect that.    It's mostly a way
commands can be invoked, not a property of each
and every command.

-t

Re: Transient Mark Mode on by default

[William Xu]

M Jared Finder <jared@hpalace.com> writes:

> I have introduced *many* people to Emacs and within the first day, I
> always hear:
>
> "How do I make Cut/Copy/Paste work like every other program?"

The Cut/Copy/Paste bindings are already different on different
platforms, for example, on Windows XP, it's C-x/c/v; on Mac OS X, it's
M-x/c/v (M represents the Apple key).  So it's really difficult for
emacs to adjust its complicated bindings to them.  We have to live with
that.

On the contrary, I wonder whether there is any global way configuring
other applications to behave as emacs, since sometimes it also pisses me
off due to this inconsistence.

-- 
William

http://williamxu.net9.org

Re: Transient Mark Mode on by default

[Stephen J. Turnbull]

David Kastrup writes:

 > > In contrary I think cua-mode is an excellent help for many
 > > beginners.
 > 
 > But should we make them beginners forever?

The people who are beginners at any given time are beginners at that
time.  This will be true forever, and at present rates of growth
that's about half as long as it will take Emacs to get a majority
share of editor users.  (I don't think that's an appropriate goal; I
simply propose that there will be plenty of people to recruit to be
beginners forever.)

 > Could be just me, but I don't remember things being that bad

Oh, wow.  I do.  I remember when ESC quit anything, except that ESC
ESC ESC ESC replied, "Buwhahaha!  Can't hack buffers, can you?" and
left you in the minibuffer.

(Actually, I'm making up the Buwhahaha! part.  The rest is true.)

But that's beside the point.  Yes, David, you are different from the
majority of beginning users of editors these days.  That is, they
want their editor to have no learning curve at all.  It should start
out easy, and stay easy, and do things for them.  They do not want to
learn anything about the editor.

Re: Transient Mark Mode on by default

[Stephen J. Turnbull]

William Xu writes:

 > On the contrary, I wonder whether there is any global way configuring
 > other applications to behave as emacs, since sometimes it also pisses me
 > off due to this inconsistence.

Bash and zsh both have Emacs keybinding themes.  There is some kind of
add-on for Firefox (and maybe other Mozilla-family products) that
emulates Emacs bindings.  I believe the ae editor has an Emacs
keybinding suite (maybe as a contrib feature).

Re: Transient Mark Mode on by default

[William Xu]

"Stephen J. Turnbull" <stephen@xemacs.org> writes:

> Bash and zsh both have Emacs keybinding themes.  There is some kind of
> add-on for Firefox (and maybe other Mozilla-family products) that
> emulates Emacs bindings.  I believe the ae editor has an Emacs
> keybinding suite (maybe as a contrib feature).

Actually I'm only concerned with Cut/Copy/Paste bindings, looking for a
system level way, not application by application.  

-- 
William

http://williamxu.net9.org

Re: Transient Mark Mode on by default

[David Kastrup]

M Jared Finder <jared@hpalace.com> writes:

> And I use cua-mode.
>
> You guys all say the experienced user never uses cua-mode,

No, we don't.  What is it in this thread that makes people invent straw
men all the time?

> and that the newb would be confused because it makes Emacs
> inconsistent with itself.  I completely disagree.  If cua-mode did not
> exist, I *would not be using Emacs today*.

So you are a newbie?

> As a solution to this whole problem, why not place an option on the
> splash screen that chooses if cua-mode is on or not?

Because the splash screen is optional and not the right place for it:
people might no longer find it.  CUA mode is right in the "Options" menu
where people looking for it can find it.

> That way newer users will always see the option, right in their face,
> and it could default to being enabled if you see the splash screen.

A user that is incapable of clicking on "Options" for setting options is
not the Emacs target audience.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Transient Mark Mode on by default

[David Kastrup]

"Stephen J. Turnbull" <stephen@xemacs.org> writes:

> David Kastrup writes:
>
>  > > In contrary I think cua-mode is an excellent help for many
>  > > beginners.
>  > 
>  > But should we make them beginners forever?
>
> The people who are beginners at any given time are beginners at that
> time.  This will be true forever, and at present rates of growth
> that's about half as long as it will take Emacs to get a majority
> share of editor users.  (I don't think that's an appropriate goal; I
> simply propose that there will be plenty of people to recruit to be
> beginners forever.)
>
>  > Could be just me, but I don't remember things being that bad
>
> Oh, wow.  I do.  I remember when ESC quit anything, except that ESC
> ESC ESC ESC replied, "Buwhahaha!  Can't hack buffers, can you?" and
> left you in the minibuffer.

"that bad" meaning the fabrication of ad-hoc straw men in list
discussions.  Combined with creative editing, you make another one here.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Transient Mark Mode on by default

[M Jared Finder]

David Kastrup wrote:
> M Jared Finder <jared@hpalace.com> writes:
>
>   
>> And I use cua-mode.
>>
>> You guys all say the experienced user never uses cua-mode,
>>     
>
> No, we don't.  What is it in this thread that makes people invent straw
> men all the time?
>
>   
>> and that the newb would be confused because it makes Emacs
>> inconsistent with itself.  I completely disagree.  If cua-mode did not
>> exist, I *would not be using Emacs today*.
>>     
>
> So you are a newbie?
>   
Repeated misunderstandings.  Is your definition of newb == someone who 
uses cua-mode?

I would not be using Emacs today, because CUA mode prevents me from 
having to context-switch in my brain from C-c/C-v/C-z to whatever Emacs 
maps those keys to.  Unlike some of you, I can not use Emacs for 
everything, and having to context switch between "using Emacs" and 
"using everything else" was too difficult for me.

>> As a solution to this whole problem, why not place an option on the
>> splash screen that chooses if cua-mode is on or not?
>>     
>
> Because the splash screen is optional and not the right place for it:
> people might no longer find it.  CUA mode is right in the "Options" menu
> where people looking for it can find it.
>   

No, you are wrong.  The splash screen is the perfect place to place 
information about CUA mode, in addition to the options menu.  One does 
not exclude the other.

