[RFC] Proposal for rebindings in Org 8.3

[RFC] Proposal for rebindings in Org 8.3

[Bastien]

The purpose is to be polite with minor modes and to not use
C-c [:punct:] keybindings, as recommended in the Elisp manual.

This is related to the issues reported here:
http://lists.gnu.org/archive/html/emacs-orgmode/2014-01/msg00866.html
http://article.gmane.org/gmane.emacs.orgmode/82010

Here is a table to summarise the proposal:

| Key   | Command                           | Proposal         | Status |
|-------+-----------------------------------+------------------+--------|
| C-c # | Checkboxes                        | C-c x            | Free   |
| C-c ~ | Cooperation                       | C-c C-~          | Free   |
| C-c , | Priorities                        | C-c C-,          | Free   |
| C-c ? | Editing and debugging formulas    | C-c C-?          | Free   |
| C-c ! | Creating timestamps               | C-c C-!          | Free   |
| C-c . | Creating timestamps               | C-c C-.          | Free   |
| C-c ` | Built-in table editor             | C-c C-`          | Free   |
|-------+-----------------------------------+------------------+--------|
| C-c ' | Editing and debugging formulas    | C-c " or C-c C-' | Free   |
| C-c ^ | Structure editing, plain lists... | C-c C-^          | Taken  |
| C-c @ | Structure editing                 | C-c <            | Taken  |

The order is from the less problematic ones to the more problematic ones.

A few comments on the last three:

- " is not a punctuation character, I find C-c " instead of C-c ' good.

- C-c C-^ is taken by org-up-element.  I'm in favor of using C-c C-u
  (currently bound to `outline-up-heading') for `org-up-element'.
  Nicolas suggested C-M-u but I find it convenient to have it when
  editing Elisp code within Org buffers.

- C-c < is bound to `org-date-from-calendar', which inserts the
  current date (or the date from the calendar buffer) at point as
  an active timestamp.

  C-c > is bound to `org-goto-calendar' and goes to the calendar,
  going to the date at point if any.  This is just a slightly more
  contextual M-x calendar RET.

  So the whole suggestion here is:
  
  - in Org-mode, remap calendar to org-goto-calendar and unbind it
    from C-c >

  - use C-c > for `org-date-from-calendar'

  - use C-c < for selecting the whole subtree, which is consistent
    with the use of `<' as a speedy command for doing the same.

Let me know how you feel about such move in general and each rebinding
in particular.  We are not forced to solve them all at once.

Thanks,

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Sebastien Vauban]

Bastien wrote:
> The purpose is to be polite with minor modes and to not use
> C-c [:punct:] keybindings, as recommended in the Elisp manual.
>
> This is related to the issues reported here:
> http://lists.gnu.org/archive/html/emacs-orgmode/2014-01/msg00866.html
> http://article.gmane.org/gmane.emacs.orgmode/82010
>
> Here is a table to summarise the proposal:
>
> | Key   | Command                           | Proposal         | Status |
> |-------+-----------------------------------+------------------+--------|
> | C-c # | Checkboxes                        | C-c x            | Free   |
> | C-c ~ | Cooperation                       | C-c C-~          | Free   |
> | C-c , | Priorities                        | C-c C-,          | Free   |
> | C-c ? | Editing and debugging formulas    | C-c C-?          | Free   |
> | C-c ! | Creating timestamps               | C-c C-!          | Free   |
> | C-c . | Creating timestamps               | C-c C-.          | Free   |
> | C-c ` | Built-in table editor             | C-c C-`          | Free   |
> |-------+-----------------------------------+------------------+--------|
> | C-c ' | Editing and debugging formulas    | C-c " or C-c C-' | Free   |
> | C-c ^ | Structure editing, plain lists... | C-c C-^          | Taken  |
> | C-c @ | Structure editing                 | C-c <            | Taken  |
>
> The order is from the less problematic ones to the more problematic ones.
>
> A few comments on the last three:
>
> - " is not a punctuation character, I find C-c " instead of C-c ' good.
>
> - C-c C-^ is taken by org-up-element.  I'm in favor of using C-c C-u
>   (currently bound to `outline-up-heading') for `org-up-element'.
>   Nicolas suggested C-M-u but I find it convenient to have it when
>   editing Elisp code within Org buffers.
>
> - C-c < is bound to `org-date-from-calendar', which inserts the
>   current date (or the date from the calendar buffer) at point as
>   an active timestamp.
>
>   C-c > is bound to `org-goto-calendar' and goes to the calendar,
>   going to the date at point if any.  This is just a slightly more
>   contextual M-x calendar RET.
>
>   So the whole suggestion here is:
>   
>   - in Org-mode, remap calendar to org-goto-calendar and unbind it
>     from C-c >
>
>   - use C-c > for `org-date-from-calendar'
>
>   - use C-c < for selecting the whole subtree, which is consistent
>     with the use of `<' as a speedy command for doing the same.

You put the finger on one important note to me: that speed commands are
"the same" as their longer counterpart.  It should be good if all of
them could be obtainable by adding "C-c C-" in front of the speed
command.

With that focus in mind, I find `C-c x' not a good one (for checkboxes);
though, in that case, speed commands don't come into play. But I find it
too different from the `C-c C-' construct used for the others?

