Bookmark the end of file?

Bookmark the end of file?

[Marcin Borkowski]

Hi all,

how to set a bookmark to the end of some file (IOW, (point-max))?  (Of
course it is possible with Bookmark+, but is there any UI for that
there?  Is it at all possible with vanilla bookmarks?)

My use case is a file which I visit (almost) every day to /add/
something to its end (e.g., a ledger file).  I can imagine other,
similar uses for that, too.

TIA,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

RE: Bookmark the end of file?

[Drew Adams]

> how to set a bookmark to the end of some file (IOW, (point-max))?  (Of
> course it is possible with Bookmark+, but is there any UI for that
> there?  Is it at all possible with vanilla bookmarks?)
> 
> My use case is a file which I visit (almost) every day to /add/
> something to its end (e.g., a ledger file).  I can imagine other,
> similar uses for that, too.

1. You can use `bookmark-after-jump-hook' with both vanilla
`bookmark.el' and Bookmark+.  Add a function that goes to
`point-max', for instance.

If you don't want to do that for all bookmarks (which I
imagine), you can define a command that binds that hook and
then jumps to a particular bookmark or lets you choose a
bookmark from some particular set of bookmarks.  IOW, create
your own `jump-then-goto-eof' command.

2. Both vanilla bookmarks and Bookmark+ bookmarks save text
before and after point, and then use that text to try to
reposition the location later.  Usually that enables relocation
correctly.

But if the location is at eof/eob and you append new text to
the file then this relocation might fail.

With Bookmark+, you can bookmark the file and then combine
that bookmark with a function bookmark that goes to `point-max'.
That is, create a sequence bookmark of: (1) the bookmark to the
file followed by (2) a function bookmark that goes to eof.

3. With Bookmark+, you can tag a bookmark with tag "bmkp-jump".
The function that is the value of a "bmkp-jump" tag is called
just after standard hook `bookmark-after-jump-hook' is invoked.

Note that this is a case where the tag is not just a string -
it is a cons with car "bmkp-jump" and with cdr the extra jump
function.  IOW, you can use tag "bmkp-jump" to invoke functions
that are specific to individual bookmarks.  In this way,
bookmarks can have their own, extra jump behavior.

Re: Bookmark the end of file?

[Dale Snell]

[-- Attachment #1: Type: text/plain, Size: 871 bytes --]

On Tue, 07 Apr 2015 01:39:03 +0200, in message
87vbh8hlns.fsf@wmi.amu.edu.pl, Marcin Borkowski wrote:

> Hi all,
> 
> how to set a bookmark to the end of some file (IOW, (point-max))?  (Of
> course it is possible with Bookmark+, but is there any UI for that
> there?  Is it at all possible with vanilla bookmarks?)
> 
> My use case is a file which I visit (almost) every day to /add/
> something to its end (e.g., a ledger file).  I can imagine other,
> similar uses for that, too.


Marcin,

Why not use "M->" ((end-of-buffer &optional ARG), which is also
bound to "C-end", and <menu-bar> <edit> <goto> <end-of-buffer>)?
There's no need to set up a bookmark, or other bit of programming.

The reverse operation is bound to "M-<" ("C-home", etc).

--Dale

-- 
Q:  Why did the Klingon color his hair blond?
A:  Because it was a good day to dye.

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

Re: Bookmark the end of file?

[Marcin Borkowski]

On 2015-04-07, at 01:59, Drew Adams <drew.adams@oracle.com> wrote:

>> how to set a bookmark to the end of some file (IOW, (point-max))?  (Of
>> course it is possible with Bookmark+, but is there any UI for that
>> there?  Is it at all possible with vanilla bookmarks?)
>> 
>> My use case is a file which I visit (almost) every day to /add/
>> something to its end (e.g., a ledger file).  I can imagine other,
>> similar uses for that, too.
>
> 1. You can use `bookmark-after-jump-hook' with both vanilla
> `bookmark.el' and Bookmark+.  Add a function that goes to
> `point-max', for instance.
>
> If you don't want to do that for all bookmarks (which I
> imagine), you can define a command that binds that hook and
> then jumps to a particular bookmark or lets you choose a
> bookmark from some particular set of bookmarks.  IOW, create
> your own `jump-then-goto-eof' command.

Thanks - but not really what I'm after (as you pointed out).  If
I wanted a command, I wouldn't need bookmarks at all.  (Though this is
not an unreasonable thing to use, either.)

