Re: Annoying new features in newer Oorts

Re: Annoying new features in newer Oorts

[Sean Neakums]

commence  Thomas Skogestad quotation:

[...]
> 2. When editing headers in a message buffer, C-a and C-e do not go
> the beginning or end of the line, but to the colon.

Personally, I like that feature.  I never change the name of the
header, always its content.  Only C-a has different behaviour, by the
way.  C-e is still bound to the vanilla end-of-line.

-- 
 /////////////////  |                  | The spark of a pin
<sneakums@zork.net> |  (require 'gnu)  | dropping, falling feather-like.
 \\\\\\\\\\\\\\\\\  |                  | There is too much noise.

Re: Annoying new features in newer Oorts

[Romain FRANCOISE]

Thomas Skogestad writes:

> * Sean Neakums
> | Personally, I like that feature.

> I hate it.

Well, I like too. I guess you can just rebind C-a to `beginning-of-line'
in Message mode, and you'll be set.

-- 
Romain FRANCOISE <romain@orebokech.com>
it's a miracle! -- http://www.orebokech.com/

Re: Annoying new features in newer Oorts

[Fabien Penso]

[-- Attachment #1: Type: text/plain, Size: 438 bytes --]


Romain on Tue, 30 Apr 2002 13:14:46 +0200 wrote:

 > Thomas Skogestad writes:
 >> * Sean Neakums
 >> | Personally, I like that feature.

 >> I hate it.

 > Well, I like too. I guess you can just rebind C-a to `beginning-of-line'
 > in Message mode, and you'll be set.

or hit C-a twice.

-- 
Fabien Penso <penso@linuxfr.org> | LinuxFr a toujours besoin de :
http://perso.LinuxFr.org/penso/  | http://linuxFr.org/dons/


[-- Attachment #2: Type: application/pgp-signature, Size: 239 bytes --]

Re: Annoying new features in newer Oorts

[Matthieu Moy]

Thomas Skogestad <tskogest@jusstud.uio.no> writes:

> I either delete a header, or when I edit a header ... I can't think of a
> time I've changed the text directly after the colon.

Did you  never change  your mind  about the recipient  of your  mail ?
Then, C-a, C-k, and type a new content. This seems natural to me. 

If you wish to go to the  beginning of line, then, C-a twice will make
you happy. 

-- 
Matthieu

Re: Annoying new features in newer Oorts

[Sean Neakums]

commence  Matthieu Moy quotation:

> If you wish to go to the  beginning of line, then, C-a twice will make
> you happy. 

Or this:

  (define-key message-mode-map "\C-a" 'beginning-of-line)

-- 
 /////////////////  |                  | The spark of a pin
<sneakums@zork.net> |  (require 'gnu)  | dropping, falling feather-like.
 \\\\\\\\\\\\\\\\\  |                  | There is too much noise.

Re: Annoying new features in newer Oorts

[Sean Neakums]

commence  Thomas Skogestad quotation:

> | If you wish to go to the beginning of line, then, C-a twice will
> | make you happy.
>
> I was happy before.

Then change the binding as suggested in two separate posts, and get on
with your life.

-- 
 /////////////////  |                  | The spark of a pin
<sneakums@zork.net> |  (require 'gnu)  | dropping, falling feather-like.
 \\\\\\\\\\\\\\\\\  |                  | There is too much noise.

Re: Annoying new features in newer Oorts

[Karl Eichwalder]

Sean Neakums <sneakums@zork.net> writes:

> Then change the binding as suggested in two separate posts, and get on
> with your life.

Note, it isn't nice to steal default keys.  Emacs, for good reason, is
very conservative.

Some Gnus hackers claimed C-a must mimick current shell-mode behavior--I
opposed because shell-mode is completely different: nobody will remove
the prompt in shell-mode, but it happens quite often that people have
to remove a complete header line.

Instead of stealing C-a you should have introduced a previously
_unused_ binding for the new behavior.

-- 
ke@suse.de (work) / keichwa@gmx.net (home):              |
http://www.suse.de/~ke/                                  |      ,__o
Free Translation Project:                                |    _-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/             |   (*)/'(*)

Re: Annoying new features in newer Oorts

[Florian Weimer]

Karl Eichwalder <ke@gnu.franken.de> writes:

> Instead of stealing C-a you should have introduced a previously
> _unused_ binding for the new behavior.

C-a already has a similar behavior in other buffers...

Re: Annoying new features in newer Oorts

[Sean Neakums]

commence  Karl Eichwalder quotation:

> Sean Neakums <sneakums@zork.net> writes:
>> Then change the binding as suggested in two separate posts, and get
>> on with your life.
>
> Note, it isn't nice to steal default keys.  Emacs, for good reason,
> is very conservative.

