Release: Org-mode 5.09

Release: Org-mode 5.09

[Carsten Dominik]

Org-mode 5.09 is available at http://orgmode.org.

Enjoy!

- Carsten

Version 5.09
------------

* Overview

   - Taking a note upon TODO state changes can be restricted to
     selected states.

   - The format in which dates are shown in the daily/weekly
     agenda can be configured.

   - The default for `org-remember-store-without-prompt' is now t.

   - `org-goto' has been made into a general lookup command.

   - Priority cycling goes back to the nil state.

   - You can store a remember note to the *last used* location.

   - On Emacs 23, the headline faces for org-mode are now
     inherited from the outline faces.

* Incompatible Changes

   - The default for `org-remember-store-without-prompt' is now
     t, in order to better match the original intent of
     remember.el (storing a note with minimum interruption of
     work flow).  I expect that many people will be hit by this
     incompatible change - nevertheless I believe it is the right
     thing to do.

* Details

   - You can now select specific states for recording a note when
     switching to that state.  With the setting

      #+SEQ_TODO: TODO(t) ORDERED(o@) INVOICE(i@) PAYED(p) | RECEIVED(r)
      #+STARTUP: lognotestate

     only the states ORDERED and INVOICE will record a timestamp
     and a note.

   - You can now set the format of the string for each day in the
     agenda and timeline buffers.  You can use a format string
     interpreted by `format-time-string', or you can write your
     own function.  Configure the new variable
     `org-agenda-format-date'.  Thanks to Levin for triggering
     this development with a patch.

   - The default for `org-remember-store-without-prompt' is now
     t, in order to better match the original intent of
     remember.el (storing a note with minimum interruption of
     work flow).  Since we can assign files and headlines to
     templates, I guess this takes care of selecting a filing
     location in most cases.  For interactive filing, you now
     need a prefix command when exiting `remember'.

   - `org-goto' (bound to `C-c C-j') now uses an indirect buffer
     and has additional commands enabled: Org-occur with `C-c /'
     or even faster with `/', and the commands needed to select
     and copy a region.  This make `org-goto' a more general
     lookup command instead of only a jumping command.  Remember
     that you can exit with `Q' to go back to the original
     location.  Thanks to William Henney for this idea.

   - Setting the priority with S-up/down now cycles back to a
     state where no priority is specified.  This was requested by
     Rick Moynihan.

   - You can store a remember note to the *last used* location.
     So if you select a location interactively once, you can
     re-use it without having to find it again.  For this, exit
     the remember buffer with `C-u C-u C-c C-c'.  The leading
     comment in the remember buffer will tell exactly where the
     note goes if you exit with a particular command.
     Thanks to Maxim Loginov for this idea.

   - On Emacs 23, the headline faces for org-mode are now
     inherited from the outline faces.  This is just a
     convenience, so that you only have to configure one set of
     faces, and that will then be outline-1 .. outline-8.  You
     will actually not see any difference in org-mode, because
     Stefan Monnier has made the outline faces in Emacs 23 to
     match the current org-mode faces.

     This change does not effect XEmacs, nor Emacs 21 and 22.

   - Bug fixes.

Re: Release: Org-mode 5.09

[Rainer Stengele]