> 2. Both vanilla bookmarks and Bookmark+ bookmarks save text
> before and after point, and then use that text to try to
> reposition the location later.  Usually that enables relocation
> correctly.
>
> But if the location is at eof/eob and you append new text to
> the file then this relocation might fail.

Exactly.

> With Bookmark+, you can bookmark the file and then combine
> that bookmark with a function bookmark that goes to `point-max'.
> That is, create a sequence bookmark of: (1) the bookmark to the
> file followed by (2) a function bookmark that goes to eof.

That was more or less what I was thinking about.  I was just curious
whether Bookmark+ has some UI for that scenario.

> 3. With Bookmark+, you can tag a bookmark with tag "bmkp-jump".
> The function that is the value of a "bmkp-jump" tag is called
> just after standard hook `bookmark-after-jump-hook' is invoked.
>
> Note that this is a case where the tag is not just a string -
> it is a cons with car "bmkp-jump" and with cdr the extra jump
> function.  IOW, you can use tag "bmkp-jump" to invoke functions
> that are specific to individual bookmarks.  In this way,
> bookmarks can have their own, extra jump behavior.

And this is probably the cleanest solution, and the one which I'm going
to use.

Thanks!

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: Bookmark the end of file?

[Marcin Borkowski]

On 2015-04-07, at 02:00, Dale Snell <ddsnell@frontier.com> wrote:

> On Tue, 07 Apr 2015 01:39:03 +0200, in message
> 87vbh8hlns.fsf@wmi.amu.edu.pl, Marcin Borkowski wrote:
>
>> Hi all,
>> 
>> how to set a bookmark to the end of some file (IOW, (point-max))?  (Of
>> course it is possible with Bookmark+, but is there any UI for that
>> there?  Is it at all possible with vanilla bookmarks?)
>> 
>> My use case is a file which I visit (almost) every day to /add/
>> something to its end (e.g., a ledger file).  I can imagine other,
>> similar uses for that, too.
>
>
> Marcin,
>
> Why not use "M->" ((end-of-buffer &optional ARG), which is also
> bound to "C-end", and <menu-bar> <edit> <goto> <end-of-buffer>)?
> There's no need to set up a bookmark, or other bit of programming.
>
> The reverse operation is bound to "M-<" ("C-home", etc).

Well, thinking in that direction, I don't need Emacs at all: a magnetic
needle and a steady hand is all that is really necessary. ;-)

But putting stupid jokes aside, if I /always/ hit M-> after visiting
some file, something is not optimal, right?  And bookmarks are
a convenient way to visit often-used files, regardless if the directory
I'm in.  (And I have a rather deep directory structure - I have
virtually no files until two levels down, and usually there are three or
four.  OTOH, there are no more than maybe two-three dozen of files I'm
working on at any particular moment.  This is a perfect use case for
bookmarks.)

> --Dale

But thanks anyway!

PS. BTW, the other bindings you mentioned are useless for me: I have my
menu-bar disabled (ok, I /could/ use M-x menu-bar-open...), and don't
have an <end> key on my netbook.

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: Bookmark the end of file?

[J. David Boyd]

Marcin Borkowski <mbork@wmi.amu.edu.pl> writes:

> On 2015-04-07, at 02:00, Dale Snell <ddsnell@frontier.com> wrote:
>
>> On Tue, 07 Apr 2015 01:39:03 +0200, in message
>> 87vbh8hlns.fsf@wmi.amu.edu.pl, Marcin Borkowski wrote:
>>
>>> Hi all,
>>>
>>> how to set a bookmark to the end of some file (IOW, (point-max))?  (Of
>>> course it is possible with Bookmark+, but is there any UI for that
>>> there?  Is it at all possible with vanilla bookmarks?)
>>>
>>> My use case is a file which I visit (almost) every day to /add/
>>> something to its end (e.g., a ledger file).  I can imagine other,
>>> similar uses for that, too.
>>
>>
>> Marcin,
>>
>> Why not use "M->" ((end-of-buffer &optional ARG), which is also
>> bound to "C-end", and <menu-bar> <edit> <goto> <end-of-buffer>)?
>> There's no need to set up a bookmark, or other bit of programming.
>>
>> The reverse operation is bound to "M-<" ("C-home", etc).
>
> Well, thinking in that direction, I don't need Emacs at all: a magnetic
> needle and a steady hand is all that is really necessary. ;-)
>
> But putting stupid jokes aside, if I /always/ hit M-> after visiting
> some file, something is not optimal, right?  And bookmarks are
> a convenient way to visit often-used files, regardless if the directory
> I'm in.  (And I have a rather deep directory structure - I have
> virtually no files until two levels down, and usually there are three or
> four.  OTOH, there are no more than maybe two-three dozen of files I'm
> working on at any particular moment.  This is a perfect use case for
> bookmarks.)
>
>> --Dale
>
> But thanks anyway!
>
> PS. BTW, the other bindings you mentioned are useless for me: I have my
> menu-bar disabled (ok, I /could/ use M-x menu-bar-open...), and don't
> have an <end> key on my netbook.