> Let me know how you feel about such move in general and each rebinding
> in particular.  We are not forced to solve them all at once.

Globally, it's fine!

Last remark about `C-c <' which I find not good:

- `<' in used for filtering categories in the agenda,

- semantically, what `C-c @' does is a `C-x h' limited to a subtree;
  maybe there could be something inspiring similar to `C-x h'?  Maybe
  `C-u C-x h'?

Best regards,
  Seb

-- 
Sebastien Vauban

Re: [RFC] Proposal for rebindings in Org 8.3

[Bastien]

Thanks for your feedback.

"Sebastien Vauban" <sva-news-D0wtAvR13HarG/iDocfnWg@public.gmane.org>
writes:

>>   - use C-c < for selecting the whole subtree, which is consistent
>>     with the use of `<' as a speedy command for doing the same.
>
> You put the finger on one important note to me: that speed commands are
> "the same" as their longer counterpart.  It should be good if all of
> them could be obtainable by adding "C-c C-" in front of the speed
> command.

I think it goes in the opposite direction: if a command is bound to
C-c KEY or C-c C-KEY then KEY should be used as a speed command.

> With that focus in mind, I find `C-c x' not a good one (for checkboxes);
> though, in that case, speed commands don't come into play.

The mnemonic here is: `x' reminds me of [x] -- let's see if something
better can be done.

> But I find it
> too different from the `C-c C-' construct used for the others?

I wanted to keep it close to C-c #, and use only C-c KEY.

>> Let me know how you feel about such move in general and each rebinding
>> in particular.  We are not forced to solve them all at once.
>
> Globally, it's fine!
>
> Last remark about `C-c <' which I find not good:
>
> - `<' in used for filtering categories in the agenda,

Categories are (most often) attached to a subtree.

When you hit the `<' speed command on this subtree then display
the agenda, you don't need to filter by category with `<' since
the narrowing already filters (by category) for you.

That's actually one of the reason for which I propose this key.

> - semantically, what `C-c @' does is a `C-x h' limited to a subtree;
>   maybe there could be something inspiring similar to `C-x h'?  Maybe
>   `C-u C-x h'?

The way to get the same behavior than `C-c @' is `C-c C-^ M-h'.
M-h is very very handy, and probably not known enough!

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Michael Brand]

Hi Bastien

On Fri, Feb 7, 2014 at 11:32 AM, Bastien <bzg@gnu.org> wrote:
> | C-c # | Checkboxes                        | C-c x            | Free   |

I did not know "C-c #" (org-update-statistics-cookies) because for me
it is easier to remember and therefore I always used this: Go to
cookie and do "C-c C-c". And "C-c x" does not comply with the Elisp
manual:

    Don't define C-c letter as a key in Lisp programs. Sequences
    consisting of C-c and a letter (either upper or lower case) are
    reserved for users; they are the only sequences reserved for
    users, so do not block them.

    Changing all the Emacs major modes to respect this convention was
    a lot of work; abandoning this convention would make that work go
    to waste, and inconvenience users. Please comply with it.

With these two reasons my suggestion is to just drop the binding for
(org-update-statistics-cookies).

Michael

Re: [RFC] Proposal for rebindings in Org 8.3

[Bastien]

Hi Michael,

Michael Brand <michael.ch.brand@gmail.com> writes:

> With these two reasons my suggestion is to just drop the binding for
> (org-update-statistics-cookies).

Agreed.  If noone objects, I'll remove C-c #.

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Nick Dokos]

Bastien <bzg@gnu.org> writes:

> The purpose is to be polite with minor modes and to not use
> C-c [:punct:] keybindings, as recommended in the Elisp manual.
>
> This is related to the issues reported here:
> http://lists.gnu.org/archive/html/emacs-orgmode/2014-01/msg00866.html
> http://article.gmane.org/gmane.emacs.orgmode/82010
>
> Here is a table to summarise the proposal:
>
> | Key   | Command                           | Proposal         | Status |
> |-------+-----------------------------------+------------------+--------|
> | C-c # | Checkboxes                        | C-c x            | Free   |
> | C-c ~ | Cooperation                       | C-c C-~          | Free   |
> | C-c , | Priorities                        | C-c C-,          | Free   |
> | C-c ? | Editing and debugging formulas    | C-c C-?          | Free   |
> | C-c ! | Creating timestamps               | C-c C-!          | Free   |
> | C-c . | Creating timestamps               | C-c C-.          | Free   |
> | C-c ` | Built-in table editor             | C-c C-`          | Free   |
> |-------+-----------------------------------+------------------+--------|
> | C-c ' | Editing and debugging formulas    | C-c " or C-c C-' | Free   |
> | C-c ^ | Structure editing, plain lists... | C-c C-^          | Taken  |
> | C-c @ | Structure editing                 | C-c <            | Taken  |
>
> The order is from the less problematic ones to the more problematic ones.
>
> A few comments on the last three:
>
> - " is not a punctuation character, I find C-c " instead of C-c ' good.
>

On a QWERTY keyboard or Dvorak keyboard, " requires Shifting, so you
have to switch horses midstream (first Control, then SHift). Not sure
about anybody else, but those are the most problematic key sequences for
me. I would much prefer C-c C-': you just keep the Control key pressed
for the duration.