This is the most ridiculous argument I have ever heard.  I suppose you
go around rebinding TAB to self-insert-command in all your programming
modes?

-- 
 /////////////////  |                  | The spark of a pin
<sneakums@zork.net> |  (require 'gnu)  | dropping, falling feather-like.
 \\\\\\\\\\\\\\\\\  |                  | There is too much noise.

Re: Annoying new features in newer Oorts

[Karl Eichwalder]

Sean Neakums <sneakums@zork.net> writes:

> I suppose you go around rebinding TAB to self-insert-command in all
> your programming modes?

Of course, I do not.  I'm using Emacs since 18.xx and TAB used to do
indentation according to mode all the years and releases.

The question simply is: where does a header line start?  Some of us
think it starts in column 0, others think it starts after ": ".  Both
parties are "right"; in those case it's good habbit to leave things as
are.

-- 
ke@suse.de (work) / keichwa@gmx.net (home):              |
http://www.suse.de/~ke/                                  |      ,__o
Free Translation Project:                                |    _-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/             |   (*)/'(*)

Re: Annoying new features in newer Oorts

[Kai Großjohann]

Karl Eichwalder <ke@gnu.franken.de> writes:

> The question simply is: where does a header line start?  Some of us
> think it starts in column 0, others think it starts after ": ".  Both
> parties are "right"; in those case it's good habbit to leave things as
> are.

This is not the question, IMHO.  The question is, when in a header,
what would be a convenient place for C-a to go to?  And the current
behavior gets you to the colon with one C-a, and to the beginning of
the line with two C-a.  The old behavior got you to the beginning to
the line with one C-a, but getting to the colon was difficult.

One could try to make a user option, but given the existence of
define-key, this seems a bit superfluous.

kai
-- 
Silence is foo!

Re: Annoying new features in newer Oorts

[Matt Armstrong]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Karl Eichwalder <ke@gnu.franken.de> writes:
>
>> The question simply is: where does a header line start?  Some of us
>> think it starts in column 0, others think it starts after ": ".  Both
>> parties are "right"; in those case it's good habbit to leave things as
>> are.
>
> This is not the question, IMHO.  The question is, when in a header,
> what would be a convenient place for C-a to go to?  And the current
> behavior gets you to the colon with one C-a, and to the beginning of
> the line with two C-a.  The old behavior got you to the beginning to
> the line with one C-a, but getting to the colon was difficult.
>
> One could try to make a user option, but given the existence of
> define-key, this seems a bit superfluous.

I like the new C-a 


Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> WINI C-a in body could have added magic, too?  In a cited line, it
> could go to the end of the ">>" prefix.
>
> kai
> -- 
> Silence is foo!
>

-- 
Don't send mail to Marcia.Smith@hole.lickey.com
The address is there for spammers to harvest.

Re: Annoying new features in newer Oorts

[Matt Armstrong]

(if I've sent this twice, disregard the first and read this one, sorry)

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Karl Eichwalder <ke@gnu.franken.de> writes:
>
>> The question simply is: where does a header line start?  Some of us
>> think it starts in column 0, others think it starts after ": ".  Both
>> parties are "right"; in those case it's good habbit to leave things as
>> are.
>
> This is not the question, IMHO.  The question is, when in a header,
> what would be a convenient place for C-a to go to?  And the current
> behavior gets you to the colon with one C-a, and to the beginning of
> the line with two C-a.  The old behavior got you to the beginning to
> the line with one C-a, but getting to the colon was difficult.
>
> One could try to make a user option, but given the existence of
> define-key, this seems a bit superfluous.

I notice that C-a toggles between the end of line and just after the
colon.  Could we change C-a to go to the beginning of the line first
and to the colon when already there?  This way you'd do C-a C-a to get
the magical behavior and people's hard coded brain pathways would not
be strained as hard.

I like the magical behavior, but I have found it hard to retrain my
brain.  I seem to always end up killing a header line while not
actually being at the beginning of the line.

The other instances of "magical C-a" in Emacs that I'm aware of are
all cases where a shell or program prompt is displayed.  It arguably
never makes sense to go the left past a prompt, while it does make
sense to do so in the message header.


Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> WINI C-a in body could have added magic, too?  In a cited line, it
> could go to the end of the ">>" prefix.

Ditto Josh's comments here.


-- 
Don't send mail to Marcia.Smith@hole.lickey.com
The address is there for spammers to harvest.

Re: Annoying new features in newer Oorts

[Karl Eichwalder]

Matt Armstrong <matt@lickey.com> writes:

> Could we change C-a to go to the beginning of the line first and to
> the colon when already there?

Thanks for your analyses.  Yes, I'm all for it.