Why not define a function to do what you want, then bind it to a keystroke?

I have stuff like

  (defun open-personal()
    (interactive) (progn(find-file "~/org/personal/personal.org")(end-of-buffer)))

and then with

  (bind-key (kbd "H-M-p") 'open-personal)


I can open my file and go where I want to with a keypress.

Dave

Re: Bookmark the end of file?

[Dale Snell]

[-- Attachment #1: Type: text/plain, Size: 1925 bytes --]

On Tue, 07 Apr 2015 03:10:56 +0200, in message
87sicchhen.fsf@wmi.amu.edu.pl, Marcin Borkowski wrote:

> But putting stupid jokes aside, if I /always/ hit M-> after visiting
> some file, something is not optimal, right?  And bookmarks are
> a convenient way to visit often-used files, regardless if the
> directory I'm in.  (And I have a rather deep directory structure - I
> have virtually no files until two levels down, and usually there are
> three or four.  OTOH, there are no more than maybe two-three dozen of
> files I'm working on at any particular moment.  This is a perfect use
> case for bookmarks.)

Are these files of a type that can take comments?  If so, you
could try adding a file-local variable to each one, telling Emacs
to go to the end of the file.  Another possibility is to Customize
the "find-file-hook" to do what you want.  (Despite its name,
find-file-hook is a "List of functions to be called after a buffer
is loaded from a file."  [From the documentation.])

Hmm... Now that I think about it, I seem to recall someone on this
list asking if one could create a command-line option to do what
you're asking.  "emacs --exec (<go to eof>) filename", or
something like that.  Of course, it would only work with the
file(s) on the command line.  (If you were the one who asked that
question, I'm going to feel very silly.)


> PS. BTW, the other bindings you mentioned are useless for me: I have
> my menu-bar disabled (ok, I /could/ use M-x menu-bar-open...), and
> don't have an <end> key on my netbook.

That's okay; while I do have the menu-bar active, I never use it
to go to the end or beginning of the file -- M-> is so much
simpler to type.  As for the <End> key, I've never used it.  :-)


I don't know if any of this will help, but maybe it will point you
in the right direction.

--Dale

-- 
"Come, muse, let us sing of rats."  -- James Grainger

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

Re: Bookmark the end of file?

[Marcin Borkowski]

On 2015-04-07, at 15:36, J. David Boyd <dboyd2@mmm.com> wrote:

> Why not define a function to do what you want, then bind it to a keystroke?
>
> I have stuff like
>
>   (defun open-personal()
>     (interactive) (progn(find-file "~/org/personal/personal.org")(end-of-buffer)))
>
> and then with
>
>   (bind-key (kbd "H-M-p") 'open-personal)
>
>
> I can open my file and go where I want to with a keypress.

Because bookmarks give me a uniform UI for "visiting a file I'm working
on", without the cognitive burden of remembering one keystroke to visit
one file, another one for another file and so on.  IOW, "One keybinding
to visit them all, one keybinding to find them." ;-).

> Dave

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: Bookmark the end of file?