Carsten Dominik schrieb:
> Org-mode 5.09 is available at http://orgmode.org.
> 
> Enjoy!
> 
> - Carsten
> 
> Version 5.09
> ------------
> 
> * Overview
> 
>   - Taking a note upon TODO state changes can be restricted to
>     selected states.
> 
>   - The format in which dates are shown in the daily/weekly
>     agenda can be configured.
> 
>   - The default for `org-remember-store-without-prompt' is now t.
> 
>   - `org-goto' has been made into a general lookup command.
> 
>   - Priority cycling goes back to the nil state.
> 
>   - You can store a remember note to the *last used* location.
> 
>   - On Emacs 23, the headline faces for org-mode are now
>     inherited from the outline faces.
> 
> * Incompatible Changes
> 
>   - The default for `org-remember-store-without-prompt' is now
>     t, in order to better match the original intent of
>     remember.el (storing a note with minimum interruption of
>     work flow).  I expect that many people will be hit by this
>     incompatible change - nevertheless I believe it is the right
>     thing to do.
> 
> * Details
> 
>   - You can now select specific states for recording a note when
>     switching to that state.  With the setting
> 
>      #+SEQ_TODO: TODO(t) ORDERED(o@) INVOICE(i@) PAYED(p) | RECEIVED(r)
>      #+STARTUP: lognotestate
> 
>     only the states ORDERED and INVOICE will record a timestamp
>     and a note.
> 
>   - You can now set the format of the string for each day in the
>     agenda and timeline buffers.  You can use a format string
>     interpreted by `format-time-string', or you can write your
>     own function.  Configure the new variable
>     `org-agenda-format-date'.  Thanks to Levin for triggering
>     this development with a patch.
> 
>   - The default for `org-remember-store-without-prompt' is now
>     t, in order to better match the original intent of
>     remember.el (storing a note with minimum interruption of
>     work flow).  Since we can assign files and headlines to
>     templates, I guess this takes care of selecting a filing
>     location in most cases.  For interactive filing, you now
>     need a prefix command when exiting `remember'.
> 
>   - `org-goto' (bound to `C-c C-j') now uses an indirect buffer
>     and has additional commands enabled: Org-occur with `C-c /'
>     or even faster with `/', and the commands needed to select
>     and copy a region.  This make `org-goto' a more general
>     lookup command instead of only a jumping command.  Remember
>     that you can exit with `Q' to go back to the original
>     location.  Thanks to William Henney for this idea.
> 
>   - Setting the priority with S-up/down now cycles back to a
>     state where no priority is specified.  This was requested by
>     Rick Moynihan.
> 
>   - You can store a remember note to the *last used* location.
>     So if you select a location interactively once, you can
>     re-use it without having to find it again.  For this, exit
>     the remember buffer with `C-u C-u C-c C-c'.  The leading
>     comment in the remember buffer will tell exactly where the
>     note goes if you exit with a particular command.
>     Thanks to Maxim Loginov for this idea.
> 
>   - On Emacs 23, the headline faces for org-mode are now
>     inherited from the outline faces.  This is just a
>     convenience, so that you only have to configure one set of
>     faces, and that will then be outline-1 .. outline-8.  You
>     will actually not see any difference in org-mode, because
>     Stefan Monnier has made the outline faces in Emacs 23 to
>     match the current org-mode faces.
> 
>     This change does not effect XEmacs, nor Emacs 21 and 22.
> 
>   - Bug fixes.
> 
> 
> 
> _______________________________________________
> Emacs-orgmode mailing list
> Remember: use `Reply All' to send replies to the list.
> Emacs-orgmode@gnu.org
> http://lists.gnu.org/mailman/listinfo/emacs-orgmode
> 

After "C-c C-j" trying to find items with a certain regex which should 
match a lot of entries finds nothing but shows:
"before first heading".
Some other searches do succeed.

Ah - now I see: Looks like the first match is in the top
:SETUP:
...
:END:
section. Which ends up with the problem
This is probably a bug?

Rainer

Re: Re: Release: Org-mode 5.09

[Carsten Dominik]

Fixed, thanks.

- Carsten

On Sep 15, 2007, at 23:25, Rainer Stengele wrote:

> Carsten Dominik schrieb:
>> Org-mode 5.09 is available at http://orgmode.org.
>> Enjoy!
>> - Carsten
>> Version 5.09
>> ------------
>> * Overview
>>   - Taking a note upon TODO state changes can be restricted to
>>     selected states.
>>   - The format in which dates are shown in the daily/weekly
>>     agenda can be configured.
>>   - The default for `org-remember-store-without-prompt' is now t.
>>   - `org-goto' has been made into a general lookup command.
>>   - Priority cycling goes back to the nil state.
>>   - You can store a remember note to the *last used* location.
>>   - On Emacs 23, the headline faces for org-mode are now
>>     inherited from the outline faces.
>> * Incompatible Changes
>>   - The default for `org-remember-store-without-prompt' is now
>>     t, in order to better match the original intent of
>>     remember.el (storing a note with minimum interruption of
>>     work flow).  I expect that many people will be hit by this
>>     incompatible change - nevertheless I believe it is the right
>>     thing to do.
>> * Details
>>   - You can now select specific states for recording a note when
>>     switching to that state.  With the setting
>>      #+SEQ_TODO: TODO(t) ORDERED(o@) INVOICE(i@) PAYED(p) | 
>> RECEIVED(r)
>>      #+STARTUP: lognotestate
>>     only the states ORDERED and INVOICE will record a timestamp
>>     and a note.
>>   - You can now set the format of the string for each day in the
>>     agenda and timeline buffers.  You can use a format string
>>     interpreted by `format-time-string', or you can write your
>>     own function.  Configure the new variable
>>     `org-agenda-format-date'.  Thanks to Levin for triggering
>>     this development with a patch.
>>   - The default for `org-remember-store-without-prompt' is now
>>     t, in order to better match the original intent of
>>     remember.el (storing a note with minimum interruption of
>>     work flow).  Since we can assign files and headlines to
>>     templates, I guess this takes care of selecting a filing
>>     location in most cases.  For interactive filing, you now
>>     need a prefix command when exiting `remember'.
>>   - `org-goto' (bound to `C-c C-j') now uses an indirect buffer
>>     and has additional commands enabled: Org-occur with `C-c /'
>>     or even faster with `/', and the commands needed to select
>>     and copy a region.  This make `org-goto' a more general
>>     lookup command instead of only a jumping command.  Remember
>>     that you can exit with `Q' to go back to the original
>>     location.  Thanks to William Henney for this idea.
>>   - Setting the priority with S-up/down now cycles back to a
>>     state where no priority is specified.  This was requested by
>>     Rick Moynihan.
>>   - You can store a remember note to the *last used* location.
>>     So if you select a location interactively once, you can
>>     re-use it without having to find it again.  For this, exit
>>     the remember buffer with `C-u C-u C-c C-c'.  The leading
>>     comment in the remember buffer will tell exactly where the
>>     note goes if you exit with a particular command.
>>     Thanks to Maxim Loginov for this idea.
>>   - On Emacs 23, the headline faces for org-mode are now
>>     inherited from the outline faces.  This is just a
>>     convenience, so that you only have to configure one set of
>>     faces, and that will then be outline-1 .. outline-8.  You
>>     will actually not see any difference in org-mode, because
>>     Stefan Monnier has made the outline faces in Emacs 23 to
>>     match the current org-mode faces.
>>     This change does not effect XEmacs, nor Emacs 21 and 22.
>>   - Bug fixes.
>> _______________________________________________
>> Emacs-orgmode mailing list
>> Remember: use `Reply All' to send replies to the list.
>> Emacs-orgmode@gnu.org
>> http://lists.gnu.org/mailman/listinfo/emacs-orgmode
>
> After "C-c C-j" trying to find items with a certain regex which should 
> match a lot of entries finds nothing but shows:
> "before first heading".
> Some other searches do succeed.
>
> Ah - now I see: Looks like the first match is in the top
> :SETUP:
> ...
> :END:
> section. Which ends up with the problem
> This is probably a bug?
>
> Rainer
>
>
>
> _______________________________________________
> Emacs-orgmode mailing list
> Remember: use `Reply All' to send replies to the 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: Release: Org-mode 5.09

[Xavier Maillard]

Hi Carsten,

   Org-mode 5.09 is available at http://orgmode.org.

      - Taking a note upon TODO state changes can be restricted to
	selected states.

Perfect !

      - The default for `org-remember-store-without-prompt' is now t.

Yeah, one more line will disappear from .emacs for me ;)

      - Priority cycling goes back to the nil state.

Perfect !

      - You can store a remember note to the *last used* location.

Perfect !

It seems this release is addressing many of my complains. Thank
you very much.

	Xavier
-- 
http://www.gnu.org
http://www.april.org
http://www.lolica.org

Re: Release: Org-mode 5.09

[Levin]

Hi Carsten,

>    - Priority cycling goes back to the nil state.

Priority cycling is a bit odd. When I use M-n to cycle, it turns 
from #C-> nil-> #C-> nil, and M-p will be #A->nil->#A->nil.

So I write this small patch to achieve: M-n will cycle from 
#A->#B->#C->nil->#A, and M-p will cycle from #C->#B->#A->nil->#C.

Hope it helps.

-Levin

;; This patch is against org 5.09

diff --git a/lisp/org/org.el b/lisp/org/org.el
index f4746b4..a5567f5 100644
--- a/lisp/org/org.el
+++ b/lisp/org/org.el
@@ -13462,7 +13462,11 @@ ACTION can be `set', `up', `down', or a character."
       (if (looking-at org-priority-regexp)
 	  (setq current (string-to-char (match-string 2))
 		have t)
-	(setq current org-default-priority))
+	(setq current (if (eq action 'up)
+			  (1+ org-lowest-priority)
+			(if (eq action 'down)
+			    (1- org-highest-priority)
+			  org-default-priority))))
       (cond
        ((or (eq action 'set) (integerp action))
 	(if (integerp action)

Re: Release: Org-mode 5.09

[Levin]

There's a small bug with org-fast-todo-selection, pressing " " will trigger an 
error instead of clearing the todo state.