Who wants CUA mode the most?  Newbies.
Who ends up actually looking at the splash screen?  People who have not 
disabled it explicity, which is a superset of newbies.

I imagine a world where the splash screen is like the "Did you know" 
popup in Gimp.  Ideally, you could place one more entry, "CUA mode 
allows you to use C-x/C-c/C-v/C-z for cut/copy/paste/undo.  [TRY IT 
NOW]", and clicking TRY IT NOW would enable cua-mode.  What's the 
disadvantage of this?  Because I can tell you MANY advantages.

  -- MJF

Re: Transient Mark Mode on by default

[David Kastrup]

M Jared Finder <jared@hpalace.com> writes:

> David Kastrup wrote:
>>
>> So you are a newbie?
>>   
> Repeated misunderstandings.  Is your definition of newb == someone who
> uses cua-mode?

It seemingly is your definition since you use yourself as an example of
the hypothetical "newbie".

> I would not be using Emacs today, because CUA mode prevents me from
> having to context-switch in my brain from C-c/C-v/C-z to whatever
> Emacs maps those keys to.

You have been posting on XEmacs/Emacs groups and lists much much longer
than cua-mode even existed.

>>> As a solution to this whole problem, why not place an option on the
>>> splash screen that chooses if cua-mode is on or not?
>>
>> Because the splash screen is optional and not the right place for it:
>> people might no longer find it.  CUA mode is right in the "Options" menu
>> where people looking for it can find it.
>
> No, you are wrong.  The splash screen is the perfect place to place
> information about CUA mode, in addition to the options menu.

We have had months of discussions about what to put on the splash screen
and what not.  You have not participated.

> I imagine a world where the splash screen is like the "Did you know"
> popup in Gimp.

Popups like that get turned off by everybody I know since they offer the
wrong information at the wrong time.

> Ideally, you could place one more entry, "CUA mode allows you to use
> C-x/C-c/C-v/C-z for cut/copy/paste/undo.  [TRY IT NOW]", and clicking
> TRY IT NOW would enable cua-mode.  What's the disadvantage of this?
> Because I can tell you MANY advantages.

Read the months of discussions on the splash screen.  The decisions of
what to put there are not arbitrary.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Transient Mark Mode on by default

[Jason Rumney]

Stephen J. Turnbull wrote:
> But that's beside the point.  Yes, David, you are different from the
> majority of beginning users of editors these days.  That is, they
> want their editor to have no learning curve at all.  It should start
> out easy, and stay easy, and do things for them.  They do not want to
> learn anything about the editor.
>   

I don't think it is true that they never want to learn anything, there 
would be no point in switching to anything from notepad if that was the 
case. But beginning users expect to be able to start doing basic text 
editing without having to learn anything, and gradually learn how to 
make better use of their chosen tool.

I started using CUA mode many years ago, for the opposite reason of  I 
expect most CUA users. Emacs keybindings were so ingrained that I had 
trouble using other software, so I decided to train myself to use the 
CUA bindings. I have honestly never encountered a problem with the CUA 
keys getting in the way (though I always used PgDn rather than C-v, I 
could see that one being a problem for some Emacs users).

Re: Transient Mark Mode on by default

[Stephen J. Turnbull]

David Kastrup writes:

 > "that bad" meaning the fabrication of ad-hoc straw men in list
 > discussions.

Really?![1]  But let's put the context back:

    > That Emacs is different does not mean that cua-mode does not help.

    What about "much" did you not understand?

    At the moment I am getting rather annoyed at the amount of strawmen I
    perceive getting thrown at me on the Emacs list, on a variety of topics.
    Could be just me, but I don't remember things being that bad.

There's nothing to understand about Emacs being "much" different in
other ways here; that's a strawman itself.

Jared is saying that *no matter how different Emacs is* in other ways,
by giving a few basic operations the same bindings in Emacs as in
other apps he uses, CUA-mode helps him to use Emacs.  Further, in his
case it was (when he was a newbie) and is (now that he is more expert)
sufficient to keep him using Emacs.  And it is implied that, even with
his current Emacs skills, without CUA he might not be using Emacs today.

IMO, this is the kind of opinion that Richard says should be included
in polls.  Jared explained why *he* feels that way, and says that he's
not unique.  You can believe him or not as you like, but I see no
strawman in Jared's post at all.


Footnotes: 
[1]  I would express what you meant with "Could be just me, but I
don't remember things *ever* being *this* bad."  (Stars highlight
differences, not for emphasis.)  The difference is slight, but to me
it is all the difference in the world.  Just like CUA mode is to Jared!

Re: Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[Richard Stallman]

    >    "You are running an Emacs configuration designed to be easy to
    >     _learn_.  When you have become somewhat proficient in its use, you
    >     may wish to switch to a standard Emacs setup, which is optimised for
    >     ease of _use_ rather than ease of learning."

    I like how you formulate the difference between ease of learn and ease of use.

Making Emacs easier to learn is an improvement, in general.
But what use is there in having a different Emacs configuration
which is easier to learn, if it is not the recommended mode of use?

That seems like telling people a shortcut leading to somewhere other
than their destination.

Re: Transient Mark Mode on by default

[Richard Stallman]

Shift-selection is fine, but I don't think we should change the
meaning of C-c, C-v and C-x.

All else being equal, it would be better to be compatible with other
programs, including in this.  But all else is not equal, and this
change would not fit into Emacs.

Re: Transient Mark Mode on by default

[Thomas Lord]

Back when people more often disagreed about what
backspace characters should mean, keyboard translation
maps helped a lot.

Here is an idea that might be simple and clean:

Let's pretend that there are three special keys on every keyboard.
These are named COPY, PASTE, and CUT.

Well, actually, on real keyboards there is no way to type those
but, through the magic of key translation, users can assign those
keys wherever they like.   So, put CUT on C-x and put C-x on F1,
say.

That doesn't give a "modal" system where sometimes C-v means
scroll-down and other times it means PASTE.   But it gives a simple
way to make those keys work how users expect with the side effect
that if a user (who has remapped those keys) types, say, C-H C-v
the system says "PASTE is bound yank [or whatever]".

