counter-intuitive key bindings

counter-intuitive key bindings

[Leo]

(info "(org)Multiple sets in one file")
| `C-S-<right>'
| `C-S-<left>'
|      These keys jump from one TODO subset to the next.  In the above
|      example, `C-S-<right>' would jump from `TODO' or `DONE' to
|      `REPORT', and any of the words in the second row to `CANCELED'.  

These key bindings are to move from one set of the TODO items to
another, which intuitively are a vertical motion. See this configure:

|      (setq org-todo-keywords
|            '((sequence "TODO" "|" "DONE")
|              (sequence "REPORT" "BUG" "KNOWNCAUSE" "|" "FIXED")
|              (sequence "|" "CANCELED")))

I would propose to change these key bindings to:

   C-S-<up/down>

HTH,
-- 
Leo <sdl.web AT gmail.com>                         (GPG Key: 9283AA3F)

Re: counter-intuitive key bindings

[Bastien]

Leo <sdl.web@gmail.com> writes:

> These key bindings are to move from one set of the TODO items to
> another, which intuitively are a vertical motion. 

Well, I don't share this intuition.

Since S-<left/right> cycles through TODO states, it's consistent to use
C-S-<left/right> to cycle through TODO-states sets. 

And since S-<up/down> changes the priority state, i guess C-S-<up/down>
could cycle through *priority-states sets* - if any. (I actually have no
idea whether priority-states sets would be useful to someone.)

Regards,

-- 
Bastien

Re: counter-intuitive key bindings

[Leo]

On 2007-07-28 12:43 +0100, Bastien wrote:
> Leo <sdl.web@gmail.com> writes:
>
>> These key bindings are to move from one set of the TODO items to
>> another, which intuitively are a vertical motion. 
>
> Well, I don't share this intuition.
>
> Since S-<left/right> cycles through TODO states, it's consistent to use
> C-S-<left/right> to cycle through TODO-states sets. 
>
> And since S-<up/down> changes the priority state, i guess C-S-<up/down>
> could cycle through *priority-states sets* - if any. (I actually have no
> idea whether priority-states sets would be useful to someone.)

In this case, I feel there should a consistent definition of key
bindings. One excellent example is vc, it is just so easy to remember.

I have been an regular user of org, and I have already forgotten that
S-up/down does priority change after one month's break.

-- 
Leo <sdl.web AT gmail.com>                         (GPG Key: 9283AA3F)

Re: Re: counter-intuitive key bindings

[Carsten Dominik]

On Jul 28, 2007, at 20:35, Leo wrote:

> On 2007-07-28 12:43 +0100, Bastien wrote:
>> Leo <sdl.web@gmail.com> writes:
>>
>>> These key bindings are to move from one set of the TODO items to
>>> another, which intuitively are a vertical motion.
>>
>> Well, I don't share this intuition.
>>
>> Since S-<left/right> cycles through TODO states, it's consistent to 
>> use
>> C-S-<left/right> to cycle through TODO-states sets.
>>
>> And since S-<up/down> changes the priority state, i guess 
>> C-S-<up/down>
>> could cycle through *priority-states sets* - if any. (I actually have 
>> no
>> idea whether priority-states sets would be useful to someone.)
>
> In this case, I feel there should a consistent definition of key
> bindings. One excellent example is vc, it is just so easy to remember.

What do you mean with "consistent definition of key bindings".

> I have been an regular user of org, and I have already forgotten that
> S-up/down does priority change after one month's break.

I have been trying very hard to make the key bindings as logical
as possible - but there is a huge amount of commands in org,
and many keys are needed.

That said, the S-up/down for priorities does leas to a couple
of inconsistencies, in particular between agenda and org-mode buffers.

- Carsten


>
> -- 
> Leo <sdl.web AT gmail.com>                         (GPG Key: 9283AA3F)
>
>
>
> _______________________________________________
> Emacs-orgmode mailing list
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode
>
>

--
Carsten Dominik
Sterrenkundig Instituut "Anton Pannekoek"
Universiteit van Amsterdam
Kruislaan 403
NL-1098SJ Amsterdam
phone: +31 20 525 7477

Re: counter-intuitive key bindings

[Leo]

On 2007-08-09 06:05 +0100, Carsten Dominik wrote:
>> In this case, I feel there should a consistent definition of key
>> bindings. One excellent example is vc, it is just so easy to remember.
>
> What do you mean with "consistent definition of key bindings".
>
>> I have been an regular user of org, and I have already forgotten that
>> S-up/down does priority change after one month's break.
>
> I have been trying very hard to make the key bindings as logical
> as possible - but there is a huge amount of commands in org,
> and many keys are needed.
>
> That said, the S-up/down for priorities does leas to a couple
> of inconsistencies, in particular between agenda and org-mode buffers.
>
> - Carsten

I know it is difficult to come up with good key bindings for so many
commands that org-mode has. That's why I gave the example of vc, which
has extremely excellent key bindings. I would never have to learn them a
second time. I hope someone can come up with something similar for org.

-- 
Leo <sdl.web AT gmail.com>                         (GPG Key: 9283AA3F)

         Gnus is one component of the Emacs operating system.

