Things that I'm removing from the todo list

Things that I'm removing from the todo list

[Lars Magne Ingebrigtsen]

These are items that I've previously said that I was going to do --
but I've changed my mind.

* multiple article buffers
Having multiple summary buffers is confusing enough; multiple article
buffers would be nice, but not that nice.

* selecting an article in a binary group should just run it through gnus-uu
I see no real use for that.  `X u' isn't that hard to type.

* allow posting with IHAVE
I've been told repeatedly that this is Not The Done Thing.

* "tapestry.el" -- allowing the user to `C-x 2' and `C-x 3' 'till
  satisfied, and then saving the window configuration
Kale Jones has already written that, so it feels stupid to
re-implement it.  

-- 
Home is where the cat is.

Re: Things that I'm removing from the todo list

[Joe Hildebrand]

> "Lars" == Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

   Lars> * "tapestry.el" -- allowing the user to `C-x 2' and `C-x 3'
   Lars> 'till satisfied, and then saving the window configuration
   Lars> Kale Jones has already written that, so it feels stupid to
   Lars> re-implement it.

did you ever figure out how to allow this window configuration, then?

-------------------------------------
| groups         | summary          |
-------------------------------------
|         article                   |
-------------------------------------

-- 
Joe Hildebrand                  Fuentez Systems Concepts
hildjj@fuentez.com              11781 Lee-Jackson Hwy, Suite 700
Lead Software Engineer          Fairfax, VA 22033
	"Breakfast recapitulates phylogeny" - Spider Robinson

Re: Things that I'm removing from the todo list

[Lars Magne Ingebrigtsen]

Joe Hildebrand <hildjj@fuentez.com> writes:

> did you ever figure out how to allow this window configuration, then?
> 
> -------------------------------------
> | groups         | summary          |
> -------------------------------------
> |         article                   |
> -------------------------------------

Yup; it'll be in September 0.23.  

The new recursive window config thing is really neat.  Just finding
the right structure for data is *so* important.  I still don't
remember who came up with this structure, but it seems very usable.  I
did a complex test -- I envisioned something that would look like:

--------------------------------
|     |                 |      |
|     |                 |      |
|     |-------------------------
|     |           |            |
------|           |            |
|     |           |            |
|     |           |            |
--------------------------------

And it came out right (almost) on the first try.

Recursive structures *rule*.  :-)

-- 
Home is where the cat is.

Re: Things that I'm removing from the todo list

[Mark Eichin]

> did a complex test -- I envisioned something that would look like:

ah, but what about

-------------------------
|	|       	|
|	----------------|
|	|      |	|
|	|      |	|
|---------------	|
|	       |	|
-------------------------

(of course, I'm pretty sure that emacs window layout can't *do* that,
but it is the standard "bane of tiling window system" example :-)

					_Mark_ <eichin@cygnus.com>
					Cygnus Support, Eastern USA

Re: Things that I'm removing from the todo list

[Lars Magne Ingebrigtsen]

Mark Eichin <eichin@cygnus.com> writes:

> ah, but what about
> 
> -------------------------
> |	|       	|
> |	----------------|
> |	|      |	|
> |	|      |	|
> |---------------	|
> |	       |	|
> -------------------------
> 
> (of course, I'm pretty sure that emacs window layout can't *do* that,
> but it is the standard "bane of tiling window system" example :-)

It is indeed impossible.  (At least I think so.)  You can't even do
something as simple as:

-----------------
|   |           |
|   |           |
-----           |
|               |
|               |
-----------------

Can you?

-- 
Home is where the cat is.

Re: Things that I'm removing from the todo list

[eichin]

>It is indeed impossible.  (At least I think so.)  You can't even do
>something as simple as:

Ahh, but the second example contained non-rectangular windows;
everything in my example was rectangular, so it should (in some
theoretical way) be "simpler". Still can't obtain it via "split"
operations though...

Re: Things that I'm removing from the todo list

[Michael Cook]

>>>>> "LMI" == Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

 LMI> You can't even do something as simple as:

 LMI> -----------------
 LMI> |   |           |
 LMI> |   |           |
 LMI> -----           |
 LMI> |               |
 LMI> |               |
 LMI> -----------------

 LMI> Can you?

Simple?  But you've redefined the problem.  That layout has a non-rectangular
window.

Michael.

Frame Configuration (was Re: Things that I'm removing...)

[Edward J. Sabol]

Excerpts from mail: (15-Dec-95) Re: Things that I'm removing from the todo list by Lars Magne Ingebrigtsen
> The new recursive window config thing is really neat.  Just finding
> the right structure for data is *so* important.  I still don't
> remember who came up with this structure, but it seems very usable.  I
> did a complex test -- I envisioned something that would look like:
> 
> --------------------------------
> |     |                 |      |
> |     |                 |      |
> |     |-------------------------
> |     |           |            |
> ------|           |            |
> |     |           |            |
> |     |           |            |
> --------------------------------
> 
> And it came out right (almost) on the first try.
> 
> Recursive structures *rule*.  :-)

Unfortunately, they are confusing. I've read and studied the documentation
you wrote for these structures, and I still can't figure it out. Maybe I
don't understand your notation. Either way, could you post the value of
gnus-frame-configuration that you used in the above test example? I think I
can get a better feel for it with an actual working example.

Thanks,
Ed

Re: Frame Configuration (was Re: Things that I'm removing...)