> - C-c C-^ is taken by org-up-element.  I'm in favor of using C-c C-u
>   (currently bound to `outline-up-heading') for `org-up-element'.
>   Nicolas suggested C-M-u but I find it convenient to have it when
>   editing Elisp code within Org buffers.
>
> - C-c < is bound to `org-date-from-calendar', which inserts the
>   current date (or the date from the calendar buffer) at point as
>   an active timestamp.
>
>   C-c > is bound to `org-goto-calendar' and goes to the calendar,
>   going to the date at point if any.  This is just a slightly more
>   contextual M-x calendar RET.
>
>   So the whole suggestion here is:
>   
>   - in Org-mode, remap calendar to org-goto-calendar and unbind it
>     from C-c >
>
>   - use C-c > for `org-date-from-calendar'
>
>   - use C-c < for selecting the whole subtree, which is consistent
>     with the use of `<' as a speedy command for doing the same.
>
> Let me know how you feel about such move in general and each rebinding
> in particular.  We are not forced to solve them all at once.
>
> Thanks,

--
Nick

Re: [RFC] Proposal for rebindings in Org 8.3

[Rasmus]

Nick Dokos <ndokos@gmail.com> writes:

> Bastien <bzg@gnu.org> writes:
>
>> The purpose is to be polite with minor modes and to not use
>> C-c [:punct:] keybindings, as recommended in the Elisp manual.
>>
>> This is related to the issues reported here:
>> http://lists.gnu.org/archive/html/emacs-orgmode/2014-01/msg00866.html
>> http://article.gmane.org/gmane.emacs.orgmode/82010
>>
>> Here is a table to summarise the proposal:
>>
>> | Key   | Command                           | Proposal         | Status |
>> |-------+-----------------------------------+------------------+--------|
>> | C-c # | Checkboxes                        | C-c x            | Free   |
>> | C-c ~ | Cooperation                       | C-c C-~          | Free   |
>> | C-c , | Priorities                        | C-c C-,          | Free   |
>> | C-c ? | Editing and debugging formulas    | C-c C-?          | Free   |
>> | C-c ! | Creating timestamps               | C-c C-!          | Free   |
>> | C-c . | Creating timestamps               | C-c C-.          | Free   |
>> | C-c ` | Built-in table editor             | C-c C-`          | Free   |
>> |-------+-----------------------------------+------------------+--------|
>> | C-c ' | Editing and debugging formulas    | C-c " or C-c C-' | Free   |
>> | C-c ^ | Structure editing, plain lists... | C-c C-^          | Taken  |
>> | C-c @ | Structure editing                 | C-c <            | Taken  |
>>
>> The order is from the less problematic ones to the more problematic ones.
>>
>> A few comments on the last three:
>>
>> - " is not a punctuation character, I find C-c " instead of C-c ' good.
>>
>
> On a QWERTY keyboard or Dvorak keyboard, " requires Shifting, so you
> have to switch horses midstream (first Control, then SHift). Not sure
> about anybody else, but those are the most problematic key sequences for
> me. I would much prefer C-c C-': you just keep the Control key pressed
> for the duration.

I agree that it's better if we can avoid letters needing a 'secondary'
key.

Unfortunately, I think this wish is hard to meet when considering
several layouts.

For me the following keys need shift or Alt-Gr: ^, ", `, !, ?, and ~.

Rasmus

-- 
If you can mix business and politics wonderful things can happen!

Re: [RFC] Proposal for rebindings in Org 8.3

[Bastien]

Rasmus <rasmus@gmx.us> writes:

> For me the following keys need shift or Alt-Gr: ^, ", `, !, ?, and
> ~.

Oh, do you really need the Alt-Gr key for `?' and `!' ?

Does that mean you don't use `C-c !' or that you rarely use it?

Of course, this may have nothing to do with the keybinding itself,
but I'm curious.

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Samuel Wales]

i'd like to point out that some of these bindings are problematic on
qwerty in a way that is not obvious to many users.

many people use shifting keys on both sides of the keyboard.

they do not issue the control of control g using left hand.  that is a
bad idea.  instead, they hold down the right control with right hand
then do g with left hand.

any sequence where one step is on one side and another step is on the
other side is fine unless they are both shifted.  at that point they
become so cumbersome they are likely to be rebound by the user.

fortunately, most of emacs does not use the cumbersome type of binding.

hih.

samuel

-- 
The Kafka Pandemic: http://thekafkapandemic.blogspot.com

The disease DOES progress.  MANY people have died from it.  ANYBODY can get it.

Denmark: free Karina Hansen NOW.

Re: [RFC] Proposal for rebindings in Org 8.3

[Bastien]

Hi Samuel,

Samuel Wales <samologist@gmail.com> writes:

> any sequence where one step is on one side and another step is on the
> other side is fine unless they are both shifted.

I understand.  I don't think any of the proposed rebindings requires
this -- except C-c " -- but C-c C-' is better anyway.

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Samuel Wales]

On 2/8/14, Bastien <bzg@gnu.org> wrote:
> I understand.  I don't think any of the proposed rebindings requires
> this -- except C-c " -- but C-c C-' is better anyway.

hi bastien,

c-c c-' is just as bad.  control is a modifier key.  try it!

samuel

-- 
The Kafka Pandemic: http://thekafkapandemic.blogspot.com

The disease DOES progress.  MANY people have died from it.  ANYBODY can get it.