[Marcin Borkowski]

On 2015-04-07, at 18:49, Dale Snell <ddsnell@frontier.com> wrote:

> On Tue, 07 Apr 2015 03:10:56 +0200, in message
> 87sicchhen.fsf@wmi.amu.edu.pl, Marcin Borkowski wrote:
>
>> But putting stupid jokes aside, if I /always/ hit M-> after visiting
>> some file, something is not optimal, right?  And bookmarks are
>> a convenient way to visit often-used files, regardless if the
>> directory I'm in.  (And I have a rather deep directory structure - I
>> have virtually no files until two levels down, and usually there are
>> three or four.  OTOH, there are no more than maybe two-three dozen of
>> files I'm working on at any particular moment.  This is a perfect use
>> case for bookmarks.)
>
> Are these files of a type that can take comments?  If so, you
> could try adding a file-local variable to each one, telling Emacs
> to go to the end of the file.  Another possibility is to Customize
> the "find-file-hook" to do what you want.  (Despite its name,
> find-file-hook is a "List of functions to be called after a buffer
> is loaded from a file."  [From the documentation.])

Nice idea (I mean, file-local vars).  Bookmarks, however, have one
advantage: I can visit bookmarked files from any directory, without
having to specify the path.  And I like putting the "jump-to-the-end"
functionality into the bookmark better than in file-local variables,
since then visiting a file with C-x C-f has uniform semantics of, well,
visiting the file, and jumping to a bookmark has a uniform semantics of,
well, visiting the file /in the place I'm interested in/ (after all,
this is what bookmarks are for).  It just happens that for some files
the "interesting" place is the EOF.

Customizing find-file-hook seems a dirty hack, since I want this
behavior only for few files.

> Hmm... Now that I think about it, I seem to recall someone on this
> list asking if one could create a command-line option to do what
> you're asking.  "emacs --exec (<go to eof>) filename", or
> something like that.  Of course, it would only work with the
> file(s) on the command line.  (If you were the one who asked that
> question, I'm going to feel very silly.)

I vaguely remember that discussion (and no, it wasn't me).  Not my use
case - I want it from within Emacs.

>> PS. BTW, the other bindings you mentioned are useless for me: I have
>> my menu-bar disabled (ok, I /could/ use M-x menu-bar-open...), and
>> don't have an <end> key on my netbook.
>
> That's okay; while I do have the menu-bar active, I never use it
> to go to the end or beginning of the file -- M-> is so much
> simpler to type.  As for the <End> key, I've never used it.  :-)

I have a small netbook, so every pixel of screen real estate is precious
for me.  No menus, no toolbar, no window decorations (I use a tiling
WM).  I even changed the mode-line face to a smaller one.
(Incidentally, more things than the mode-line use the mode-line face,
btw.)

> I don't know if any of this will help, but maybe it will point you
> in the right direction.
>
> --Dale

Thanks for your suggestions, even if I won't use them - they might be of
use for someone else!

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: Bookmark the end of file?

[Stefan Monnier]

> how to set a bookmark to the end of some file (IOW, (point-max))?  (Of
> course it is possible with Bookmark+, but is there any UI for that
> there?  Is it at all possible with vanilla bookmarks?)

I don't think there's support for that currently.

OTOH, I think it'd be OK to refine the current code such that a bookmark
that's at a position equal to point-max would be marked as "go the EOB"
instead of "go to the position that happened to be EOB back when we
created the bookmark".

It shouldn't be too difficult either: patch welcome,


        Stefan

RE: Bookmark the end of file?

[Drew Adams]

> I think it'd be OK to refine the current code such that a bookmark
> that's at a position equal to point-max would be marked as "go the EOB"
> instead of "go to the position that happened to be EOB back when we
> created the bookmark".

Unless that's not what the user wants. ;-)

The usual behavior of a bookmark is to take you to a recorded
position, which is understood not to be absolute but relative to the
surrounding context (i.e., located by that context).

A buffer (file) position is what is bookmarked.  But the intention
is for the bookmark to be automatically repositioned to stay, if
possible, surrounded by the same text context.  The recorded position
is used only as a starting point.  The actual jump position is
adjusted to try to place it in the same context.

So is "relative to the surrounding context" to be understood in this
particular case as just after the text that precedes the bookmarked
place?  Or is it to be understood as just before the (nonexistent)
text that follows it, i.e., at eob?

Do you assume that most users want to track eob in such a case?
Or do you assume that they want to track the text that preceded the
bookmarked position?

Consider also this comment for the code that automatically relocates
a bookmark when you jump to it, to keep it within/near its originally
recorded context:

;; Go searching forward first.  Then, if forward-str exists and
;; was found in the file, we can search backward for behind-str.
;; Rationale is that if text was inserted between the two in the
;; file, it's better to be put before it so you can read it,
;; rather than after and remain perhaps unaware of the changes.