-- 
ke@suse.de (work) / keichwa@gmx.net (home):              |
http://www.suse.de/~ke/                                  |      ,__o
Free Translation Project:                                |    _-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/             |   (*)/'(*)

Re: Annoying new features in newer Oorts

[Sean Neakums]

commence  Matt Armstrong quotation:

> Could we change C-a to go to the beginning of the line first
> and to the colon when already there?  This way you'd do C-a C-a to get
> the magical behavior and people's hard coded brain pathways would not
> be strained as hard.

I think it's better that C-a always moves point to the left, or not at
all if it's at the beginning of the line.  The current behaviour of
C-a in header lines is also easily "discoverable" by new users.  Who
would ever imagine that typing C-a at the beginning of a line would
move point to the right?

-- 
 /////////////////  |                  | The spark of a pin
<sneakums@zork.net> |  (require 'gnu)  | dropping, falling feather-like.
 \\\\\\\\\\\\\\\\\  |                  | There is too much noise.

Re: Annoying new features in newer Oorts

[Florian Weimer]

Sean Neakums <sneakums@zork.net> writes:

> Who would ever imagine that typing C-a at the beginning of a line
> would move point to the right?

BRIEF users? ;-)

Re: Annoying new features in newer Oorts

[Sean Neakums]

commence  Florian Weimer quotation:

> Sean Neakums <sneakums@zork.net> writes:
>> Who would ever imagine that typing C-a at the beginning of a line
>> would move point to the right?
>
> BRIEF users? ;-)

I knew a BRIEF user once.  I finally got an excuse to use up the fava
beans I had been keeping.

-- 
 /////////////////  |                  | The spark of a pin
<sneakums@zork.net> |  (require 'gnu)  | dropping, falling feather-like.
 \\\\\\\\\\\\\\\\\  |                  | There is too much noise.

Re: Annoying new features in newer Oorts

[Kai Großjohann]

Matt Armstrong <matt@lickey.com> writes:

> I notice that C-a toggles between the end

ITYM beginning?

> of line and just after the colon.  Could we change C-a to go to the
> beginning of the line first and to the colon when already there?

Maybe that's a good idea.  Though it seems strange for me to see C-a
move forward.  (I didn't know it was doing that until just now, when
I tried.)  Hm.

kai
-- 
Silence is foo!

Re: Annoying new features in newer Oorts

[Steinar Bang]

>>>>> Matt Armstrong <matt@lickey.com>:

> I notice that C-a toggles between the end of line and just after the
> colon.  Could we change C-a to go to the beginning of the line first
> and to the colon when already there?  This way you'd do C-a C-a to
> get the magical behavior and people's hard coded brain pathways
> would not be strained as hard.

Good suggestion.  I'm for it.

Re: Annoying new features in newer Oorts

[Simon Josefsson]

Steinar Bang <sb@dod.no> writes:

>>>>>> Matt Armstrong <matt@lickey.com>:
>
>> I notice that C-a toggles between the end of line and just after the
>> colon.  Could we change C-a to go to the beginning of the line first
>> and to the colon when already there?  This way you'd do C-a C-a to
>> get the magical behavior and people's hard coded brain pathways
>> would not be strained as hard.
>
> Good suggestion.  I'm for it.

I like this approach better as well.

OTOH, it seems as if Emacs in general is moving away from C-a meaning
start of line, so the new message behaviour is consistent with the
rest of modern emacs.  I like the behaviour in the shell mode, and in
minibuffer prompts, but it feels wrong in message mode.

The same behaviour could be used when jumping passed quote prefixes as
well. C-a goes to start of line, C-a again goes to start of quoted
text.

Re: Annoying new features in newer Oorts

[Steinar Bang]

>>>>> Sean Neakums <sneakums@zork.net>:

> commence  Karl Eichwalder quotation:
>> Sean Neakums <sneakums@zork.net> writes:

>>> Then change the binding as suggested in two separate posts, and
>>> get on with your life.

>> Note, it isn't nice to steal default keys.  Emacs, for good reason,
>> is very conservative.

> This is the most ridiculous argument I have ever heard.

I don't think it is ridiculous.  I think his argument makes perfect
sense.  The current `C-a' in message mode, breaks with my expectation
of `C-a'.

I know I can rebind the keybinding, and I even know how to do it.

The discussion is about whether the current `C-a' in message mode is a
good binding or not.  My vote is that it isn't.

Re: Annoying new features in newer Oorts

[Steinar Bang]

>>>>> Thomas Skogestad <tskogest@jusstud.uio.no>:

> * Sean Neakums
>> Personally, I like that feature.

> I hate it.

So do I.

[snip!]
> I either delete a header, or when I edit a header ... I can't think
> of a time I've changed the text directly after the colon.

AOL!