[Lars Magne Ingebrigtsen]

"Edward J. Sabol" <sabol@thuban.gsfc.nasa.gov> writes:

> > --------------------------------
> > |     |                 |      |
> > |     |                 |      |
> > |     |-------------------------
> > |     |           |            |
> > ------|           |            |
> > |     |           |            |
> > |     |           |            |
> > --------------------------------
> > 
> > Recursive structures *rule*.  :-)
> 
> Unfortunately, they are confusing. I've read and studied the
> documentation you wrote for these structures, and I still can't
> figure it out.

Well, the old version of `gnus-buffer-configuration' wasn't recursive,
which is why is was confusing.  :-)

> Maybe I don't understand your notation. Either way, could you post
> the value of gnus-frame-configuration that you used in the above
> test example?

(I've gone back to calling it `gnus-buffer-configuration' since the
new format is (mostly) a generalization of the old format.  The old
format should work with the new function, I hope.)

(gnus-configure-frame
  '(horizontal 1.0
    (vertical 10
      (group 1.0)
      (article 0.3 point))
    (vertical 1.0
      (horizontal 4
        (article 1.0)
        (group 10))
      (horizontal 1.0
        (group 1.0)
        (article .5)))))

Both the group buffer and the article buffer is here displayed
thrice, as you can see.

Just `C-x C-e' at the end of that form and be amused.

Despite the name and the documentation, multiple frames are still not
supported.  They will be in the next release.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@ifi.uio.no * Lars Ingebrigtsen

Re: Things that I'm removing from the todo list

[Mark Eichin]

> * multiple article buffers
> Having multiple summary buffers is confusing enough; multiple article
> buffers would be nice, but not that nice.

Umm, I've always felt that one of the big problems with RMAIL (and
most of the other emacs mail interfaces) is that you can only look at
the current article and no others. Oh well, just my random
opinion... not like I've ever been motivated enough to *fix* it or
anything :-)
					_Mark_ <eichin@cygnus.com>
					Cygnus Support, Eastern USA

Re: Things that I'm removing from the todo list

[Edward J. Sabol]

Excerpts from mail: (14-Dec-95) Things that I'm removing from the todo list by Lars Magne Ingebrigtsen
> These are items that I've previously said that I was going to do --
> but I've changed my mind.
> 
> * multiple article buffers
> Having multiple summary buffers is confusing enough; multiple article
> buffers would be nice, but not that nice.

ARGH!!!! No, say it ain't so! I really, *really* need this feature!

(Have a variable so people who are easily confused can turn it off to get the
current behavior.)

Re: Things that I'm removing from the todo list

[Lars Magne Ingebrigtsen]

"Edward J. Sabol" <sabol@thuban.gsfc.nasa.gov> writes:

> > * multiple article buffers
> > Having multiple summary buffers is confusing enough; multiple article
> > buffers would be nice, but not that nice.
> 
> ARGH!!!! No, say it ain't so! I really, *really* need this feature!
> 
> (Have a variable so people who are easily confused can turn it off to get the
> current behavior.)

I meant that it's confusing for me (as a programmer) to have to deal
with all these buffer-local variables.  But thinking some more about
it, I don't think adding multiple article buffers will add
significantly more complexity to the code.  The article buffers will
always be coupled with the summary buffers (so to speak), so perhaps
it would even simplify certain things.

So I've put it back in the todo list again.  

(Wishy-washy?  Moi?)

-- 
Home is where the cat is.

Re: Things that I'm removing from the todo list

[Hallvard B Furuseth]

[Lars Magne Ingebrigtsen]

> * multiple article buffers
> Having multiple summary buffers is confusing enough; multiple article
> buffers would be nice, but not that nice.

[Edward J. Sabol]

> ARGH!!!! No, say it ain't so! I really, *really* need this feature!
> 
> (Have a variable so people who are easily confused can turn it off to get the
> current behavior.)

A variable (or keystroke) to turn it *on*, you mean.  Don't confuse
people unless they request to be confused.  Complexifying features that
most people seldom or never use, should be off by default.


Regards,

Hallvard

Re: Things that I'm removing from the todo list

[Steven L. Baur]

>>>>> "Lars" == Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

    Lars> These are items that I've previously said that I was going
    Lars> to do -- but I've changed my mind.

    Lars> * multiple article buffers Having multiple summary buffers
    Lars> is confusing enough; multiple article buffers would be nice,
    Lars> but not that nice.

Hmm.  Having multiple articles buffers would help in assembling
various messages into summaries.  Right now I am resorting to xterm,
grep and less to look up multiple messages when researching some
messages, or the Gnus FAQ.

Perhaps you could put this on the TODO list for April Gnus?

-- 
steve@miranova.com baur

Re: Things that I'm removing from the todo list

[Per Abrahamsen]

>>>>> "LMI" == Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

LMI> * "tapestry.el" -- allowing the user to `C-x 2' and `C-x 3' 'till
LMI>   satisfied, and then saving the window configuration

LMI> Kale Jones has already written that, so it feels stupid to
LMI> re-implement it.  

Was he willing to sign papers for it?

How 'bout an interface to tapestry.el from Gnus?

Re: Things that I'm removing from the todo list

[Lars Magne Ingebrigtsen]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> LMI> Kale Jones has already written that, so it feels stupid to
> LMI> re-implement it.  
> 
> Was he willing to sign papers for it?

Uhm...  I think "no" would be the most accurate answer for that one. 

> How 'bout an interface to tapestry.el from Gnus?

Well -- yes.  But I'm not really interested in doing lots of work for
things that can only be used under XEmacs when it could work under
Emacs as well, but won't work for political reasons.  (*sigh*).
Working on toolbars and graphics and such for XEmacs feels ok for me,
though, since that can't work under Emacs.  Not that that makes any
sense what-so-ever.

-- 
Home is where the cat is.