Emacs documentation will still be saying things like "Use C-x f to
open a file."    That's a burden on new users who elect to remap
C-x to CUT (and some other key to C-x).    But it's a small burden
because it's just those few keys and the rules about how to type those
characters apply consistently, all the time.

-t


Richard Stallman wrote:
> Shift-selection is fine, but I don't think we should change the
> meaning of C-c, C-v and C-x.
>
> All else being equal, it would be better to be compatible with other
> programs, including in this.  But all else is not equal, and this
> change would not fit into Emacs.
>
>
>
>
>   

Re: Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[Alan Mackenzie]

'Evening, Richard!

On Mon, Mar 31, 2008 at 12:24:42PM -0400, Richard Stallman wrote:
>     >    "You are running an Emacs configuration designed to be easy to
>     >    _learn_.  When you have become somewhat proficient in its use,
>     >    you may wish to switch to a standard Emacs setup, which is
>     >    optimised for ease of _use_ rather than ease of learning."

>     I like how you formulate the difference between ease of learn and
>     ease of use.

> Making Emacs easier to learn is an improvement, in general.  But what
> use is there in having a different Emacs configuration which is easier
> to learn, if it is not the recommended mode of use?

It is useful if it makes it easier to get to the final result, getting
there in two stages rather than going straight there - a bit like first
teaching skiers to turn by skidding their skis on the snow, to give them
confidence, then in the advanced classes getting them to edge their skis
and get proper carved turns.

> That seems like telling people a shortcut leading to somewhere other
> than their destination.

The "somewhere else" should be regarded as an intermediate stepping
stone.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 1765 bytes --]

On Mon, 31 Mar 2008 12:24:42 -0400
Richard Stallman <rms@gnu.org> wrote:

>     >    "You are running an Emacs configuration designed to be easy
>     > to _learn_.  When you have become somewhat proficient in its
>     > use, you may wish to switch to a standard Emacs setup, which is
>     > optimised for ease of _use_ rather than ease of learning."
> 
>     I like how you formulate the difference between ease of learn and
> ease of use.
> 
> Making Emacs easier to learn is an improvement, in general.
> But what use is there in having a different Emacs configuration
> which is easier to learn, if it is not the recommended mode of use?
> 
> That seems like telling people a shortcut leading to somewhere other
> than their destination.

That is the essence of my disagreement by endorsing t-m-m with default
status. Since the user's of things like t-m-m can't articulate how
the mode enhances their productivity beyond force of habit, expecting
them to articulate where they are going with it
(other than emulation - a poor standard: emulating which program/system?), 
and how it's useful is not possible.

The fact that the region abstraction is thoroughly broken as well because
they tried to make persistent marks transient, instead of making regions
transient is a disaster piled on top of confusion.

I won't argue with force of habit, but I will argue with quick and dirty hacks
where the Right Way: a new region abstraction, was eschewed for implementation
speed leaving a legacy of nasty complexity for others to sort out.

Others sorting it out implies someone who cares enough about making it simple
enough that bugs are obvious doing a considerable amount of work. The alternative
is more hair and bugs.

> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Transient Mark Mode on by default

[Alan Mackenzie]

Hi, Jari!

On Sat, Mar 29, 2008 at 11:01:47AM +0200, Jari Aalto wrote:
> * Mon 2008-03-24 Alan Mackenzie <acm@muc.de> gmane.emacs.devel
> * Message-Id: 20080324115510.GA1563@muc.de
> > Hi, Yidong and Emacs!

> > On Sun, Mar 23, 2008 at 07:00:10PM -0400, Chong Yidong wrote:
> >> that Transient Mark Mode is now turned on by default, in the trunk.

> > I feel I must protest here as strongly as I can.

> Just like the font-lock, the starting users expect certain features to
> be default in programs.

Most users also expect certain things to be on a new PC - Microsoft
Windows, for example.  If you were to take this argument to extremes (OK,
I know you wouldn't), we could make Emacs maximally attractive to newbies
by dumbing it down to look like any other Lowest Common Denominator
editor, but none of us would want to use or maintain such a thing.  So,
how far down this road do we want to go?

Transient Mark Mode is, I have argued (in the very post you are
answering), objectively less good than the traditional Emacs mark and
region - it is (i) over-complicated, (ii) modal (in the vi sense), and
(iii) rude.  Please either accept this or take issue with it.  Ignoring
it won't make it go away.

Further, Transient Mark Mode seems to be incoherent.  For a start, by
default (when mark-even-if-inactive is set), T-M-M is a stupid name - a
better one would be "Transient Region Highlighting Mode".  The mark is
ALWAYS "active", for any reasonable value of "active".  Richard's (and
other people's) view that T-M-M absolutely requires m-even-if-i set to be
tolerable is really the view that Transient Mark Mode is intolerable.
For that matter, how coherent is the name "mark-even-if-inactive"?  A
more accurate name would be "mark-active-even-if-inactive" (which
approaches jibberish), or better "mark-always-active".

To see this confusion, just peruse this sentence from the page "Transient
Mark" in the Emacs Manual:

       If the variable `mark-even-if-inactive' is non-`nil' in Transient
    Mark mode, then commands can use the mark and the region even when it
    is inactive.

- "active" and "inactive" seem to have lost all connection with their
normal meanings in this sentence.  Unless you're Lewis Carroll's Humpty
Dumpty, words mean things, and their meanings are important.

I would contend that this manual page is one of the worst - you cannot
grok T-M-M by reading this page; you can only manage this by
experimenting.

It's not clear to me why anybody would want to make the mark inactive.
Does anybody at all _really_ want to be beeped with "The mark is not
active now" when trying M-w?  That's not a rhetorical question.

In truth, T-M-M is a ragbag of features arbitrarily conflated into a
single option: There's (i) region highlighting; (ii) a variant of
narrowing, for certain commands; (iii) disabling the mark.  Anything else
I've missed?

> This is good news and helps introducing Emacs to the wider audience.

