usability issue

usability issue

[Tuomo Takkula]

        Hi.

The department I am working on just switched emacs from 20.xx?? to
21.2.1, with the new cool look. However, the keystrokes <Home> and
<End> changed their meaning as well, from 'go-to-beginning/end-of
document' to 'goto-begin/end-of line', respectively. So, I thought,
lets make an usability test and try to change that back.

So I look under options, find nothing useful, then I try under
'Cutomize Emacs' and find the good old mess that I remembered hating
so much: gazillions of customization groups which have all have in
common that they do not offer the option they promise to offer, with
more than 50% of them having names which I cannot associate any
meaning to (and I am a fairly seasoned emacs user). And of course, no
customization group called 'keystrokes' or such.

So, two questions:
- is there any chance that customization becomes more user friendly? A
good start wood be to avoid for instance names like 'Wp' for
customization groups. What the heck is 'Wp'?
- _is_ there a way to get to the old behaviour of <Home> and <End> via
custimization groups, and if so, where?


        Cheers
        Tuomo

-- 
___
   "Microsoft OS's are good because they encourage Intel to produce
    faster CPUs for the rest of us to run Unix on."
							 George Dau

Re: usability issue

[David Kastrup]

Tuomo Takkula <tuomo@quirm.cs.chalmers.se> writes:

> The department I am working on just switched emacs from 20.xx?? to
> 21.2.1, with the new cool look. However, the keystrokes <Home> and
> <End> changed their meaning as well, from 'go-to-beginning/end-of
> document' to 'goto-begin/end-of line', respectively. So, I thought,
> lets make an usability test and try to change that back.

Why?  Just use C-home and C-end.  You don't need them that often, eh?

> - _is_ there a way to get to the old behaviour of <Home> and <End> via
> custimization groups, and if so, where?

Unlikely.  Those are keystrokes.  Just use

(global-set-key [home] 'beginning-of-buffer)
(global-set-key [end] 'end-of-buffer)

in your .emacs file.  That's all.  But why you should want it when
those functions are available on C-home and C-end escapes me.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: usability issue

[Henrik Enberg]

Tuomo Takkula <tuomo@quirm.cs.chalmers.se> writes:

[...]

> - is there any chance that customization becomes more user friendly? A
> good start wood be to avoid for instance names like 'Wp' for
> customization groups. What the heck is 'Wp'?

Probably, If someone can come up with better names.  The Emacs
maintainers are quite responsive to suggestions of improvements.

> - _is_ there a way to get to the old behaviour of <Home> and <End> via
> custimization groups, and if so, where?

No.  Customize doesn't yet support binding keys.  Putting the following
in your ~/.emacs.el should get the old behavior back.

(global-set-key (kbd "<home>") 'beginning-of-buffer)
(global-set-key (kbd "<end>") 'end-of-buffer)

Re: usability issue

[Alan Shutko]

David Kastrup <dak@gnu.org> writes:

> in your .emacs file.  That's all.  But why you should want it when
> those functions are available on C-home and C-end escapes me.

Because we're used to it?  Why do people want beginning-of-line and
end-of-line on Home and End when they're on C-a and C-e anyway?

-- 
Alan Shutko <ats@acm.org> - In a variety of flavors!
Lower taxes, raise the dead.

Re: usability issue

[Miles Bader]

Alan Shutko <ats@acm.org> writes:
> Because we're used to it?  Why do people want beginning-of-line and
> end-of-line on Home and End when they're on C-a and C-e anyway?

The reason it was changed was that the bol/eol bindings are apparently
what the majority of windows programs use (I don't know if it's actually
true!).

Of course the change is going to annoy users of emacs (or mozilla!) that
are used to the old bindings, but users that are used to windows are
more likely to use those special keys than `veteran' emacs users, and
also less skillful at customization.

-Miles
-- 
`Suppose Korea goes to the World Cup final against Japan and wins,' Moon said.
`All the past could be forgiven.'   [NYT]

Re: usability issue

[Bruce Ingalls]

Tuomo Takkula wrote:
...
You might wish to look at <url: http://emacro.sf.net/ >
which makes Emacs more useable.

It does not solve your problem. However, I am *slowly* getting around to 
the idea of packaging keybinding themes.
These would be standard bindings, such as cua, gnome, classic emacs, etc.

I sure could use some coding help.

As well, some things I am thinking about, include:
Do I package it with EMacro?
Should I bind C-x & M-x functions to the F12 key & modifiers?
How do I deal with I18N and dead key issues?