In the special case we are considering now, you seem to assume that
it is better for the cursor to be put *after* any text inserted
since bookmarking, not that "it's better to be put before it so you
can read it".

I'd say there are different use cases to consider.

One approach is to offer users some ways to obtain one or the other
behavior after the (ordinary) bookmark has been created.  (Bookmark+
does that.)

Another approach could be to let users choose the behavior when a
bookmark is created at eob, and save that behavior choice in the
bookmark.  Of course then the bookmark is essentially of a different
type (different recorded info), and if a user wants to later switch
it to the other behavior then sh?e will need to jump through some hoops.

Re: Bookmark the end of file?

[Stefan Monnier]

> Do you assume that most users want to track eob in such a case?
[...]
> Consider also this comment for the code that automatically relocates
> a bookmark when you jump to it, to keep it within/near its originally
> recorded context:

The EOB is an extremely important part of the context.
So, yes, if point was at EOB when you set the bookmark, I think the user
will more often be satisfied if we store "EOB" rather than if we store
"the position that was EOB when the bookmark was set".


        Stefan

RE: Bookmark the end of file?

[Drew Adams]

> > > I think it'd be OK to refine the current code such that a bookmark
> > > that's at a position equal to point-max would be marked as "go the EOB"
> > > instead of "go to the position that happened to be EOB back when we
> > > created the bookmark".
> >
> > Unless that's not what the user wants. ;-)

> > Do you assume that most users want to track eob in such a case? ...
> > Consider also this comment for the code that automatically
> > relocates a bookmark when you jump to it, to keep it within/near
> > its originally recorded context:
> >
> > ;; ... it's better to be put before it so you can read it,
> > ;; rather than after and remain perhaps unaware of the changes.
> 
> The EOB is an extremely important part of the context.  So, yes, if
> point was at EOB when you set the bookmark, I think the user will
> more often be satisfied if we store "EOB" rather than if we store
> "the position that was EOB when the bookmark was set".

Unless that's not what the user wants. ;-)

As I said, 

> > there are different use cases to consider.  One approach is to
> > offer users some ways to obtain one or the other behavior after the
> > (ordinary) bookmark has been created....  Another approach could be
> > to let users choose the behavior when a bookmark is created at eob

Even if you were right that one use case is more common, why
wouldn't we want to let users control the behavior, instead of
deciding it for them at coding time?

IOW, instead of guessing what will make users "more often be
satisified", why not let them decide what satisfies them in any
given context?

Re: Bookmark the end of file?

[Stefan Monnier]

> Even if you were right that one use case is more common, why
> wouldn't we want to let users control the behavior, instead of
> deciding it for them at coding time?

Because offering more choices doesn't always help the user.


        Stefan

RE: Bookmark the end of file?

[Drew Adams]

> > Even if you were right that one use case is more common, why
> > wouldn't we want to let users control the behavior, instead of
> > deciding it for them at coding time?
> 
> Because offering more choices doesn't always help the user.

No, of course not.  But neither does offering fewer choices.

How about speaking to this particular question, instead just
throwing out truisms?

(Why paint that boat orange?  Because it's not always best to
paint boats blue...)

Re: Bookmark the end of file?

[Stefan Monnier]

> But neither does offering fewer choices.

I'm not suggesting to offer fewer choices.  Just not to add
a new choice.

Anyway, I don't need to convince you, luckily.  I'm just making it clear
that if someone submits a patch which changes the behavior such that
bookmarks "stay at EOB", I'll gladly accept it.


        Stefan

Do we want more choice? (was Bookmark the end of file?)

[Rusi]

On Monday, April 13, 2015 at 3:05:07 AM UTC+5:30, Drew Adams wrote:
> > > Even if you were right that one use case is more common, why
> > > wouldn't we want to let users control the behavior, instead of
> > > deciding it for them at coding time?
> > 
> > Because offering more choices doesn't always help the user.
> 
> No, of course not.  But neither does offering fewer choices.

There is a general realization nowadays, that everything being equal,
less choice is likely better than more choice:
http://en.wikipedia.org/wiki/The_Paradox_of_Choice

In the case of emacs that gets a hefty multiplier.

eg. To set up thunderbird one needs to point it to the right source and maybe a 
couple more settings
To set up gnus one has to read a 500 page manual of choices!!
Do you not think that corresponds to the number of thunderbird to gnus users?