Yet to get this wider audience, what this wider audience is getting is
getting less and less like Emacs.

> The opposite can be easily be configured in ~/.emacs

This is of no relevance - we're talking about the DEFAULT Emacs in this
thread - what newcomers should be seeing.

As somebody who prizes simplicity, clarity and logical cohesion, making
this psuedo-T-M-M default fills me with revulsion.  There's got to be
better ways of attracting new users.

> Jari

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Transient Mark Mode on by default

[Sebastian Rose]

Alan Mackenzie <acm@muc.de> writes:
> To see this confusion, just peruse this sentence from the page "Transient
> Mark" in the Emacs Manual:
>
>        If the variable `mark-even-if-inactive' is non-`nil' in Transient
>     Mark mode, then commands can use the mark and the region even when it
>     is inactive.

...and the rest of the truth is the documentation for `mark':

       mark
         Function: Return this buffer's mark value as integer, or nil if never set.

       In Transient Mark mode, this function signals an error if
       the mark is not active.  However, if `mark-even-if-inactive' is non-nil,
       or the argument FORCE is non-nil, it disregards whether the mark
       is active, and returns an integer or nil in the usual way.

       If you are using this in an editing command, you are most likely making
       a mistake; see the documentation of `set-mark'.

Here is, where the former words get their sence.



> In truth, T-M-M is a ragbag of features arbitrarily conflated into a
> single option: There's (i) region highlighting; (ii) a variant of
> narrowing, for certain commands; (iii) disabling the mark.  Anything else
> I've missed?


Maybe using transient-mark-mode and have some commands narrowing to the
region ;) This is something I use all the time. Having replace-string
replacing from point to end of buffer is useless in many cases. And if I
wont to do that, I just don't create a region. Here TMM adds an editing
feature to emacs, IMHO.



> Yet to get this wider audience, what this wider audience is getting is
> getting less and less like Emacs.


They're getting TMM since I use emacs. And since then I have to turn TMM
on, if I want to use it. This was never a propblem.


> As somebody who prizes simplicity, clarity and logical cohesion, making
> this psuedo-T-M-M default fills me with revulsion.  There's got to be
> better ways of attracting new users.


I don't care about defaults at all as long they don't change all the
time. But I think TMM is somewhat underestimated here.

Visualization is what attracs users today. Make the application think
for me makes it a usefull tool. There are about 7 registers in short
time memory in an avarage brain. I rember the point and the mark even
after a 2 hours phone call if the region is highlighted. And *ploink*
remember why.

right-click to create a highlighted region, and by more right clicks on
any end of that region resize it (this could not be done without
highlighting). Or cut the region by double-right-click (and inserting by
middle-click as usual). Very convinient in some cases.




Regards,

    Sebastian

Re: Transient Mark Mode on by default

[Kim F. Storm]

Thomas Lord <lord@emf.net> writes:

> That doesn't give a "modal" system where sometimes C-v means
> scroll-down and other times it means PASTE.   

With CUA on, C-v always means paste, so it is not modal.  
Neither is C-z (undo).

C-x and C-c are modal with CUA on, but have you actually tried it?
In practice, it is very rarely notiable.  

Of course, I've used Emacs before CUA came along - but I wrote CUA --
and still use it -- because it drove me crazy to have to use different
control sequences in Emacs than in every other application I used.

And since Emacs was the only application that was different, I decided
to implement the necessary functions to make it work _good enough_ to
make me switch comfortably between Emacs and other applications.

> Emacs documentation will still be saying things like "Use C-x f to
> open a file."    That's a burden on new users who elect to remap
> C-x to CUT (and some other key to C-x).    But it's a small burden
> because it's just those few keys and the rules about how to type those
> characters apply consistently, all the time.

So to avoid the (tranparent) modal behaviour of CUA, you shuffle everything
around instead.  I don't see that as a better - or simpler - solution.
At least it is a solution to a problem which doesn't exist IMHO.

>
> -t
>
>
> Richard Stallman wrote:
>> Shift-selection is fine, but I don't think we should change the
>> meaning of C-c, C-v and C-x.
>>
>> All else being equal, it would be better to be compatible with other
>> programs, including in this.  But all else is not equal, and this
>> change would not fit into Emacs.

In practice, it fits very well, but I have no problem with it being
an option which you have to turn on explicitly.  But it would be
nice to mention it on the splash screen.

Also, if shift-select is implemented as default, I think many users
will be utterly confused if C-x doesn't do cut and C-c doesn't copy.

So all-in-all I really don't see why everybody is making a lot of
fuzz over making shift-select a 1st class emacs feature -- when
we could just as well just leave it to CUA mode to DTRT, but possibly
make a few enhancements to basic Emacs functionalities to assist CUA
mode to do its work.

But once again, I know this is a lost battle...

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Transient Mark Mode on by default

[Lennart Borgman (gmail)]

Kim F. Storm wrote:
> In practice, it fits very well, but I have no problem with it being
> an option which you have to turn on explicitly.  But it would be
> nice to mention it on the splash screen.
> 
> Also, if shift-select is implemented as default, I think many users
> will be utterly confused if C-x doesn't do cut and C-c doesn't copy.
> 
> So all-in-all I really don't see why everybody is making a lot of
> fuzz over making shift-select a 1st class emacs feature -- when
> we could just as well just leave it to CUA mode to DTRT, but possibly
> make a few enhancements to basic Emacs functionalities to assist CUA
> mode to do its work.
> 
> But once again, I know this is a lost battle...


I agree, but in the long term it is not a lost battle (unless Emacs does 
not survive of course).