Denmark: free Karina Hansen NOW.

Re: [RFC] Proposal for rebindings in Org 8.3

[Samuel Wales]

well, almost as bad :).  you have to press shift also, but that's a
minor difference :).

Re: [RFC] Proposal for rebindings in Org 8.3

[Bastien]

Samuel Wales <samologist@gmail.com> writes:

> c-c c-' is just as bad.  control is a modifier key.  try it!

I tried it and for me it is better than C-c '

Anyway, let's try to find the best replacement.

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Samuel Wales]

hi bastien,

i meant that c-c c-' and c-c c-" are both cumbersome for those users
who press c-c by holding down the control key with the right hand and
then pressing c with the left hand.  c-c ' is not cumbersome for those
users.

samuel

-- 
The Kafka Pandemic: http://thekafkapandemic.blogspot.com

The disease DOES progress.  MANY people have died from it.  ANYBODY can get it.

Denmark: free Karina Hansen NOW.

Re: [RFC] Proposal for rebindings in Org 8.3

[Bastien]

Hi Samuel,

Samuel Wales <samologist@gmail.com> writes:

> i meant that c-c c-' and c-c c-" are both cumbersome for those users
> who press c-c by holding down the control key with the right hand and
> then pressing c with the left hand.  c-c ' is not cumbersome for those
> users.

I understand better now, thanks.

What I don't understand is why keeping the right control key between
C-c and C-' is harder than releasing the control key between C-c and
' (or " as also proposed.)  My experience (which seems the same than
Nick's) is that holding the control key down is easier/faster.

(I wonder if any serious ergonomical study has been done for Emacs
beybindings depending on keyboards.)

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Achim Gratz]

Bastien writes:
> What I don't understand is why keeping the right control key between
> C-c and C-' is harder than releasing the control key between C-c and
> ' (or " as also proposed.)  My experience (which seems the same than
> Nick's) is that holding the control key down is easier/faster.

That depends on the keyboard layout, but if pressing "'" requires the
use of the right hand, then you'd need to switch to the laft for
pressing Control (unless you were chording it with the right hand, which
is probably a bad thing for anyone inflicted with RSI).

> (I wonder if any serious ergonomical study has been done for Emacs
> beybindings depending on keyboards.)

Quite certainly not.  There are oodles of keybindings that really don't
work sensibly with anything but a UNIX keyboard and the whole concept is
clearly inspired by the Lisp Machine "Space Cadet" keyboard.  I wish I'd
still have my Sun Type4 keyboard… Similarly, you really don't get why
anybody thought "\" was a good path separator until you take a look at
the original PC/XT keyboard (it was also a non-modifier key on most
terminal keyboards at the time, but "/" was more prominent there).


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Factory and User Sound Singles for Waldorf rackAttack:
http://Synth.Stromeko.net/Downloads.html#WaldorfSounds

Assumptions on user's environment (was: [RFC] Proposal for rebindings in Org 8.3)

[Karl Voit]

* Samuel Wales <samologist@gmail.com> wrote:

Hi!

> i meant that c-c c-' and c-c c-" are both cumbersome for those users
> who press c-c by holding down the control key with the right hand and
> then pressing c with the left hand.  c-c ' is not cumbersome for those
> users.

Oh, this is a sensitive subject :-)

For example: I am happily removing CAPS LOCK on all of my computers
and replace it with an additional CTRL key. This way, I am using my
left pinkie for all kinds of CTRL-combinations. As long as it is not
combined with [`123q~] or TAB, I am fine.

And: although I am living in a German speaking country, I am using
US_intl keyboard settings. Most keyboard shortcuts make more sense
since I switched. Unfortunately, software developers who define
keyboard shortcuts have either settle for their own keyboard layout
(mostly en_US) or keyboard shortcuts are part of the i18n layer
which has also some drawbacks IMHO.

The point is: when you are settling for keyboard shortcuts, you are
going to do some assumptions on the environment of your users. So:
what are these assumptions for Emacs/Org-mode? I guess this is the
root question we should try to answer (and document).

-- 
mail|git|SVN|photos|postings|SMS|phonecalls|RSS|CSV|XML to Org-mode:
       > get Memacs from https://github.com/novoid/Memacs <

https://github.com/novoid/extract_pdf_annotations_to_orgmode + more on github

Re: Assumptions on user's environment

[Bastien]

Hi Karl,

Karl Voit <devnull@Karl-Voit.at> writes:

> So: what are these assumptions for Emacs/Org-mode? I guess this is
> the root question we should try to answer (and document).

Well, this is a general Emacs issue.  I guess the assumption is that
keybindings are optimized for US keyboards layouts - that said, some
core keybindings are based on mnemonic (C-f and C-b to move forward
and backward), so "optimized" seems a bit fuzzy anyway...

-- 
 Bastien

Re: Assumptions on user's environment

[Karl Voit]

* Bastien <bzg@gnu.org> wrote:
> Hi Karl,

Hi Bastien!

> Karl Voit <devnull@Karl-Voit.at> writes:
>
>> So: what are these assumptions for Emacs/Org-mode? I guess this is
>> the root question we should try to answer (and document).
>
> Well, this is a general Emacs issue.  