At a more coarse-grained level:
When I was scouting around for something like org/planner I asked here,
I got a lot of lengthy and diplomatic answers.
Wasted a couple of months deciding.
Now after some years of using org, I can give a 1-line answer that would have
been way more helpful to me then:

"Planner core-devs have switched to using org themselves"

In short planner vs org is a bogus choice

tl;dr If I have any choice in the matter, please take away the spurious choices

RE: Do we want more choice? (was Bookmark the end of file?)

[Drew Adams]

> > > > Even if you were right that one use case is more common, why
> > > > wouldn't we want to let users control the behavior, instead
> > > > of deciding it for them at coding time?
> > >
> > > Because offering more choices doesn't always help the user.
> >
> > No, of course not.  But neither does offering fewer choices.
> > How about speaking to this particular question, instead just
> > throwing out truisms?
> 
> There is a general realization nowadays, that everything being
> equal, less choice is likely better than more choice:

"A general realization." "Nowadays." Ooooh. Can't argue with that.

Consideration of one particular choice has now been successfully
diverted toward vacuous generalizations about choice in general -  
platitudes of Polonius.  Too bad.

Not to mention that no one proposed forcing users to choose
anything here.  Define a default behavior.  Provide an easy
way to get other, optional behavior.  No imposition.  No user
paralysis before a terrifying and difficult dilemma.  Perfectly
ignorable choice; zero choosing needed.

I'm afraid you've missed the point. But you were led by example
to miss it.

Re: Do we want more choice? (was Bookmark the end of file?)

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Sun, Apr 12, 2015 at 10:35:21PM -0700, Drew Adams wrote:

[...]

> "A general realization." "Nowadays." Ooooh. Can't argue with that.
> 
> Consideration of one particular choice has now been successfully
> diverted toward vacuous generalizations about choice in general -  
> platitudes of Polonius.  Too bad.

I must say that I agree strongly with this point of view.

More choice is *always* better. The hard part is presenting this
choice to the user in a way it doesn't overwhelm her/him.

I have the strong suspicion that the "less choice is good" movement
we are seeing stems from being too lazy (or at a loss) wrt thinking
hard about how to guide the user through a rich set of choices. The
latter would be, in my view, the right thing to do.

- -- t
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlUrbZUACgkQBcgs9XrR2kZqHgCfWCmDs7p2DrL7BCSWVv5DPIQg
msIAnA4Ui+l9/ltU8UCRuabvCjdRBLBA
=Q2jP
-----END PGP SIGNATURE-----

Re: Do we want more choice? (was Bookmark the end of file?)

[Marcin Borkowski]

On 2015-04-13, at 07:35, Drew Adams <drew.adams@oracle.com> wrote:

>> There is a general realization nowadays, that everything being
>> equal, less choice is likely better than more choice:
>
> "A general realization." "Nowadays." Ooooh. Can't argue with that.

This made me laugh.  And I thought of this:
http://www.paulgraham.com/say.html immediately.

> Not to mention that no one proposed forcing users to choose
> anything here.  Define a default behavior.  Provide an easy
> way to get other, optional behavior.  No imposition.  No user
> paralysis before a terrifying and difficult dilemma.  Perfectly
> ignorable choice; zero choosing needed.

+1.  I guess that Gnus users are exactly the ones dissatisfied by the
/lack/ of choices in Thunderbird.  (Though I admit that the difficulty
of setup of Gnus scared me off, and I'm happy with mu4e.)

OTOH, I agree that reasonable defaults is what should be pursued.

BTW, there is at least one community which commonly regards "too few
choices" as something very bad: boardgamers.  And the intersection of
boardgamers and computer geeks is not only nonempty, but rather big,
I would guess.

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: Do we want more choice? (was Bookmark the end of file?)

[Marcin Borkowski]

On 2015-04-13, at 09:17, tomas@tuxteam.de wrote:

>> "A general realization." "Nowadays." Ooooh. Can't argue with that.
>> 
>> Consideration of one particular choice has now been successfully
>> diverted toward vacuous generalizations about choice in general -  
>> platitudes of Polonius.  Too bad.
>
> I must say that I agree strongly with this point of view.
>
> More choice is *always* better. The hard part is presenting this
> choice to the user in a way it doesn't overwhelm her/him.