Re: Transient Mark Mode on by default

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 4519 bytes --]

On Tue, 01 Apr 2008 02:14:01 +0200
Sebastian Rose <sebastian_rose@gmx.de> wrote:

> Alan Mackenzie <acm@muc.de> writes:
> > To see this confusion, just peruse this sentence from the page
> > "Transient Mark" in the Emacs Manual:
> >
> >        If the variable `mark-even-if-inactive' is non-`nil' in
> > Transient Mark mode, then commands can use the mark and the region
> > even when it is inactive.
> 
> ...and the rest of the truth is the documentation for `mark':
> 
>        mark
>          Function: Return this buffer's mark value as integer, or nil
> if never set.
> 
>        In Transient Mark mode, this function signals an error if
>        the mark is not active.  However, if `mark-even-if-inactive'
> is non-nil, or the argument FORCE is non-nil, it disregards whether
> the mark is active, and returns an integer or nil in the usual way.
> 
>        If you are using this in an editing command, you are most
> likely making a mistake; see the documentation of `set-mark'.
> 
> Here is, where the former words get their sence.
> 
> 
> 
> > In truth, T-M-M is a ragbag of features arbitrarily conflated into a
> > single option: There's (i) region highlighting; (ii) a variant of
> > narrowing, for certain commands; (iii) disabling the mark.
> > Anything else I've missed?
> 
> 
> Maybe using transient-mark-mode and have some commands narrowing to
> the region ;) This is something I use all the time. Having
> replace-string replacing from point to end of buffer is useless in
> many cases. And if I wont to do that, I just don't create a region.
> Here TMM adds an editing feature to emacs, IMHO.
> 
> 
> 
> > Yet to get this wider audience, what this wider audience is getting
> > is getting less and less like Emacs.
> 
> 
> They're getting TMM since I use emacs. And since then I have to turn
> TMM on, if I want to use it. This was never a propblem.
> 
> 
> > As somebody who prizes simplicity, clarity and logical cohesion,
> > making this psuedo-T-M-M default fills me with revulsion.  There's
> > got to be better ways of attracting new users.
> 
> 
> I don't care about defaults at all as long they don't change all the
> time. But I think TMM is somewhat underestimated here.
> 
> Visualization is what attracs users today. Make the application think
> for me makes it a usefull tool. There are about 7 registers in short
> time memory in an avarage brain. I rember the point and the mark even
> after a 2 hours phone call if the region is highlighted. And *ploink*
> remember why.

No tool thinks for you, but tools can help. When ediff highlights the
part of a diff that differs (refining the diff), then highlighting shows
me something I was looking for.

When t-m-m highlights what I already know to be the region it's not helping,
just flushing more of those precious registers you spoke of with an
angry fruit salad of color. Since it's marking where I have been, it's 
literally riding the horse backwards ala Lewis Carrol.

In fact I never remember where I set a mark, because I am busy looking
for the *other bounds*. If I need to change the other bounds I can do it
with swap point and mark. When I need to change it is the only time I even
recall the mark, much less expend brain power keeping track of it.

No one has yet to explain how highlighting where you have been points
to where you are going, because it's bloody impossible. Also you have
to be so careful not to accidentally de-activate the region. Many windows
users I have observed require several tries to mark a region successfully
precisely because it is transient - meaning you have to get it right
in one shot.

If it helps you manage the burden of habit, then so much the better. But
it is hardly the sort of thing you would teach to a child:
"Walk backwards so you can see where you have been"

I am not trying to convince you to give up t-m-m, a man convinced against
his will is of the same opinion still. But I would ask that the mark
and mark ring be left out of this transient stuff. If you really want it
transient why is the mark being molested ? invent something else !
 
> right-click to create a highlighted region, and by more right clicks
> on any end of that region resize it (this could not be done without
> highlighting). Or cut the region by double-right-click (and inserting
> by middle-click as usual). Very convinient in some cases.
> 
> 
> 
> 
> Regards,
> 
>     Sebastian
> 
> 
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Transient Mark Mode on by default

[Mike Mattie]

[-- Attachment #1: Type: text/plain, Size: 3173 bytes --]

On Tue, 01 Apr 2008 02:14:01 +0200
Sebastian Rose <sebastian_rose@gmx.de> wrote:

> Alan Mackenzie <acm@muc.de> writes:
> > To see this confusion, just peruse this sentence from the page
> > "Transient Mark" in the Emacs Manual:
> >
> >        If the variable `mark-even-if-inactive' is non-`nil' in
> > Transient Mark mode, then commands can use the mark and the region
> > even when it is inactive.
> 
> ...and the rest of the truth is the documentation for `mark':
> 
>        mark
>          Function: Return this buffer's mark value as integer, or nil
> if never set.
> 
>        In Transient Mark mode, this function signals an error if
>        the mark is not active.  However, if `mark-even-if-inactive'
> is non-nil, or the argument FORCE is non-nil, it disregards whether
> the mark is active, and returns an integer or nil in the usual way.
> 
>        If you are using this in an editing command, you are most
> likely making a mistake; see the documentation of `set-mark'.
> 
> Here is, where the former words get their sence.
> 
> 
> 
> > In truth, T-M-M is a ragbag of features arbitrarily conflated into a
> > single option: There's (i) region highlighting; (ii) a variant of
> > narrowing, for certain commands; (iii) disabling the mark.
> > Anything else I've missed?
> 
> 
> Maybe using transient-mark-mode and have some commands narrowing to
> the region ;) This is something I use all the time. Having
> replace-string replacing from point to end of buffer is useless in
> many cases. And if I wont to do that, I just don't create a region.
> Here TMM adds an editing feature to emacs, IMHO.
> 
> 
> 
> > Yet to get this wider audience, what this wider audience is getting
> > is getting less and less like Emacs.
> 
> 
> They're getting TMM since I use emacs. And since then I have to turn
> TMM on, if I want to use it. This was never a propblem.
> 
> 
> > As somebody who prizes simplicity, clarity and logical cohesion,
> > making this psuedo-T-M-M default fills me with revulsion.  There's
> > got to be better ways of attracting new users.
> 
> 
> I don't care about defaults at all as long they don't change all the
> time. But I think TMM is somewhat underestimated here.
> 
> Visualization is what attracs users today. Make the application think
> for me makes it a usefull tool. There are about 7 registers in short
> time memory in an avarage brain. I rember the point and the mark even
> after a 2 hours phone call if the region is highlighted. And *ploink*
> remember why.
> 
> right-click to create a highlighted region, and by more right clicks
> on any end of that region resize it (this could not be done without
> highlighting). Or cut the region by double-right-click (and inserting
> by middle-click as usual). Very convinient in some cases.