Sure.

> I guess the assumption is that keybindings are optimized for US
> keyboards layouts - that said, some core keybindings are based on
> mnemonic (C-f and C-b to move forward and backward), so
> "optimized" seems a bit fuzzy anyway...

Fair enough. So there has to be a decision whether or not to invest
time/effort for users of QWERTZ keyboard layouts.

If we should decide to ignore non US_intl/en_US-issues, I vote for
making this decision very clear to new users with a prominent
sentence in our documentation.

Personally, I do think that tech-savvy users of non English speaking
countries should definitely consider switching to US_intl layout for
many reasons.  From my experience, only a minority of text-savvy
users are doing so. But this is also true for applying live-hacking
as a habit in general :-(

-- 
mail|git|SVN|photos|postings|SMS|phonecalls|RSS|CSV|XML to Org-mode:
       > get Memacs from https://github.com/novoid/Memacs <

https://github.com/novoid/extract_pdf_annotations_to_orgmode + more on github

Re: Assumptions on user's environment

[Rasmus]

Karl Voit <devnull@Karl-Voit.at> writes:

> Personally, I do think that tech-savvy users of non English speaking
> countries should definitely consider switching to US_intl layout for
> many reasons.  

What if the set of letters in English is a subset of the set of
letters in your the tech-savvy users language?

> From my experience, only a minority of text-savvy users are doing
> so.

Perhaps the objective that these users are maximizing another
objective than the one you have in mind?

—Rasmus

-- 
If you can mix business and politics wonderful things can happen!

Re: Assumptions on user's environment

[Karl Voit]

* Rasmus <rasmus@gmx.us> wrote:
> Karl Voit <devnull@Karl-Voit.at> writes:
>
>> Personally, I do think that tech-savvy users of non English speaking
>> countries should definitely consider switching to US_intl layout for
>> many reasons.  
>
> What if the set of letters in English is a subset of the set of
> letters in your the tech-savvy users language?

This is the case with the German language. However, there are ways
to enter German umlauts with us_intl layout (digraph).

With Microsoft Windows as an exception, it is possible to switch
keyboard layouts on keyboard shortcuts.

However, when I am coding, I am only using English variable names
and comments. So most of the time, I am happy with 7-bit ASCII
characters.

If you do not code nor use strange environments like LaTeX, you
might as well ignore my comments about tech-savvy people which was
maybe a bit too general.

>> From my experience, only a minority of text-savvy users are doing
>> so.
>
> Perhaps the objective that these users are maximizing another
> objective than the one you have in mind?

Yes, you are absolutely right. So "tech-savvy" is probably too
unspecific as a term. However, "coders" seems to be narrow as well.
Maybe you can think of a term in between? ;-)

-- 
mail|git|SVN|photos|postings|SMS|phonecalls|RSS|CSV|XML to Org-mode:
       > get Memacs from https://github.com/novoid/Memacs <

https://github.com/novoid/extract_pdf_annotations_to_orgmode + more on github

Re: Assumptions on user's environment (was: [RFC] Proposal for rebindings in Org 8.3)

[Samuel Wales]

hi karl,

it is true that there are assumptions about the user's environment.
the strongest is probably qwerty, followed by european language,
english, and common layouts.  as a native english speaker, i am aware
that i am fortunate in that regard.  i get to use utf-8 without making
my text files larger, unlike some asian speakers, for example.

HOWEVER: my point was not about the user's environment at all.  it
applies to whatever environment is chosen for the assumptions.

on qwerty, c is on the lhs.  ' is on the rhs.  /that/ is why c-c c-'
is more cumbersome for using two hands for modifiers.  remember: using
one hand for both modifier and key is never an option.

===

hope that clarifies.

samuel


On 2/9/14, Karl Voit <devnull@karl-voit.at> wrote:
> * Samuel Wales <samologist@gmail.com> wrote:
>
> Hi!
>
>> i meant that c-c c-' and c-c c-" are both cumbersome for those users
>> who press c-c by holding down the control key with the right hand and
>> then pressing c with the left hand.  c-c ' is not cumbersome for those
>> users.
>
> Oh, this is a sensitive subject :-)
>
> For example: I am happily removing CAPS LOCK on all of my computers
> and replace it with an additional CTRL key. This way, I am using my
> left pinkie for all kinds of CTRL-combinations. As long as it is not
> combined with [`123q~] or TAB, I am fine.
>
> And: although I am living in a German speaking country, I am using
> US_intl keyboard settings. Most keyboard shortcuts make more sense
> since I switched. Unfortunately, software developers who define
> keyboard shortcuts have either settle for their own keyboard layout
> (mostly en_US) or keyboard shortcuts are part of the i18n layer
> which has also some drawbacks IMHO.
>
> The point is: when you are settling for keyboard shortcuts, you are
> going to do some assumptions on the environment of your users. So:
> what are these assumptions for Emacs/Org-mode? I guess this is the
> root question we should try to answer (and document).
>
> --
> mail|git|SVN|photos|postings|SMS|phonecalls|RSS|CSV|XML to Org-mode:
>        > get Memacs from https://github.com/novoid/Memacs <
>
> https://github.com/novoid/extract_pdf_annotations_to_orgmode + more on
> github
>
>
>


-- 
The Kafka Pandemic: http://thekafkapandemic.blogspot.com

The disease DOES progress.  MANY people have died from it.  ANYBODY can get it.