Hm.  Don't want to sound trollish, but as a mathematician I'm strongly
against using the general quantifier where it doesn't belong.  So, not
*always*, but often.  (A good source of counterexamples is ethics.
I wouldn't consider giving my child -- or myself, for that matter --
giving more choice wrt to e.g. drugs a good thing.)

> I have the strong suspicion that the "less choice is good" movement
> we are seeing stems from being too lazy (or at a loss) wrt thinking
> hard about how to guide the user through a rich set of choices. The
> latter would be, in my view, the right thing to do.

+1.

Best,

-- 
Marcin Borkowski
http://octd.wmi.amu.edu.pl/en/Marcin_Borkowski
Faculty of Mathematics and Computer Science
Adam Mickiewicz University

Re: Do we want more choice? (was Bookmark the end of file?)

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Mon, Apr 13, 2015 at 09:53:12AM +0200, Marcin Borkowski wrote:

[...]

> > More choice is *always* better [...]

> Hm.  Don't want to sound trollish, but as a mathematician I'm strongly
> against using the general quantifier where it doesn't belong [...]

"All generalizations suck" ;-)

best
- -- t
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlUremsACgkQBcgs9XrR2kashQCdGsHJLpLY8LeUCIIjXoi+acr9
hFgAn0SyQjGqHdmV3aMQnNfWP6wrG9ln
=0WJF
-----END PGP SIGNATURE-----

Re: Do we want more choice? (was Bookmark the end of file?)

[Eli Zaretskii]

> Date: Mon, 13 Apr 2015 09:17:42 +0200
> From: <tomas@tuxteam.de>
> Cc: Rusi <rustompmody@gmail.com>, help-gnu-emacs@gnu.org
> 
> More choice is *always* better. The hard part is presenting this
> choice to the user in a way it doesn't overwhelm her/him.
> 
> I have the strong suspicion that the "less choice is good" movement
> we are seeing stems from being too lazy (or at a loss) wrt thinking
> hard about how to guide the user through a rich set of choices. The
> latter would be, in my view, the right thing to do.

Volunteers are welcome to come aboard and work on the Emacs manuals
and doc strings to make all this happen.  If the number of people who
work on documentation increases significantly, it's quite probable
that the Emacs maintainers will change their views about "less choice
is good" issue.

Thanks in advance.

Re: Do we want more choice? (was Bookmark the end of file?)

[Stefan Monnier]

> I have the strong suspicion that the "less choice is good" movement
> we are seeing stems from being too lazy (or at a loss) wrt thinking
> hard about how to guide the user through a rich set of choices.  The
> latter would be, in my view, the right thing to do.

There are many ways to interpret "less choice is good".  For me, what it
means is that adding a custom variable, a C-u prefix, or a new command
just to "give more choice" is often actually either laziness or
a reflection of the fact that it's very hard to find a better UI.

Of course, in Emacs we get to enjoy even more cop-outs and more choices,
since most of the "missing choices" are available to the end user,
modulo a few advice-add and enough motivation ;-)


        Stefan

Re: Do we want more choice? (was Bookmark the end of file?)

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Mon, Apr 13, 2015 at 05:52:03PM +0300, Eli Zaretskii wrote:
> > Date: Mon, 13 Apr 2015 09:17:42 +0200
> > From: <tomas@tuxteam.de>
> > Cc: Rusi <rustompmody@gmail.com>, help-gnu-emacs@gnu.org
> > 
> > MOAR CHOICE!

> Volunteers are welcome to come aboard and work on the Emacs manuals
> and doc strings to make all this happen.  If the number of people who
> work on documentation increases significantly, it's quite probable
> that the Emacs maintainers will change their views about "less choice
> is good" issue.

Very well put. Point taken.

Thanks
- -- tomás
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlUsF8UACgkQBcgs9XrR2kaB6ACeJbj8kIkklqoyamQZpYe2X8vg
fT4AnAnpJj5rKJyxwQ+99Pj9y4MSBV4V
=kxcJ
-----END PGP SIGNATURE-----

Re: Do we want more choice? (was Bookmark the end of file?)

[tomas]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Mon, Apr 13, 2015 at 01:20:15PM -0400, Stefan Monnier wrote:
> > I have the strong suspicion that the "less choice is good" movement
> > we are seeing stems from being too lazy [...]