I refute: swap point and mark can do that. highlighting has nothing to
do with it. (swap, move)+ works. You did get closer to the one legit
scenario for highlighting when setting a bound. Hint: the mouse's resolution
is beyond that of the character cell.

> 
> 
> 
> Regards,
> 
>     Sebastian
> 
> 
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Transient Mark Mode on by default

[Thomas Lord]

[-- Attachment #1: Type: text/plain, Size: 3053 bytes --]

Hi, Kim.

No, I am not a CUA user although, from hearing it described
here I have a pretty good sense of how it works.

I don't mean the following effusive praise to be condescending
but, really, it sounds like excellent work.  Except.....

I gather you "inherited" TMM.   CUA would probably be just as
excellent work, and probably simpler, if TMM had been better
designed.  

I'm trying not to be *too* much of a pill on the dev list of
a project I'm not otherwise active in but I do think that the
accumulated mistake of the original TMM can probably be
usefully fixed, as an alternative to making it the default behavior.

-t


Kim F. Storm wrote:
> Thomas Lord <lord@emf.net> writes:
>
>   
>> That doesn't give a "modal" system where sometimes C-v means
>> scroll-down and other times it means PASTE.   
>>     
>
> With CUA on, C-v always means paste, so it is not modal.  
> Neither is C-z (undo).
>
> C-x and C-c are modal with CUA on, but have you actually tried it?
> In practice, it is very rarely notiable.  
>
> Of course, I've used Emacs before CUA came along - but I wrote CUA --
> and still use it -- because it drove me crazy to have to use different
> control sequences in Emacs than in every other application I used.
>
> And since Emacs was the only application that was different, I decided
> to implement the necessary functions to make it work _good enough_ to
> make me switch comfortably between Emacs and other applications.
>
>   
>> Emacs documentation will still be saying things like "Use C-x f to
>> open a file."    That's a burden on new users who elect to remap
>> C-x to CUT (and some other key to C-x).    But it's a small burden
>> because it's just those few keys and the rules about how to type those
>> characters apply consistently, all the time.
>>     
>
> So to avoid the (tranparent) modal behaviour of CUA, you shuffle everything
> around instead.  I don't see that as a better - or simpler - solution.
> At least it is a solution to a problem which doesn't exist IMHO.
>
>   
>> -t
>>
>>
>> Richard Stallman wrote:
>>     
>>> Shift-selection is fine, but I don't think we should change the
>>> meaning of C-c, C-v and C-x.
>>>
>>> All else being equal, it would be better to be compatible with other
>>> programs, including in this.  But all else is not equal, and this
>>> change would not fit into Emacs.
>>>       
>
> In practice, it fits very well, but I have no problem with it being
> an option which you have to turn on explicitly.  But it would be
> nice to mention it on the splash screen.
>
> Also, if shift-select is implemented as default, I think many users
> will be utterly confused if C-x doesn't do cut and C-c doesn't copy.
>
> So all-in-all I really don't see why everybody is making a lot of
> fuzz over making shift-select a 1st class emacs feature -- when
> we could just as well just leave it to CUA mode to DTRT, but possibly
> make a few enhancements to basic Emacs functionalities to assist CUA
> mode to do its work.
>
> But once again, I know this is a lost battle...
>
>   


[-- Attachment #2: Type: text/html, Size: 3764 bytes --]

Re: Transient Mark Mode on by default

[Stefan Monnier]

> So all-in-all I really don't see why everybody is making a lot of
> fuzz over making shift-select a 1st class emacs feature -- when
> we could just as well just leave it to CUA mode to DTRT, but possibly
> make a few enhancements to basic Emacs functionalities to assist CUA
> mode to do its work.

I'd tend to agree.  Maybe we should make cua-mode more visible and be
done with it.  As for enhancements to basic Emacs functionality, that's
of course possible, but I can't think of what that would look like.


        Stefan

Re: Transient Mark Mode on by default

[David Kastrup]

Stefan Monnier <monnier@IRO.UMontreal.CA> writes:

>> So all-in-all I really don't see why everybody is making a lot of
>> fuzz over making shift-select a 1st class emacs feature -- when
>> we could just as well just leave it to CUA mode to DTRT, but possibly
>> make a few enhancements to basic Emacs functionalities to assist CUA
>> mode to do its work.
>
> I'd tend to agree.  Maybe we should make cua-mode more visible and be
> done with it.

It is right there in the (sparsely populated) "Options" menu.  In my
opinion, that is as good as it gets.

We might mention it in the tutorial, but that implies adapting the
tutorial where this would affect it.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Transient Mark Mode on by default

[Robert J. Chassell]

        So all-in-all I really don't see why everybody is making a lot
         of fuzz over making shift-select a 1st class emacs feature
         ...

Except that is not what the arguments concerning Transient Mark mode
are about.  The question is whether to activate a region.  The
question comes up because the current action makes visible a mode of
navigation that some people do not use.

The arguments are not whether to have or what user interface to have
to combine two commands.  For example, an argument could regard a user
interface, whether to use a shift key in a graphic environment when
you move point and so at the same time select the region moved over.
Instead of that argument or whether to have the commands combined, the
arguments are different.

One argument is whether to use an `interactive' function for that
interface or to use something else, such as properties.

A second argument is whether to continue to use the old meaning of
mark.  (The first argument looks forward twenty years and more; the
second looks back twenty years and more.)

A third argument is whether to activate a region with the same
highlighting and under the same conditions that others use for
navigation and/or discover with `exchange-point-and-mark' with the
user interface `C-x C-x'.  (That argument looks both forward and
back.)

-- 
    Robert J. Chassell                          GnuPG Key ID: 004B4AC8
    bob@rattlesnake.com                         bob@gnu.org
    http://www.rattlesnake.com                  http://www.teak.cc

Re: Transient Mark Mode on by default

[Richard Stallman]

We should make shift-select a standard feature of Emacs
if that would be an improvement.  Perhaps it would be.