Re: usability issue

[Tuomo Takkula]

David Kastrup <dak@gnu.org> writes:

> Tuomo Takkula <tuomo@quirm.cs.chalmers.se> writes:
> 
> > The department I am working on just switched emacs from 20.xx?? to
> > 21.2.1, with the new cool look. However, the keystrokes <Home> and
> > <End> changed their meaning as well, from 'go-to-beginning/end-of
> > document' to 'goto-begin/end-of line', respectively. So, I thought,
> > lets make an usability test and try to change that back.
> 
> Why?  Just use C-home and C-end.  You don't need them that often, eh?

That has to do with my age. I am used to Home and End jumping up and
down, and I am not used to software telling me what I am supposed to,
as in 'der Schwanz hat nicht mit dem Hund zu wedeln' and such ;-)

> > - _is_ there a way to get to the old behaviour of <Home> and <End> via
> > custimization groups, and if so, where?
> 
> Unlikely.  Those are keystrokes.  Just use
> 
> (global-set-key [home] 'beginning-of-buffer)
> (global-set-key [end] 'end-of-buffer)
> 
> in your .emacs file.  That's all.  

Thanks. Copied. Cool - works fine!

> But why you should want it when those functions are available on
> C-home and C-end escapes me.

I was used to the old behaviour and I might switch to the new if I see
any advantage in the new behaviour. Right now, I just feel annoyed,
for no benefit, so I stick to the old. That's reasonable (in my age,
y'know...)


        Cheers
        Tuomo


PS. Btw, thanks for the great work on PreviewLateX, and the advice you
give in c.t.t in general


> -- 
> David Kastrup, Kriemhildstr. 15, 44793 Bochum

--
Chalmers University of Technology
Computing Science
Eklandagatan 86, S-41296 Göteborg, Sweden
tuomo@cs.chalmers.se, +46-31-772 1052 phone, +46-31-165 655 fax

Re: usability issue

[Ehud Karni]

On Mon, 06 Jan 2003 22:52:55 GMT, Alan Shutko <ats@acm.org> wrote:
> 
> Why do people want beginning-of-line and
> end-of-line on Home and End when they're on C-a and C-e anyway?

Because beginning-of-line and end-of-line are more useful and we
want to use only 1 key for these functions.

`beginning-of-buffer' and `end-of-buffer' where on M-< and M->
forever, why you avoided using them ?

One of the first things I did when I started to use Emacs (~1992)
was to change the binding of Home and End to beginning/end of line !
(That was binding of the Control Data FSE I used up to that date).

Ehud.


-- 
 Ehud Karni           Tel: +972-3-7966-561  /"\
 Mivtach - Simon      Fax: +972-3-7966-667  \ /  ASCII Ribbon Campaign
 Insurance agencies   (USA) voice mail and   X   Against   HTML   Mail
 http://www.mvs.co.il  FAX:  1-815-5509341  / \
 mailto:ehud@unix.mvs.co.il                  Better  Safe  Than  Sorry

Re: usability issue

[Kai Großjohann]

"Ehud Karni" <ehud@unix.mvs.co.il> writes:

> Because beginning-of-line and end-of-line are more useful and we
> want to use only 1 key for these functions.

I once wrote something that allows you to hit <home> once to go to
beginning of line, twice in a row to go to the beginning of the
window, and thrice to go to the beginning of the buffer.

The implementation sucks somewhat, I'm afraid.

But I think the feature is good, so WIBNI this functionality came
standard with Emacs?  What do people think.
-- 
Ambibibentists unite!

Re: usability issue

[David Kastrup]

kai.grossjohann@uni-duisburg.de (Kai Großjohann) writes:

> "Ehud Karni" <ehud@unix.mvs.co.il> writes:
> 
> > Because beginning-of-line and end-of-line are more useful and we
> > want to use only 1 key for these functions.
> 
> I once wrote something that allows you to hit <home> once to go to
> beginning of line, twice in a row to go to the beginning of the
> window, and thrice to go to the beginning of the buffer.
> 
> The implementation sucks somewhat, I'm afraid.
> 
> But I think the feature is good, so WIBNI this functionality came
> standard with Emacs?  What do people think.

I'd think beginning of window odious.  I would recommend start of
line, start of section/defun, start of buffer.

As to the implementation: I would recommend to make it stateless: the
implementation will simply check whether the cursor currently is at
the start of line/section and move to the beginning of the smallest
unit for which it is not yet already at the beginning.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: usability issue

[Andrew]

David Kastrup <dak@gnu.org> writes:

> kai.grossjohann@uni-duisburg.de (Kai Großjohann) writes:
> 
> > "Ehud Karni" <ehud@unix.mvs.co.il> writes:
> > 
> > > Because beginning-of-line and end-of-line are more useful and we
> > > want to use only 1 key for these functions.
> > 
> > I once wrote something that allows you to hit <home> once to go to
> > beginning of line, twice in a row to go to the beginning of the
> > window, and thrice to go to the beginning of the buffer.
> > 
> > The implementation sucks somewhat, I'm afraid.
> > 
> > But I think the feature is good, so WIBNI this functionality came
> > standard with Emacs?  What do people think.
> 
> I'd think beginning of window odious.  I would recommend start of
> line, start of section/defun, start of buffer.
> 
> As to the implementation: I would recommend to make it stateless: the
> implementation will simply check whether the cursor currently is at
> the start of line/section and move to the beginning of the smallest
> unit for which it is not yet already at the beginning.
> 
> -- 
> David Kastrup, Kriemhildstr. 15, 44793 Bochum

I saw this message and thought it would be an interesting (and useful) feature 
to code up, as I am attempting to learn elisp; i just don't know how to find
if I'm at the beginning of a line. Is there a function; or does it need to be 
written specially? 

Cheers,
Andrew

Re: usability issue

[John Paul Wallington]

Andrew <andrew@uberwald.discnet> wrote:

> i just don't know how to find if I'm at the beginning of a line. Is
> there a function; or does it need to be written specially?

See `bolp'.

-- 
John Paul Wallington

Re: usability issue

[Andrew]

John Paul Wallington wrote:
> Andrew <andrew@uberwald.discnet> wrote:
> 
> 
>>i just don't know how to find if I'm at the beginning of a line. Is
>>there a function; or does it need to be written specially?
> 
> 
> See `bolp'.
> 

for the sack of politness; heres my version of the functions (only line 
and buffer sorry) feel free to do whatever.

(defun beginning-of-line-or-buffer ()
   "Sends point to the beginning of the current line, or, if already there,
    the beginning of the buffer"
   (interactive)
   (if (bolp)
       (goto-char (point-min))
     (beginning-of-line)
     ))

(defun end-of-line-or-buffer ()
   "Sends point to the beginning of the current line, or, if already there,
    the beginning of the buffer"
   (interactive)
   (if (eolp)
       (goto-char (point-max))
     (end-of-line)
     ))

Andrew

Re: usability issue

[David Kastrup]

Andrew <gr1dl0ck@deviantart.com> writes:

> John Paul Wallington wrote:
> > Andrew <andrew@uberwald.discnet> wrote:
> > 
> >>i just don't know how to find if I'm at the beginning of a line. Is
> >>there a function; or does it need to be written specially?
> > See `bolp'.
> > 
> 
> for the sack of politness; heres my version of the functions (only
> line and buffer sorry) feel free to do whatever.
> 
> (defun beginning-of-line-or-buffer ()
>    "Sends point to the beginning of the current line, or, if already there,
>     the beginning of the buffer"
>    (interactive)
>    (if (bolp)
>        (goto-char (point-min))
>      (beginning-of-line)
>      ))
> 
> (defun end-of-line-or-buffer ()
>    "Sends point to the beginning of the current line, or, if already there,
>     the beginning of the buffer"
>    (interactive)
>    (if (eolp)
>        (goto-char (point-max))
>      (end-of-line)
>      ))

I think at least the far jumps should place a mark so that C-x C-x
will get you back.  So you would want to use beginning-of-buffer
instead of (goto-char (point-min)) and the corresponding for the end.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: usability issue

[Andrew Brehaut]

David Kastrup wrote:
> Andrew <gr1dl0ck@deviantart.com> writes:
> 
> 
>>John Paul Wallington wrote:
>>
>>>Andrew <andrew@uberwald.discnet> wrote:
>>>
>>>
>>>>i just don't know how to find if I'm at the beginning of a line. Is
>>>>there a function; or does it need to be written specially?
>>>
>>>See `bolp'.
>>>
>>
>>for the sack of politness; heres my version of the functions (only
>>line and buffer sorry) feel free to do whatever.
>>
>>(defun beginning-of-line-or-buffer ()
>>   "Sends point to the beginning of the current line, or, if already there,
>>    the beginning of the buffer"
>>   (interactive)
>>   (if (bolp)
>>       (goto-char (point-min))
>>     (beginning-of-line)
>>     ))
>>
>>(defun end-of-line-or-buffer ()
>>   "Sends point to the beginning of the current line, or, if already there,
>>    the beginning of the buffer"
>>   (interactive)
>>   (if (eolp)
>>       (goto-char (point-max))
>>     (end-of-line)
>>     ))
> 
> 
> I think at least the far jumps should place a mark so that C-x C-x
> will get you back.  So you would want to use beginning-of-buffer
> instead of (goto-char (point-min)) and the corresponding for the end.
> 

Thanks for that; the info pages were kinda missleading (perhaps only for 
a beginner though).

Andrew

Re: usability issue

[John Paul Wallington]

Andrew <gr1dl0ck@deviantart.com> wrote:

>> See `bolp'.

> for the sack of politness; 

My reply was brusque.  I hope you didn't find it impolite.