Re: Re: counter-intuitive key bindings

[Carsten Dominik]

On Aug 15, 2007, at 18:57, Leo wrote:

> On 2007-08-09 06:05 +0100, Carsten Dominik wrote:
>>> In this case, I feel there should a consistent definition of key
>>> bindings. One excellent example is vc, it is just so easy to 
>>> remember.
>>
>> What do you mean with "consistent definition of key bindings".
>>
>>> I have been an regular user of org, and I have already forgotten that
>>> S-up/down does priority change after one month's break.
>>
>> I have been trying very hard to make the key bindings as logical
>> as possible - but there is a huge amount of commands in org,
>> and many keys are needed.
>>
>> That said, the S-up/down for priorities does leas to a couple
>> of inconsistencies, in particular between agenda and org-mode buffers.
>>
>> - Carsten
>
> I know it is difficult to come up with good key bindings for so many
> commands that org-mode has. That's why I gave the example of vc, which
> has extremely excellent key bindings. I would never have to learn them 
> a
> second time. I hope someone can come up with something similar for org.

Well, this is not really a fair comparison, because vc has maybe 13
commands???  Sure, not too hard to make these stick in your head.

Obviously I have been trying to make the best key bindings I can.
Besides making them easy to remember you are of course limited
by Emacs conventions, and by the wish to avoid 3 keypress bindings for
speed.  And maybe some compromises because of the historical
development, to keep the habits of users.

I am certainly open for suggestions.

- Carsten


>
> -- 
> Leo <sdl.web AT gmail.com>                         (GPG Key: 9283AA3F)
>
>          Gnus is one component of the Emacs operating system.
>
>
>
> _______________________________________________
> Emacs-orgmode mailing list
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode
>
>

--
Carsten Dominik
Sterrenkundig Instituut "Anton Pannekoek"
Universiteit van Amsterdam
Kruislaan 403
NL-1098SJ Amsterdam
phone: +31 20 525 7477

Re: counter-intuitive key bindings

[Adam Spiers]

On Sat, Jul 28, 2007 at 12:17:06PM +0100, Leo wrote:
> (info "(org)Multiple sets in one file")
> | `C-S-<right>'
> | `C-S-<left>'
> |      These keys jump from one TODO subset to the next.  In the above
> |      example, `C-S-<right>' would jump from `TODO' or `DONE' to
> |      `REPORT', and any of the words in the second row to `CANCELED'.  
> 
> These key bindings are to move from one set of the TODO items to
> another, which intuitively are a vertical motion. See this configure:
> 
> |      (setq org-todo-keywords
> |            '((sequence "TODO" "|" "DONE")
> |              (sequence "REPORT" "BUG" "KNOWNCAUSE" "|" "FIXED")
> |              (sequence "|" "CANCELED")))
> 
> I would propose to change these key bindings to:
> 
>    C-S-<up/down>

Agreed, this is more logical, although the modified cursor key
bindings are always going to be contended.  Personally I have a
"global" standard set of key bindings for outline navigation and
editing which applies across several modes: folding-mode, org-mode,
outline-minor-mode (and allout), orgstruct-mode.  Having to remember
different bindings for how to hide/reveal/zoom to a subtree in each
mode is a nightmare.  I would love to see some standardization here
with more mode authors talking to each other.

On a related topic, if I have S-{left,right} on my org-disputed-keys
list, this breaks using them for changing a timestamp by one-day
increments, even though I don't consider that a clashing binding.  I
guess I am arguing that this is a minor bug with `org-key' being too
indiscriminate in when it filters bindings.

Re: counter-intuitive key bindings

[Carsten Dominik]

On Jul 28, 2007, at 14:52, Adam Spiers wrote:

> On Sat, Jul 28, 2007 at 12:17:06PM +0100, Leo wrote:
>> (info "(org)Multiple sets in one file")
>> | `C-S-<right>'
>> | `C-S-<left>'
>> |      These keys jump from one TODO subset to the next.  In the above
>> |      example, `C-S-<right>' would jump from `TODO' or `DONE' to
>> |      `REPORT', and any of the words in the second row to `CANCELED'.
>>
>> These key bindings are to move from one set of the TODO items to
>> another, which intuitively are a vertical motion. See this configure:
>>
>> |      (setq org-todo-keywords
>> |            '((sequence "TODO" "|" "DONE")
>> |              (sequence "REPORT" "BUG" "KNOWNCAUSE" "|" "FIXED")
>> |              (sequence "|" "CANCELED")))
>>
>> I would propose to change these key bindings to:
>>
>>    C-S-<up/down>
>
> Agreed, this is more logical, although the modified cursor key
> bindings are always going to be contended.  Personally I have a
> "global" standard set of key bindings for outline navigation and
> editing which applies across several modes: folding-mode, org-mode,
> outline-minor-mode (and allout), orgstruct-mode.  Having to remember
> different bindings for how to hide/reveal/zoom to a subtree in each
> mode is a nightmare.  I would love to see some standardization here
> with more mode authors talking to each other.
>
> On a related topic, if I have S-{left,right} on my org-disputed-keys
> list, this breaks using them for changing a timestamp by one-day
> increments, even though I don't consider that a clashing binding.  I
> guess I am arguing that this is a minor bug with `org-key' being too
> indiscriminate in when it filters bindings.

Why do you then have these keys on your list?  What are you
using them for?  How is org-mode supposed to know in what
situations you don't what org-mode to use them?

- Carsten

Re: counter-intuitive key bindings

[Adam Spiers]

Carsten Dominik (dominik@science.uva.nl) wrote:
> On Jul 28, 2007, at 14:52, Adam Spiers wrote:
> >On a related topic, if I have S-{left,right} on my org-disputed-keys
> >list, this breaks using them for changing a timestamp by one-day
> >increments, even though I don't consider that a clashing binding.  I
> >guess I am arguing that this is a minor bug with `org-key' being too
> >indiscriminate in when it filters bindings.
> 
> Why do you then have these keys on your list?  What are you
> using them for?  How is org-mode supposed to know in what
> situations you don't what org-mode to use them?

You're quite right of course :-) It's up to me to override them in the
specific maps in which I want them overridden.