> There are many ways to interpret "less choice is good".  For me, what it
> means is that adding a custom variable, a C-u prefix, or a new command
> just to "give more choice" is often actually either laziness or
> a reflection of the fact that it's very hard to find a better UI.
> 
> Of course, in Emacs we get to enjoy even more cop-outs and more choices,
> since most of the "missing choices" are available to the end user,
> modulo a few advice-add and enough motivation ;-)

Definitely. It takes more thought to offer more choice in a useful
way. I think Emacs is one of the more advanced interfaces wrt this
aspect.

- -- t
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlUsGFcACgkQBcgs9XrR2kZdbACfTyChB0cwxbOQs8p18trIAf0L
npsAn0wiSEUlXW7ZMkGVyOGzgviS0Avt
=XqFB
-----END PGP SIGNATURE-----

Re: Do we want more choice? (was Bookmark the end of file?)

[Gian Uberto Lauri]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 593 bytes --]

Eli Zaretskii writes:

 > Volunteers are welcome to come aboard and work on the Emacs manuals
 > and doc strings to make all this happen.

Where is the enlistment office? :)

-- 
 /\           ___                                    Ubuntu: ancient
/___/\_|_|\_|__|___Gian Uberto Lauri_____               African word
  //--\| | \|  |   Integralista GNUslamico            meaning "I can
\/                 coltivatore diretto di software       not install
     già sistemista a tempo (altrui) perso...                Debian"

Warning: gnome-config-daemon considered more dangerous than GOTO

Re: Do we want more choice? (was Bookmark the end of file?)

[Eli Zaretskii]

> Date: Tue, 14 Apr 2015 09:38:05 +0200
> Cc: help-gnu-emacs@gnu.org
> From: "Gian Uberto Lauri" <saint@eng.it> 
> 
> Eli Zaretskii writes:
> 
>  > Volunteers are welcome to come aboard and work on the Emacs manuals
>  > and doc strings to make all this happen.
> 
> Where is the enlistment office? :)

On emacs-devel@gnu.org.  But you don't have to enlist, you can
contribute changes regardless.

Re: Do we want more choice? (was Bookmark the end of file?)

[Gian Uberto Lauri]

Eli Zaretskii writes:
 > > Date: Tue, 14 Apr 2015 09:38:05 +0200
 > > Cc: help-gnu-emacs@gnu.org
 > > From: "Gian Uberto Lauri" <saint@eng.it> 
 > > 
 > > Eli Zaretskii writes:
 > > 
 > >  > Volunteers are welcome to come aboard and work on the Emacs manuals
 > >  > and doc strings to make all this happen.
 > > 
 > > Where is the enlistment office? :)
 > 
 > On emacs-devel@gnu.org.  But you don't have to enlist, you can
 > contribute changes regardless.

Thank you for the info!

I was joking when I talked about enlisting, BTW they just re run
"Captain America" on tv here...

-- 
 /\           ___                                    Ubuntu: ancient
/___/\_|_|\_|__|___Gian Uberto Lauri_____               African word
  //--\| | \|  |   Integralista GNUslamico            meaning "I can
\/                 coltivatore diretto di software       not install
     già sistemista a tempo (altrui) perso...                Debian"

Warning: gnome-config-daemon considered more dangerous than GOTO

Re: Do we want more choice? (was Bookmark the end of file?)

[Eli Zaretskii]

> Date: Tue, 14 Apr 2015 16:42:09 +0200
> Cc: help-gnu-emacs@gnu.org
> From: "Gian Uberto Lauri" <saint@eng.it> 
> 
> I was joking when I talked about enlisting

What makes you think I wasn't?

Re: Do we want more choice? (was Bookmark the end of file?)

[Gian Uberto Lauri]

Eli Zaretskii writes:
 > > Date: Tue, 14 Apr 2015 16:42:09 +0200
 > > Cc: help-gnu-emacs@gnu.org
 > > From: "Gian Uberto Lauri" <saint@eng.it> 
 > > 
 > > I was joking when I talked about enlisting
 > 
 > What makes you think I wasn't?

What, me thinking?

-- 
 /\           ___                                    Ubuntu: ancient
/___/\_|_|\_|__|___Gian Uberto Lauri_____               African word
  //--\| | \|  |   Integralista GNUslamico            meaning "I can
\/                 coltivatore diretto di software       not install
     già sistemista a tempo (altrui) perso...                Debian"

Warning: gnome-config-daemon considered more dangerous than GOTO