Is this small fix ok?

-Levin

diff --git a/lisp/org/org.el b/lisp/org/org.el
index a5567f5..edb5cb2 100644
--- a/lisp/org/org.el
+++ b/lisp/org/org.el
@@ -13058,7 +13058,7 @@ Returns the new TODO keyword, or nil if no state 
change should occur."
        ((or (= c ?\C-g)
 	    (and (= c ?q) (not (rassoc c fulltable))))
 	(setq quit-flag t))
-       ((= c ?\ ) 'none)
+       ((= c ?\ ) nil)
        ((setq e (rassoc c fulltable) tg (car e))
 	tg)
        (t (setq quit-flag t))))))

Re: Release: Org-mode 5.09

[Levin]

Another thing, if no todo state speedkey is specified, C-u C-c C-t will not 
function correctly.

Test file:
------8<--------8<--------
#+SEQ_TODO: TODO STARTED WAITING DELEGATED APPT | DONE DEFERRED CANCELLED

* task1
* task2
------8<--------8<--------


-Levin

Re: Release: Org-mode 5.09

[Carsten Dominik]

On Sep 18, 2007, at 11:00, Levin wrote:

> Hi Carsten,
>
>>    - Priority cycling goes back to the nil state.
>
> Priority cycling is a bit odd. When I use M-n to cycle, it turns
> from #C-> nil-> #C-> nil, and M-p will be #A->nil->#A->nil.

I did it like this on purpose, because I wanted S-up to
immediately increase the priority.  The default priority is
#B.  I know this works differently from the other cycling mechanisms
in Org.  However, with you implementation you need to press

S-up S-up S-up to finally get to #A, or you oddly need to use S-down(!)
in increase the priority.  I can see argument for both, open to 
discussion!

- Carsten

>
> So I write this small patch to achieve: M-n will cycle from
> #A->#B->#C->nil->#A, and M-p will cycle from #C->#B->#A->nil->#C.
>
> Hope it helps.
>
> -Levin
>
> ;; This patch is against org 5.09
>
> diff --git a/lisp/org/org.el b/lisp/org/org.el
> index f4746b4..a5567f5 100644
> --- a/lisp/org/org.el
> +++ b/lisp/org/org.el
> @@ -13462,7 +13462,11 @@ ACTION can be `set', `up', `down', or a 
> character."
>        (if (looking-at org-priority-regexp)
>  	  (setq current (string-to-char (match-string 2))
>  		have t)
> -	(setq current org-default-priority))
> +	(setq current (if (eq action 'up)
> +			  (1+ org-lowest-priority)
> +			(if (eq action 'down)
> +			    (1- org-highest-priority)
> +			  org-default-priority))))
>        (cond
>         ((or (eq action 'set) (integerp action))
>  	(if (integerp action)
>
>
> _______________________________________________
> Emacs-orgmode mailing list
> Remember: use `Reply All' to send replies to the 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: Release: Org-mode 5.09

[Carsten Dominik]

On Sep 18, 2007, at 11:27, Levin wrote:

> There's a small bug with org-fast-todo-selection, pressing " " will 
> trigger an
> error instead of clearing the todo state.
>
> Is this small fix ok?

Yes, thanks.

- Carsten

>
> -Levin
>
> diff --git a/lisp/org/org.el b/lisp/org/org.el
> index a5567f5..edb5cb2 100644
> --- a/lisp/org/org.el
> +++ b/lisp/org/org.el
> @@ -13058,7 +13058,7 @@ Returns the new TODO keyword, or nil if no 
> state
> change should occur."
>         ((or (= c ?\C-g)
>  	    (and (= c ?q) (not (rassoc c fulltable))))
>  	(setq quit-flag t))
> -       ((= c ?\ ) 'none)
> +       ((= c ?\ ) nil)
>         ((setq e (rassoc c fulltable) tg (car e))
>  	tg)
>         (t (setq quit-flag t))))))
>
>
> _______________________________________________
> Emacs-orgmode mailing list
> Remember: use `Reply All' to send replies to the 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: Release: Org-mode 5.09

[Carsten Dominik]

Fixed, thanks.