>                            heres my version of the functions (only
> line and buffer sorry) feel free to do whatever.

Your functions are neat.  Two nits: the first line of a doc string
should "stand on it own" (ideally ending in a period) because
`apropos' will return it in its output, and that the stray parens look
funky because the convention is to cram all the closing parens
together on the same line.

-- 
John Paul Wallington

Re: usability issue

[David Kastrup]

Andrew Brehaut <gr1dl0ck@deviantart.com> writes:

> David Kastrup wrote:
> > Andrew <gr1dl0ck@deviantart.com> writes:
> > 
> >>John Paul Wallington wrote:
> >>
> >>>Andrew <andrew@uberwald.discnet> wrote:
> >>>
> >>>
> >>>>i just don't know how to find if I'm at the beginning of a line. Is
> >>>>there a function; or does it need to be written specially?
> >>>
> >>>See `bolp'.
> >>>
> >>
> >>for the sack of politness; heres my version of the functions (only
> >>line and buffer sorry) feel free to do whatever.
> >>
> >>(defun beginning-of-line-or-buffer ()
> >>   "Sends point to the beginning of the current line, or, if already there,
> >>    the beginning of the buffer"
> >>   (interactive)
> >>   (if (bolp)
> >>       (goto-char (point-min))
> >>     (beginning-of-line)
> >>     ))
> >>
> >>(defun end-of-line-or-buffer ()
> >>   "Sends point to the beginning of the current line, or, if already there,
> >>    the beginning of the buffer"
> >>   (interactive)
> >>   (if (eolp)
> >>       (goto-char (point-max))
> >>     (end-of-line)
> >>     ))
> > I think at least the far jumps should place a mark so that C-x C-x
> > will get you back.  So you would want to use beginning-of-buffer
> > instead of (goto-char (point-min)) and the corresponding for the end.
> > 
> 
> Thanks for that; the info pages were kinda missleading (perhaps only
> for a beginner though).

And one last refinement: beginning-of-buffer is able to take a prefix
argument with several different meanings.  While beginning-of-line
can also take such an argument, it is more or less a safe bet that
you would rarely do so in interactive use, and you still could with a
combination of cursor-up with argument and the Pos1 key, anyhow.

So if we get a prefix-argument, we may assume
a) the beginning-of-buffer functionality is wanted
b) it is wanted with the given argument.

So you would want to make something like

(defun beginning-of-line-or-buffer (&optional arg)
   "Send point to beginning of line or buffer.
If point is not yet at the beginning of the current line, it is
placed there.  If already there, it is sent to the beginning of
the buffer using `beginning-of-buffer'.  If a prefix argument ARG
is given, it is passed to `beginning-of-buffer' which gets then
called regardless of whether point is at the beginning of a line."
   (interactive "P")
   (if (or arg (bolp))
       (beginning-of-buffer arg)
     (beginning-of-line)))

Notice that I also changed the incorrect documentation string.  You
can check those sort of strings by using
M-x checkdoc RET

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: usability issue

[Andrew Brehaut]

Thanks for all the pointers; Its really appreciated.
Interestingly, both on IRC and newsgroups, LISP coders seem to be the 
most polite and helpful toward beginners.

Andrew


> 
> And one last refinement: beginning-of-buffer is able to take a prefix
> argument with several different meanings.  While beginning-of-line
> can also take such an argument, it is more or less a safe bet that
> you would rarely do so in interactive use, and you still could with a
> combination of cursor-up with argument and the Pos1 key, anyhow.
> 
> So if we get a prefix-argument, we may assume
> a) the beginning-of-buffer functionality is wanted
> b) it is wanted with the given argument.
> 
> So you would want to make something like
> 
> (defun beginning-of-line-or-buffer (&optional arg)
>    "Send point to beginning of line or buffer.
> If point is not yet at the beginning of the current line, it is
> placed there.  If already there, it is sent to the beginning of
> the buffer using `beginning-of-buffer'.  If a prefix argument ARG
> is given, it is passed to `beginning-of-buffer' which gets then
> called regardless of whether point is at the beginning of a line."
>    (interactive "P")
>    (if (or arg (bolp))
>        (beginning-of-buffer arg)
>      (beginning-of-line)))
> 
> Notice that I also changed the incorrect documentation string.  You
> can check those sort of strings by using
> M-x checkdoc RET
> 

Re: usability issue

[Colin Marquardt]

jpw@shootybangbang.com (John Paul Wallington) writes:

> Your functions are neat.  Two nits: the first line of a doc string
> should "stand on it own" (ideally ending in a period) because
> `apropos' will return it in its output, and that the stray parens look
> funky because the convention is to cram all the closing parens
> together on the same line.

FWIW, M-x checkdoc can help to spot such things, and also develock.

Cheers,
  Colin