Denmark: free Karina Hansen NOW.

Re: Assumptions on user's environment

[Bastien]

Samuel Wales <samologist@gmail.com> writes:

> remember: using
> one hand for both modifier and key is never an option.

Why?  For me C-c C-' is very easy with one hand, I don't even
need to move the fingers.

-- 
 Bastien

Re: Assumptions on user's environment

[Samuel Wales]

rsi.

one of the worst things you can do is use a single hand for more than one key.


On 2/9/14, Bastien <bzg@gnu.org> wrote:
> Samuel Wales <samologist@gmail.com> writes:
>
>> remember: using
>> one hand for both modifier and key is never an option.
>
> Why?  For me C-c C-' is very easy with one hand, I don't even
> need to move the fingers.
>
> --
>  Bastien
>


-- 
The Kafka Pandemic: http://thekafkapandemic.blogspot.com

The disease DOES progress.  MANY people have died from it.  ANYBODY can get it.

Denmark: free Karina Hansen NOW.

Re: Assumptions on user's environment

[Achim Gratz]

Bastien writes:
> Samuel Wales <samologist@gmail.com> writes:
>
>> remember: using
>> one hand for both modifier and key is never an option.
>
> Why?  For me C-c C-' is very easy with one hand, I don't even
> need to move the fingers.

I'd think you need to curl the pinky at least unless you're talking
about a laptop keyboard with a favorable position for the Ctrl key.

It's really a bad thing if you've developed certain types of RSI and may
even be impossible if you have coordinative disabilities or can only use
one hand.  This is what "sticky" or "locked" modifiers were invented for
(check how your favourite OS lets you set up "accessibility features").


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptation for Waldorf rackAttack V1.04R1:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada

Re: Assumptions on user's environment

[Karl Voit]

* Achim Gratz <Stromeko@nexgo.de> wrote:
> Bastien writes:
>> Samuel Wales <samologist@gmail.com> writes:
>>
>>> remember: using
>>> one hand for both modifier and key is never an option.
>>
>> Why?  For me C-c C-' is very easy with one hand, I don't even
>> need to move the fingers.
>
> I'd think you need to curl the pinky at least unless you're talking
> about a laptop keyboard with a favorable position for the Ctrl key.

I am using Lenovo UltraNav USB keyboards on Windows and Linux
machines and an Apple BT keyboard on a Mac. On all of them I can
type C-c with a *completely relaxed* left pinkie and a left index
finger thanks to an additional CTRL key instead of the CAPS LOCK. My
left pinkie seldom moves away from the left CAPS LOCK position.

In fact, I never use the original left CTRL key at all. And more or
less thanks to bad habit, I never use the right CTRL key.

Unfortunately, I do not touch type all the time. And I know, that my
typing behavior is not optimized. However, it's that fast that
changing my typing habits at this point seems to be a hard thing to
do for very little benefit. I never had any physical issues related
to typing. I tend to think that this is because of the CAPSLOCK/CTRL
trick.


I get the impression from this thread, that this varies from user to
user.

-- 
mail|git|SVN|photos|postings|SMS|phonecalls|RSS|CSV|XML to Org-mode:
       > get Memacs from https://github.com/novoid/Memacs <

https://github.com/novoid/extract_pdf_annotations_to_orgmode + more on github

Re: Assumptions on user's environment

[Bastien]

Karl Voit <devnull@Karl-Voit.at> writes:

> In fact, I never use the original left CTRL key at all. And more or
> less thanks to bad habit, I never use the right CTRL key.

Yep, I'm exactly in the same boat.

-- 
 Bastien

Re: Assumptions on user's environment

[Bastien]

Achim Gratz <Stromeko@nexgo.de> writes:

> I'd think you need to curl the pinky at least unless you're talking
> about a laptop keyboard with a favorable position for the Ctrl key.
>
> It's really a bad thing if you've developed certain types of RSI and may
> even be impossible if you have coordinative disabilities or can only use
> one hand.  This is what "sticky" or "locked" modifiers were invented for
> (check how your favourite OS lets you set up "accessibility features").

Yes, I see, thanks.

What do you think of `C-c :', as suggested in my reply to Thomas?

-- 
 Bastien

Re: Assumptions on user's environment

[Sebastien Vauban]

Bastien wrote:
> What do you think of `C-c :', as suggested in my reply to Thomas?

I'd favor a common `C-c C-' prefix for all moved key bindings.

Best regards,
  Seb

-- 
Sebastien Vauban

Re: Assumptions on user's environment

[Samuel Wales]

perhaps this is possible:

  c-c ' -> c-c c-x '

Re: Assumptions on user's environment

[Karl Voit]

* Samuel Wales <samologist@gmail.com> wrote:
> perhaps this is possible:
>
>   c-c ' -> c-c c-x '

If you are using US_intl with digraph (deadkeys) in order to type
special characters, ' needs actually two keys: ' + SPACE

So I guess this is not that great for common commands typed often a
day.

-- 
mail|git|SVN|photos|postings|SMS|phonecalls|RSS|CSV|XML to Org-mode:
       > get Memacs from https://github.com/novoid/Memacs <

https://github.com/novoid/extract_pdf_annotations_to_orgmode + more on github

Re: [RFC] Proposal for rebindings in Org 8.3

[Bastien]

Hi Nick,

Nick Dokos <ndokos@gmail.com> writes:

> On a QWERTY keyboard or Dvorak keyboard, " requires Shifting, so you
> have to switch horses midstream (first Control, then SHift). Not sure
> about anybody else, but those are the most problematic key sequences for
> me. I would much prefer C-c C-': you just keep the Control key pressed
> for the duration.

That's actually the same on an AZERTY keyboard, I prefer to keep the
control key pressed.

Also, the change to C-c C-' is easier to remember.

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Thomas S. Dye]

Bastien <bzg@gnu.org> writes:

> A few comments on the last three:
>
> - " is not a punctuation character, I find C-c " instead of C-c ' good.

Is it true that " is not a punctuation character? 

In any event, C-c " is likely to pose the same problem as C-c '.  For
icicle users, C-c " will be shadowed by the keybinding for
icicle-search-text-property.

At least that's my current working hypothesis.  Feel free to correct me
if I'm wrong.

All the best,
Tom
-- 
Thomas S. Dye
http://www.tsdye.com

Re: [RFC] Proposal for rebindings in Org 8.3

[Bastien]

Hi Thomas,

tsd@tsdye.com (Thomas S. Dye) writes:

> Bastien <bzg@gnu.org> writes:
>
>> A few comments on the last three:
>>
>> - " is not a punctuation character, I find C-c " instead of C-c ' good.
>
> Is it true that " is not a punctuation character?

It is a string in both fundamental-mode and org-mode, that's what I'm
checking.  But it is a punctuation character in other modes and maybe
in our "minds".

I agree the intuition is misleading here, because ' is a punctuation
character.

> In any event, C-c " is likely to pose the same problem as C-c '.  For
> icicle users, C-c " will be shadowed by the keybinding for
> icicle-search-text-property.

... which, strictly speaking, is not an org-mode problem if " is not
a punctuation character.

> At least that's my current working hypothesis.  Feel free to correct me
> if I'm wrong.

It encourages me to find a completely different solution.

The one I have in mind now is to use C-c : (and to move the current
C-c : binding to C-c C-:).  The reasoning is that we have

- fixed-width
- example blocks
- source code blocks

which often achieve a similar goal: to present some code snippets.
Source blocks are more powerful than example blocks which are more
powerful than fixed-width regions.  But you can currently edit them
all with C-c '.

So C-c : would call org-edit-src-code and C-c C-: would convert the
region to fixed-width region.

What you (and others) think?

Eric, what's your take on this, as the "father" of C-c ' ?

-- 
 Bastien

Re: [RFC] Proposal for rebindings in Org 8.3

[Achim Gratz]

Bastien writes:
> So C-c : would call org-edit-src-code and C-c C-: would convert the
> region to fixed-width region.

You cannot enter C-: in some terminals because it would require
simultaneous processing of shift and control (these terminals ignore
shift while control is pressed).


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptation for Waldorf microQ V2.22R2:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada

terminal emulators (was: Re: Proposal for rebindings in Org 8.3)

[Gregor Zattler]

Hi Achim,
* Achim Gratz <Stromeko@nexgo.de> [10. Feb. 2014]:
> Bastien writes:
>> So C-c : would call org-edit-src-code and C-c C-: would convert the
>> region to fixed-width region.
> 
> You cannot enter C-: in some terminals because it would require
> simultaneous processing of shift and control (these terminals ignore
> shift while control is pressed).

this is true for xterm, rxvt-unicode, gnome-terminal, konsole and
the linux console.

Actually it's possible to start emacs -Q -nw within ansi-term but
then Shift-Control-: is not recognized either.

It would be great if it's possible to avoid such key combos ...

Ciao, Gregor
-- 
 -... --- .-. . -.. ..--.. ...-.-

Re: terminal emulators

[Brett Viren]

[-- Attachment #1: Type: text/plain, Size: 902 bytes --]

Gregor Zattler <telegraph@gmx.net> writes:

> Hi Achim,
> * Achim Gratz <Stromeko@nexgo.de> [10. Feb. 2014]:
>> You cannot enter C-: in some terminals because it would require
>> simultaneous processing of shift and control (these terminals ignore
>> shift while control is pressed).
>
> this is true for xterm, rxvt-unicode, gnome-terminal, konsole and
> the linux console.

This terminal "feature" surprised me so I checked a few terminals I have
here.

xterm on Debian (278-4) does pass the C-:.  Testing with C-c C-: after
starting "emacs -nw -q" in that xterm I get the expected:

  C-c C-: is undefined

But, I confirm that mate-terminal (1.6.1-1.1+7.wheezy) and rxvt
(1:2.6.4-14) strip off the Ctrl.  There the test produces:

  C-c : is undefined

I note that mate-terminal does process Ctrl+Shift as some of its
shortcuts use this combo.  So, it's "actively" stripping the Ctrl away.

-Brett.

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: terminal emulators

[Achim Gratz]

Gregor Zattler writes:
>> You cannot enter C-: in some terminals because it would require
>> simultaneous processing of shift and control (these terminals ignore
>> shift while control is pressed).
>
> this is true for xterm, rxvt-unicode, gnome-terminal, konsole and
> the linux console.

Terminal emulators and certainly all of those you've listed usually pass
this on correctly unless they are explicitly configured (by the user or
the application using them) not to (at least when in an UTF-8 locale).
Since Emacs tries to use the full terminal capabilities you must have
configured something that prevents this from being enabled by Emacs if
it doesn't work for you.  Note that some X kbd configurations take over
Shift+Control (usually just for left or right) to emulate hyper or
super, but that would cause Emacs to recognize these events instead, I'd
think.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptations for Waldorf Q V3.00R3 and Q+ V3.54R2:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada

Re: terminal emulators

[Gregor Zattler]

Hi Achim,
* Achim Gratz <Stromeko@nexgo.de> [11. Feb. 2014]:
> Gregor Zattler writes:
>>> You cannot enter C-: in some terminals because it would require
>>> simultaneous processing of shift and control (these terminals ignore
>>> shift while control is pressed).
>>
>> this is true for xterm, rxvt-unicode, gnome-terminal, konsole and
>> the linux console.
> 
> Terminal emulators and certainly all of those you've listed usually pass
> this on correctly unless they are explicitly configured (by the user or
> the application using them) not to (at least when in an UTF-8 locale).

Huh?  I added a new user account to my debian/testing system with
no further customization whatsoever and renamed xorg.conf and
/etc/default/keyboard.  Then I started X as this new user, opened
a xterm and a rxtv-unicode, started emacs -Q -nw, did C-h k
(describe key) and hit C-: which is Shift-Control-. on a german
qwertz keyboard.

It says

: runs the command self-insert-command, which is an interactive
built-in function in `C source code'.

I never used a terminal emulator which would let me use
Shift-Control key combos.  Actually I would be very interested in
working Shift-Ctrl key combos.

> Since Emacs tries to use the full terminal capabilities you must have
> configured something that prevents this from being enabled by Emacs if
> it doesn't work for you.  Note that some X kbd configurations take over
> Shift+Control (usually just for left or right)

ATM I can't test with different (left, right) control keys since
I'm at a MacBook, there sadly is only one control key.

> to emulate hyper or
> super, but that would cause Emacs to recognize these events instead, I'd
> think.


You are able to type C-: in an emacs running in a terminal emulator?


Ciao; Gregor

Re: terminal emulators

[Achim Gratz]

Gregor Zattler writes:
> I never used a terminal emulator which would let me use
> Shift-Control key combos.  Actually I would be very interested in
> working Shift-Ctrl key combos.

I'm not sure when and where I've had that working or I may be
misremembering another combination.  In any case, in konsole
Ctrl-Shift-. is a Hotkey and can be mapped to Ctrl-:, so whatever
application uses the terminal never gets to see that combo anyway.
Similarly in GNOME a lot of the Ctrl-Shift plane is taken up by standard
shortcuts (which you will have to disable first).  In any case, to make
it work you'd need to switch the terminal mode so that it communicates
the modifier keys out-of-band (as CSI escape sequences AFAIK) to the
application.  There's no way to have it in normal mode and be able to
send Ctrl-Shift anything since there simply aren't any codings for this.
In XTerm this can be controlled via the modifyKeys resource IIRC.  There
are other modes like the old X10 terminal mode that synthesize key
events, too.

> You are able to type C-: in an emacs running in a terminal emulator?

Right now, no.  I haven't configured this in years and thankfully it has
been even longer I've had to program through an actual serial line (at
1200baud Emacs isn't going to work well and you're very thankful for a
vi, BTW).


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Wavetables for the Terratec KOMPLEXER:
http://Synth.Stromeko.net/Downloads.html#KomplexerWaves

Re: [RFC] Proposal for rebindings in Org 8.3

[Greg Troxel]

[-- Attachment #1: Type: text/plain, Size: 1495 bytes --]


Bastien <bzg@gnu.org> writes:

> | Key   | Command                           | Proposal         | Status |
> |-------+-----------------------------------+------------------+--------|
> | C-c # | Checkboxes                        | C-c x            | Free   |
> | C-c ~ | Cooperation                       | C-c C-~          | Free   |
> | C-c , | Priorities                        | C-c C-,          | Free   |
> | C-c ? | Editing and debugging formulas    | C-c C-?          | Free   |
> | C-c ! | Creating timestamps               | C-c C-!          | Free   |

I tried C-c C-! in my environment, and it fails, no noticing the C-!
(which involves shift) keypress (with ^H k; I get it that this is
proposed).  I have C-1 bound in my window manager to switch desktops,
since that binding doesn't take away the ability to generate any ASCII
character.

I'm running "emacs -nw" under tmux on one machine (netbsd), connected
via ssh from a mac using Terminal.

While one can argue that various emulations are broken, org should be
fully usable with a 7-bit terminal connection, and non-kludgy with an
8-bit connection.  In general, I find that emacs works fine with that,
although one has to prefix with ESC instead of the meta key.

I've always been bothered by keybindings like C-S-left, which while
useful, cause there to be no available keystroke sequence to perform the
function.

I find this surprising; I'd expect within emacs/org culture there to be
more people using terminal-mode emacs.

[-- Attachment #2: Type: application/pgp-signature, Size: 180 bytes --]