- Carsten

On Sep 18, 2007, at 11:38, Levin wrote:

> Another thing, if no todo state speedkey is specified, C-u C-c C-t 
> will not
> function correctly.
>
> Test file:
> ------8<--------8<--------
> #+SEQ_TODO: TODO STARTED WAITING DELEGATED APPT | DONE DEFERRED 
> CANCELLED
>
> * task1
> * task2
> ------8<--------8<--------
>
>
> -Levin
>
>
> _______________________________________________
> Emacs-orgmode mailing list
> Remember: use `Reply All' to send replies to the 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: Release: Org-mode 5.09

[Bastien]

Carsten Dominik <carsten.dominik@gmail.com> writes:

>> Priority cycling is a bit odd. When I use M-n to cycle, it turns
>> from #C-> nil-> #C-> nil, and M-p will be #A->nil->#A->nil.
>
> I did it like this on purpose, because I wanted S-up to
> immediately increase the priority.  

I would vote for something very simple: 

<S-up> increases priority, <S-down> decreases priority.

<S-up>   : nil -> [#A] -> [#B] -> [#C] -> nil
           [#A] -> [#B] -> [#C] -> nil -> [#A]
           [#B] -> [#C] -> nil -> [#A] -> [#B]
           [...]

<S-down> : nil -> [#C] -> [#B] -> [#A] -> nil
           [#A] -> nil -> [#C] -> [#B] -> [#A]
           [#B] -> [#A] -> nil -> [#C] -> [#B]
           [...]

I know it wouldn't be aware of the default priority, but I would still
prefer this behavior over another one.

For making the priority setting aware of the default priority I would
better use this:

<S-up>   : increase priority (by with normal cycling)
<S-down> : switch default priority / nil

But I still prefer the very simple interface... as I tend to think
priority handling should be as smooth (and sober) as possible.

-- 
Bastien

Re: Release: Org-mode 5.09

[Carsten Dominik]

On Sep 19, 2007, at 2:49, Bastien wrote:

> Carsten Dominik <carsten.dominik@gmail.com> writes:
>
>>> Priority cycling is a bit odd. When I use M-n to cycle, it turns
>>> from #C-> nil-> #C-> nil, and M-p will be #A->nil->#A->nil.
>>
>> I did it like this on purpose, because I wanted S-up to
>> immediately increase the priority.
>
> I would vote for something very simple:
>
> <S-up> increases priority, <S-down> decreases priority.
>
> <S-up>   : nil -> [#A] -> [#B] -> [#C] -> nil
>            [#A] -> [#B] -> [#C] -> nil -> [#A]
>            [#B] -> [#C] -> nil -> [#A] -> [#B]
>            [...]
>
> <S-down> : nil -> [#C] -> [#B] -> [#A] -> nil
>            [#A] -> nil -> [#C] -> [#B] -> [#A]
>            [#B] -> [#A] -> nil -> [#C] -> [#B]
>            [...]


I do find this *totally* confusing.  In this, S-up will *decrease*
the priority.

- Carsten



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

Re: Release: Org-mode 5.09

[Carsten Dominik]

On Sep 19, 2007, at 2:49, Bastien wrote:

> Carsten Dominik <carsten.dominik@gmail.com> writes:
>
>>> Priority cycling is a bit odd. When I use M-n to cycle, it turns
>>> from #C-> nil-> #C-> nil, and M-p will be #A->nil->#A->nil.
>>
>> I did it like this on purpose, because I wanted S-up to
>> immediately increase the priority.
>
> I would vote for something very simple:
>
> <S-up> increases priority, <S-down> decreases priority.
>
> <S-up>   : nil -> [#A] -> [#B] -> [#C] -> nil
>            [#A] -> [#B] -> [#C] -> nil -> [#A]
>            [#B] -> [#C] -> nil -> [#A] -> [#B]
>            [...]
>
> <S-down> : nil -> [#C] -> [#B] -> [#A] -> nil
>            [#A] -> nil -> [#C] -> [#B] -> [#A]
>            [#B] -> [#A] -> nil -> [#C] -> [#B]
>            [...]

How about this:

S-up   : nil -> [#A] -> nil -> [#C] -> [#B] -> [#A] -> nil -> [#C] ...

S-sown : nil -> [#C] -> nil -> [#A] -> [#B] -> [#C] -> nil -> [#A] ...

So the first keypress will really get you up/down from the default
priority, and after that it is normal cycling.  Maybe it is clearer
in an example with 5 priorities, A..E, default C:

S-up : nil -> [#B] -> [#A] -> nil -> [#E] -> [#D] ->
               [#C] -> [#B] -> [#A] -> nil -> [#E] ...