Re: Transient Mark Mode on by default

[Richard Stallman]

I don't think we should even consider the idea of changing central
Emacs commands such as C-v or C-x.

Re: Transient Mark Mode on by default

[Chong Yidong]

David Kastrup <dak@gnu.org> writes:

>>> So all-in-all I really don't see why everybody is making a lot of
>>> fuzz over making shift-select a 1st class emacs feature -- when
>>> we could just as well just leave it to CUA mode to DTRT, but possibly
>>> make a few enhancements to basic Emacs functionalities to assist CUA
>>> mode to do its work.
>>
>> I'd tend to agree.  Maybe we should make cua-mode more visible and be
>> done with it.
>
> It is right there in the (sparsely populated) "Options" menu.  In my
> opinion, that is as good as it gets.

cua-selection-mode is not available in the Options menu.  I wouldn't
mind adding an additional menu item for it, but if so, I think it should
be on by default.

There are two good reasons for providing shift-selection by default.
Firstly, new users expect shift-selection.  Secondly, unlike the other
features provided by cua-mode, shift-selection does not change the
existing Emacs control scheme in any significant way, since shift-arrow
keys are currently undefined key sequences (which, for convenience, we
translate into unshifted keys).

The reason that we started this long discussion about making
shift-selection a 1st class feature is that people weren't happy about
the way cua-selection-mode works (e.g., using the post-command hooks),
which prevents it from being turned on by default.

Furthermore, making shift-selection a 1st class feature isn't too
difficult, as I've previously shown.  It also has the advantage of
integrating with and complementing mouse-selection (e.g., you can select
a region with the mouse, then extend it with shift-selection).

Re: Transient Mark Mode on by default

[Thomas Lord]

Richard Stallman wrote:
> I don't think we should even consider the idea of changing central
> Emacs commands such as C-v or C-x.
>
>   


I am not sure if that is in reply to me.

I agree, though.   And my proposals are consistent with that.
If it were the case you thought I said something otherwise,
I would want to make it clear that you misunderstood.

This doesn't address the question of whether Emacs should
take up *my* suggestions.     I'm just pointing out some of the
logic.

-t

Re: Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[Richard Stallman]

    > Making Emacs easier to learn is an improvement, in general.  But what
    > use is there in having a different Emacs configuration which is easier
    > to learn, if it is not the recommended mode of use?

    It is useful if it makes it easier to get to the final result, getting
    there in two stages rather than going straight there - a bit like first
    teaching skiers to turn by skidding their skis on the snow, to give them
    confidence, then in the advanced classes getting them to edge their skis
    and get proper carved turns.

I am skeptical that it will work this way.  I think that most people
will learn the easy-to-learn version and stop there.

Re: Transient Mark Mode on by default

[Thomas Lord]

It occurred to me that this might be misunderstood:

Thomas Lord wrote:
>
> This doesn't address the question of whether Emacs should
> take up *my* suggestions. 


I don't mean to suggest that the Emacs project is prejudiced
against me.   I mean to apologize for volunteering some
(imho) good ideas in the awkward context of not being
able to actually, well, you know *volunteer* for the project.

-t

Re: Honoring traditional defaults - how to do it. [was: Transient Mark Mode on bydefault]

[Alan Mackenzie]

Hi, Richard!

On Tue, Apr 01, 2008 at 10:53:59PM -0400, Richard Stallman wrote:
>     > Making Emacs easier to learn is an improvement, in general.  But
>     > what use is there in having a different Emacs configuration which
>     > is easier to learn, if it is not the recommended mode of use?

>     It is useful if it makes it easier to get to the final result,
>     getting there in two stages rather than going straight there - a
>     bit like first teaching skiers to turn by skidding their skis on
>     the snow, to give them confidence, then in the advanced classes
>     getting them to edge their skis and get proper carved turns.

> I am skeptical that it will work this way.  I think that most people
> will learn the easy-to-learn version and stop there.

We seem to have come full circle.

My proposal was to have several commands to start Emacs, in particular:
(i) "emacs-easy" would start Emacs with Transient Mark Mode enabled, and
  display a message suggesting the newby change to "emacs".
(ii) "emacs" would retain the traditional default here.

You have endorsed making T-M-M on by default.  I have argued that T-M-M
is objectively inferior to !T-M-M, particularly for experienced Emacsers.

Identifying "easy-to-learn version" with "T-M-M as default", I think
newbies are more likely to move to "traditional emacs" by my suggestion
than by the current state (T-M-M on by default).

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Transient Mark Mode on by default

[Kim F. Storm]

Chong Yidong <cyd@stupidchicken.com> writes:

> Furthermore, making shift-selection a 1st class feature isn't too
> difficult, as I've previously shown.  It also has the advantage of
> integrating with and complementing mouse-selection (e.g., you can select
> a region with the mouse, then extend it with shift-selection).

I've also suggested ways to make shift-selection a 1st class feature
based on the CUA approach - by moving part of the stuff that is currently
in the pre-/post-command-hooks into the command loop, and only call the
(modified) CUA-functions when certain conditions are true.

It seems that Richard is in favour of a scheme which supports both ^ in
the interactive spec and a command property (for external packages),
but IMHO, this is overkill -- 

If we name the property something like shift-select, then the help
system can easily tell people that "applying the Shift modifier to
this command will start or extend the active region".  

I actually think this is much easier to accomplish than having to add
that text explicitly to the doc string of each command which has ^ in
the interactive spec.

So no matter what the arguments for "cleanliness" are, I think the
property approach is simpler, better - and has already proven to work
just fine for MANY years!

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Transient Mark Mode on by default

[Chong Yidong]

storm@cua.dk (Kim F. Storm) writes:

> Chong Yidong <cyd@stupidchicken.com> writes:
>
>> Furthermore, making shift-selection a 1st class feature isn't too
>> difficult, as I've previously shown.  It also has the advantage of
>> integrating with and complementing mouse-selection (e.g., you can select
>> a region with the mouse, then extend it with shift-selection).
>
> I've also suggested ways to make shift-selection a 1st class feature
> based on the CUA approach - by moving part of the stuff that is currently
> in the pre-/post-command-hooks into the command loop, and only call the
> (modified) CUA-functions when certain conditions are true.