Re: Annoying new features in newer Oorts

[Martin Sandiford]

On Tue, 30 Apr 2002, tskogest@jusstud.uio.no wrote:
> 2. When editing headers in a message buffer, C-a and C-e do not go the
> beginning or end of the line, but to the colon.

C-e still works the same way, I think.

Two C-a in succession will take you to the beginning of the line.

Martin

-- 

Re: Annoying new features in newer Oorts

[Kai Großjohann]

Thomas Skogestad <tskogest@jusstud.uio.no> writes:

> 1. When entering a group buffer, and I specify that I want to read read
> messages also, Gnus goes to the first unread message, not to the beginning
> of the buffer.

There's a variable for this.  Hm.  It's not gnus-auto-select-first,
but similar.  Can't remember right now.  But I'm sure it's in the
GNUS-NEWS file.

> 2. When editing headers in a message buffer, C-a and C-e do not go the
> beginning or end of the line, but to the colon.

You can hit C-a twice to go to the beginning of the line.

I really like this feature, because it makes it easy to edit the
contents of a header.  But maybe that's because I have another
keybinding (S-<backspace>) which deletes the current line, so I use
that for removing headers.

kai
-- 
Silence is foo!

Re: Annoying new features in newer Oorts

[Andreas Fuchs]

On 2002-04-30, Kai Großjohann <Kai.Grossjohann@cs.uni-dortmund.de>
wrote:
> Thomas Skogestad <tskogest@jusstud.uio.no> writes:
>> 1. When entering a group buffer, and I specify that I want to read
>>    read messages also, Gnus goes to the first unread message, not to
>>    the beginning of the buffer.

> There's a variable for this.  Hm.  It's not gnus-auto-select-first,
> but similar.  Can't remember right now.  But I'm sure it's in the
> GNUS-NEWS file.

That's strange, it isn't in GNUS-NEWS, and neither could I find it it
with apropos. I actually quite liked this feature, but as of recent CVS
gnusae, it just doesn't go to the first unread article. It feels like
the feature just wasn't ever there. I also grepped through the
ChangeLog and CVS logs of gnus-sum, and couldn't find anything.

Is there any trace of this feature? (-:

Thanks,
-- 
Andreas Fuchs, <asf@acm.org>, asf@jabber.at, antifuchs

Re: Annoying new features in newer Oorts

[Mark Thomas]

On Tue, 30 Apr 2002, Kai.Grossjohann@CS.Uni-Dortmund.DE wrote:

> Thomas Skogestad <tskogest@jusstud.uio.no> writes:
> 
>> 1. When entering a group buffer, and I specify that I want to read
>>    read messages also, Gnus goes to the first unread message, not
>>    to the beginning of the buffer.
> 
> There's a variable for this.  Hm.  It's not gnus-auto-select-first,
> but similar.  Can't remember right now.  But I'm sure it's in the
> GNUS-NEWS file.

Are you thinking of gnus-auto-select-subject?

  Documentation:
  *Says what subject to place under point when entering a group.
  
  This variable can either be the symbols `first' (place point on the
  first subject), `unread' (place point on the subject line of the first
  unread article), `best' (place point on the subject line of the
  higest-scored article), `unseen' (place point on the subject line of
  the first unseen article), 'unseen-or-unread' (place point on the subject
  line of the first unseen article or, if all article have been seen, on the
  subject line of the first unread article), or a function to be called to
  place point on some subject line.

-Mark

Re: Annoying new features in newer Oorts

[Andreas Fuchs]

[-- Attachment #1: Type: text/plain, Size: 498 bytes --]

Today, Mark Thomas <swoon@bellatlantic.net> wrote:
>> There's a variable for this.  Hm.  It's not gnus-auto-select-first,
>> but similar.  Can't remember right now.  But I'm sure it's in the
>> GNUS-NEWS file.
> 
> Are you thinking of gnus-auto-select-subject?
> 

Yes, and I now know why it had just went away. I have had an old
version of CVS gnus lying around (Oort v0.04), and didn't notice
it. Silly me.

Thanks a lot,
-- 
Andreas Fuchs, <asf@acm.org>, asf@jabber.at, antifuchs

[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

Re: Annoying new features in newer Oorts

[Kai Großjohann]

Mark Thomas <swoon@bellatlantic.net> writes:

> Are you thinking of gnus-auto-select-subject?

Thank you!  That's it.

kai
-- 
Silence is foo!

Re: Annoying new features in newer Oorts

[Kai Großjohann]

Thomas Skogestad <tskogest@jusstud.uio.no> writes:

> 2. When editing headers in a message buffer, C-a and C-e do not go the
> beginning or end of the line, but to the colon.

(require 'message)
(define-key message-mode-map (kbd "C-a") nil)

kai
-- 
Silence is foo!