Comments?

- Carsten



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

Re: Release: Org-mode 5.09

[Bastien]

Carsten Dominik <dominik@science.uva.nl> writes:

> How about this:
>
> S-up   : nil -> [#A] -> nil -> [#C] -> [#B] -> [#A] -> nil -> [#C] ...
> S-sown : nil -> [#C] -> nil -> [#A] -> [#B] -> [#C] -> nil -> [#A] ...

Yes, the general scheme looks fine to me. 

The only thing that still tickles me here is that the *default* priority
is not the *easiest* to assign.  So why not this:

S-up   : nil -> [#B] -> [#A] -> nil -> [#C] -> [#B] -> [#A] -> nil...
S-sown : nil -> [#B] -> [#C] -> nil -> [#A] -> [#B] -> [#C] -> nil...

As obvious, the problem in this case is that different keys first assign
the same (default) priority.

Anyway, I won't linger on this further since anyone seems to be okay
with the proposed solution!

-- 
Bastien

Re: Release: Org-mode 5.09

[Nuutti Kotivuori]

Bastien wrote:
> The only thing that still tickles me here is that the *default* priority
> is not the *easiest* to assign.  So why not this:

I've never understood what's the difference between a line with the
default priority and a line without a priority at all. That is, with
the default settings:

* [#B] Foo
* Bar

Are these not equivalent in priority sorting? If so, why should the
default priority be ever explicitly said?

-- Naked

Re: Release: Org-mode 5.09

[Bernt Hansen]

Nuutti Kotivuori <naked@iki.fi> writes:

> Bastien wrote:
>> The only thing that still tickles me here is that the *default* priority
>> is not the *easiest* to assign.  So why not this:
>
> I've never understood what's the difference between a line with the
> default priority and a line without a priority at all. That is, with
> the default settings:
>
> * [#B] Foo
> * Bar
>
> Are these not equivalent in priority sorting? If so, why should the
> default priority be ever explicitly said?

I never use [#B] priority.  I only set A and C.  I'm in the habit of
globally replacing [#B] with nothing since I think the task item looks
much cleaner without it (since it shows up in timelog reports and on the
agenda).

Bernt

Re: Re: Release: Org-mode 5.09

[Bastien]

Nuutti Kotivuori <naked@iki.fi> writes:

> Bastien wrote:
>> The only thing that still tickles me here is that the *default* priority
>> is not the *easiest* to assign.  So why not this:
>
> I've never understood what's the difference between a line with the
> default priority and a line without a priority at all. 

Depends on what "default" stands for.   It can be either:

1. the first available state when setting priorities
2. the default state you *want* to use when you need to set a priority;
3. the state that neither increase or decrease the priority rating

>  That is, with
> the default settings:
>
> * [#B] Foo
> * Bar
>
> Are these not equivalent in priority sorting? 

For sorting with org-sort, yes. But I use this convention that any item
that has a priority - even the default priority - should be done before
other items.  For example:

* [#A] Foo
* [#B] Bar
* [#C] Fooo
* Baar

In this case, * Baar and * [#B] are not equivalent.

> If so, why should the default priority be ever explicitly said?

To quickly be able to choose this default priority when cycling? :)

-- 
Bastien

Re: Re: Release: Org-mode 5.09

[John Wiegley]

Bernt Hansen <bernt@alumni.uwaterloo.ca> writes:

> I never use [#B] priority.  I only set A and C.  I'm in the habit of
> globally replacing [#B] with nothing since I think the task item looks much
> cleaner without it (since it shows up in timelog reports and on the agenda).

Heh, I'm exactly the opposite.  I use the "no priority" state as a way of
letting me know that I haven't considered the task yet.  And even then,
defaulting to B makes things work out nicely.

John

Re: Release: Org-mode 5.09

[Rick Moynihan]

Bastien wrote:
> Carsten Dominik <dominik@science.uva.nl> writes:
 >
> S-up   : nil -> [#B] -> [#A] -> nil -> [#C] -> [#B] -> [#A] -> nil...
> S-sown : nil -> [#B] -> [#C] -> nil -> [#A] -> [#B] -> [#C] -> nil...
> 
> As obvious, the problem in this case is that different keys first assign
> the same (default) priority.
> 

Hi all, sorry for chipping in late on these discussions (holiday), but I 
think this proposal is closest to what I had in mind.

R.