Right, but Stefan also wanted to modify the way momentary selection
(a.k.a. tmm `only' mode) works, to make it less fragile.

> If we name the property something like shift-select, then the help
> system can easily tell people that "applying the Shift modifier to
> this command will start or extend the active region".  

It wouldn't be any harder to make describe-function describe the `^'
spec.  Actually, it ought to decribe the `@' and `*' specs anyway.

Re: Transient Mark Mode on by default

[Richard Stallman]

    It seems that Richard is in favour of a scheme which supports both ^ in
    the interactive spec and a command property (for external packages),
    but IMHO, this is overkill -- 

It is cleaner to have this in `interactive' than in a separate property.

    If we name the property something like shift-select, then the help
    system can easily tell people that "applying the Shift modifier to
    this command will start or extend the active region".  

It can do that based on the interactive spec, too.  That is a good idea.

RE: Transient Mark Mode on by default

[Drew Adams]

>     It seems that Richard is in favour of a scheme which 
>     supports both ^ in the interactive spec and a command
>     property (for external packages),
>     but IMHO, this is overkill -- 
> 
> It is cleaner to have this in `interactive' than in a 
> separate property.
>
>     If we name the property something like shift-select, then the help
>     system can easily tell people that "applying the Shift modifier to
>     this command will start or extend the active region".  
>
> It can do that based on the interactive spec, too.  That is a 
> good idea.

I have not followed this shift discussion closely; sorry.

I assumed that users will be able to easily turn this feature off, so that Shift
will then not automatically act on cursor-moving commands to extend the region.
Is that not the case?

If it is the case, then how does that fit with Help explaining, for each such
command, that Shift extends the region? 

Will there be a user option to turn this feature off? In that case, Help could
make the info conditional by mentioning that user option (If `foo' is non-nil,
then applying the Shift modifier to this command will start or extend the active
region".

[Just for the record, I'm not crazy about having Shift-extends-the-region turned
on by default. Among other things, it could have the effect of discouraging
libraries and perhaps even users from using Shift with cursor-moving keys for
something else. That is a lot of key sequences that will be lost/discouraged,
for little gain, IMO.]

Re: Transient Mark Mode on by default

[Chong Yidong]

Richard Stallman <rms@gnu.org> writes:

>     It seems that Richard is in favour of a scheme which supports both ^ in
>     the interactive spec and a command property (for external packages),
>     but IMHO, this is overkill -- 
>
> It is cleaner to have this in `interactive' than in a separate property.
>
>     If we name the property something like shift-select, then the help
>     system can easily tell people that "applying the Shift modifier to
>     this command will start or extend the active region".  
>
> It can do that based on the interactive spec, too.  That is a good idea.

I've committed to CVS the shift-selection mechanism based on the
`interactive' property.  See the new variable `shift-select-mode'.

As per Stefan's suggestion, the shift-selection is a little more
persistent here than in the old `only' system: only unshifted point
motion commands and mark-deactivating commands will deactivate the mark.
For example, something like C-x 2 won't deactivate the mark.  This
behavior extends to mouse selection too.

This should mesh correctly with the ordinary C-SPC system, as well as
drag-mouse, mouse-3, C-u C-SPC, C-x C-x, and the mark-* commands.  Give
it a try, and let me know if there are any problems.  The shift-select
property and describe-function changes remain unchanged, so far.

I've also taken the liberty of turning off shift-select-mode when
cua-mode is on, so that cua-selection behavior should be unchanged.

Re: Transient Mark Mode on by default

[Kim F. Storm]

Chong Yidong <cyd@stupidchicken.com> writes:

> I've also taken the liberty of turning off shift-select-mode when
> cua-mode is on, so that cua-selection behavior should be unchanged.

Thank you!

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: Transient Mark Mode on by default

[Richard Stallman]

    I assumed that users will be able to easily turn this feature off, so that Shift
    will then not automatically act on cursor-moving commands to extend the region.
    Is that not the case?

None of this is the case now, since it is just a proposal, but that
suggestion seems like a good one.

    If it is the case, then how does that fit with Help explaining, for each such
    command, that Shift extends the region? 

Whatever option enables and disables that feature could be checked
by the help commands too.

cua-selection-mode by default (was: Transient Mark Mode on by default)

[David Kastrup]

Chong Yidong <cyd@stupidchicken.com> writes:

> David Kastrup <dak@gnu.org> writes:
>
>>>> So all-in-all I really don't see why everybody is making a lot of
>>>> fuzz over making shift-select a 1st class emacs feature -- when
>>>> we could just as well just leave it to CUA mode to DTRT, but possibly
>>>> make a few enhancements to basic Emacs functionalities to assist CUA
>>>> mode to do its work.
>>>
>>> I'd tend to agree.  Maybe we should make cua-mode more visible and be
>>> done with it.
>>
>> It is right there in the (sparsely populated) "Options" menu.  In my
>> opinion, that is as good as it gets.
>
> cua-selection-mode is not available in the Options menu.  I wouldn't
> mind adding an additional menu item for it, but if so, I think it should
> be on by default.
>
> There are two good reasons for providing shift-selection by default.

[...]

Oh please.  If you want to discuss something different, open a new
thread instead of feigning disagreement with a different posting of mine
(I was talking about cua-mode quite explicitly).  We had enough strawmen
in this thread.

I am perfectly fine with making cua-selection-mode (which I never tried)
the default without even an option menu entry as long as

a) it does not interfere with preexisting bindings (whoever is capable
of extending those bindings should be able to customize cua-selection
off if he really needs to)

b) turning transient-mark-mode off leaves one with a reasonable set of
features/bindings.  This implies that cua-selection-mode should not be
made to rely on full transient-mark-mode.  If no feature of it makes any
sense at all without transient-mark-mode, cua-selection-mode should
either be inactive without transient-mark-mode, or interact usefully
with temporary transient-mark-mode.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum