Those `mailcrypt' "push OK" boxes

Those `mailcrypt' "push OK" boxes

[François Pinard]

Hi, gang.

`mailcrypt' has this habit to write nice boxes with an OK button as windows
over the Emacs frame, forcing my hand to leave the keyboard, play with
the mouse and click on the said button.  This is surely nice, but neither
efficient nor ergonomic.  Maybe I'm not ready for user-friendliness! :-)

So, I decided to counter-attack with the following kludge.  Maybe it could
serve other people as well?



;; Override message-box (so we avoid the mouse from mailcrypt, for example).
(defun message-box (fmt &rest args)
  (let ((string (apply (function format) fmt args)))
    (ediff-select-lowest-window)
    (split-window-vertically)
    (ediff-select-lowest-window)
    (with-temp-buffer
      (switch-to-buffer (current-buffer))
      (insert string)
      (goto-char (point-min))
      (let ((window-min-height 2))
	(shrink-window-if-larger-than-buffer))
      (let ((cursor-in-echo-area t)
	    key)
	(while (not (memq key '(?\C-m ?\C-j)))
	  (setq key (read-char-exclusive (format "OK? [Enter] "))))))
    (delete-window)))
(autoload 'ediff-select-lowest-window "ediff-wind")

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: Those `mailcrypt' "push OK" boxes

[Gareth Jones]

Francois Pinard <pinard@iro.umontreal.ca> writes:

> `mailcrypt' has this habit to write nice boxes with an OK button as windows
> over the Emacs frame, forcing my hand to leave the keyboard, play with
> the mouse and click on the said button.

Only if the mailcrypt window has focus.  If the main Emacs window has
focus then any key-press will remove the window.

-- 
Gareth Jones

What is written without effort is in general read without pleasure.
S. Johnson

Re: Those `mailcrypt' "push OK" boxes

[François Pinard]

gdj1@gdjones.demon.co.uk (Gareth Jones) writes:

> Francois Pinard <pinard@iro.umontreal.ca> writes:

> > `mailcrypt' has this habit to write nice boxes with an OK button as windows
> > over the Emacs frame, forcing my hand to leave the keyboard, play with
> > the mouse and click on the said button.

> Only if the mailcrypt window has focus.  If the main Emacs window has
> focus then any key-press will remove the window.

Indeed?  If I understand you fully, this might be considered as even worse,
as when one types a bit and fast (or in some case, use key macros), and
such things occur in practice, the window might be displayed and removed
without the user noticing.  And moreover, the cursor might randomly sit
where the mailcrypt window will occur, or not, which is not nice either.

P.S. - For enriched mode editing for sending message, I have something
that almost work now, unless a bad bug hits me.  It is short and nice code
so far.  If I'm lucky, I'll send you the results a bit later this afternoon
(I have to go fore and back to a neighbouring town _now_).

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: Those `mailcrypt' "push OK" boxes

[Stainless Steel Rat]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

* François Pinard <pinard@iro.umontreal.ca>  on Fri, 23 Apr 1999
| `mailcrypt' has this habit to write nice boxes with an OK button as windows
| over the Emacs frame, forcing my hand to leave the keyboard, play with
| the mouse and click on the said button.

It does?  Funny, it does not do that to me.

But then, I built my XEmacs without things like dialogue boxes, so
*NOTHING* can do anything like that to me.  I believe FSF Emacs can also be 
built without dialogues.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v0.9.5 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE3IyH0gl+vIlSVSNkRAh2oAJ9ts7dJvwuKNXWRBkUSjPjmq00SgACghHWT
igxv3QMIqIHh4yK/PFFEvl8=
=78v+
-----END PGP SIGNATURE-----

-- 
Rat <ratinox@peorth.gweep.net>    \ Do not use Happy Fun Ball on concrete.
Minion of Nathan - Nathan says Hi! \ 
PGP Key: at a key server near you!  \ 

Re: Those `mailcrypt' "push OK" boxes

[François Pinard]

Stainless Steel Rat <ratinox@peorth.gweep.net> écrit:

> * François Pinard <pinard@iro.umontreal.ca>  on Fri, 23 Apr 1999
> | `mailcrypt' has this habit to write nice boxes with an OK button as windows
> | over the Emacs frame, forcing my hand to leave the keyboard, play with
> | the mouse and click on the said button.

> It does?  Funny, it does not do that to me.

> But then, I built my XEmacs without things like dialogue boxes, so
> *NOTHING* can do anything like that to me.  I believe FSF Emacs can also be 
> built without dialogues.

I might not find optimal the way the user interface is done, yet I would
not dismiss that there is a need for some feedback to users about what
PGP has to say about signature verification, say.  How do you get it, then?

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: Those `mailcrypt' "push OK" boxes

[Stainless Steel Rat]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

* François Pinard <pinard@iro.umontreal.ca>  on Sun, 25 Apr 1999
| I might not find optimal the way the user interface is done, yet I would
| not dismiss that there is a need for some feedback to users about what
| PGP has to say about signature verification, say.  How do you get it,
| then?

In the minibuffer, the way <insert deity here> Intended Things To Be.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v0.9.5 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE3I4wxgl+vIlSVSNkRAnyfAJ4iq9Nb5amVIV7VscFaAzIrsTZSOQCg4aLr
h1Lika45D/dRfxCnyBzV53k=
=Br7z
-----END PGP SIGNATURE-----

-- 
Rat <ratinox@peorth.gweep.net>    \ When not in use, Happy Fun Ball should be
Minion of Nathan - Nathan says Hi! \ returned to its special container and
PGP Key: at a key server near you!  \ kept under refrigeration.

Re: Those `mailcrypt' "push OK" boxes

[François Pinard]

Stainless Steel Rat <ratinox@peorth.gweep.net> écrit:

> | I might not find optimal the way the user interface is done, yet I would
> | not dismiss that there is a need for some feedback to users about what
> | PGP has to say about signature verification, say.  How do you get it,
> | then?

> In the minibuffer, the way <insert deity here> Intended Things To Be.

And then, you get an accepting interaction there?  Something that merely
flashes in the mini-buffer and disappears at the next random keystroke could
easily go un-noticed.

Also, is the minibuffer guaranteed to extend for holding and displaying
the full diagnostic?  It is often multi-line.  Is there any required magic
for it?  (I use the resize-minibuffer mode, but it does not work perfectly
in all circumstances, at least from my experience.)

I'm surely open to any solution, as long as it does work dependably! :-)

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: Those `mailcrypt' "push OK" boxes

[Stainless Steel Rat]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

* François Pinard <pinard@iro.umontreal.ca>  on Sun, 25 Apr 1999
| And then, you get an accepting interaction there?  Something that merely
| flashes in the mini-buffer and disappears at the next random keystroke
| could easily go un-noticed.

I get an error message, just like any other error condition in Emacs, the
way <insert deity> intended things to be.

| Also, is the minibuffer guaranteed to extend for holding and displaying
| the full diagnostic?  It is often multi-line.  Is there any required magic
| for it?  (I use the resize-minibuffer mode, but it does not work perfectly
| in all circumstances, at least from my experience.)

Yeah, well, that is a flaw in FSF Emacs.  XEmacs minibuffer resizing
appears to be much saner.

| I'm surely open to any solution, as long as it does work dependably! :-)

Get used to pretty, gooey dialog boxes.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v0.9.5 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE3I78Sgl+vIlSVSNkRAtPNAJ0R6ZrSVqQ6mjuO4LbrVgTZrGUSuwCgw4Ky
ytASnTN05H2nwgkfskEf1fo=
=XCSc
-----END PGP SIGNATURE-----

-- 
Rat <ratinox@peorth.gweep.net>    \ Caution: Happy Fun Ball may suddenly
Minion of Nathan - Nathan says Hi! \ accelerate to dangerous speeds.
PGP Key: at a key server near you!  \ 

Re: Those `mailcrypt' "push OK" boxes

[François Pinard]

Stainless Steel Rat <ratinox@peorth.gweep.net> écrit:

> | And then, you get an accepting interaction there?  Something that merely
> | flashes in the mini-buffer and disappears at the next random keystroke
> | could easily go un-noticed.

> I get an error message, just like any other error condition in Emacs, the
> way <insert deity> intended things to be.

Most error messages in Emacs do not require confirmation.  So, if I interpret
your answer, you are saying that you have no required confirmation that
you read the error message?  It might then disappear at the next random
keystroke.

> | Also, is the minibuffer guaranteed to extend for holding and displaying
> | the full diagnostic?  It is often multi-line.  Is there any required magic
> | for it?  (I use the resize-minibuffer mode, but it does not work perfectly
> | in all circumstances, at least from my experience.)

> Yeah, well, that is a flaw in FSF Emacs.  XEmacs minibuffer resizing
> appears to be much saner.

OK, well, I'm using Emacs most of the times, and XEmacs much less often.

> | I'm surely open to any solution, as long as it does work dependably! :-)
> Get used to pretty, gooey dialog boxes.

I think we are making circles, here.  I explained in a previous messages
a few reasons why I do not consider them to be good things.  I suggested
a replacement in which the mouse interaction and the special X-windows for
dialog were getting avoided.  If I remember correctly, you then did suggest
to just use the mini-buffer instead.  After further verification with you,
it seems using the mini-buffer does not really solve any of the problems.

The idea is not to push forward solutions just so they oppose each other,
this is not productive.  We should work towards ideas which improve things.
Let us mere <humans> try to do better than any <deity> out there! :-)

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: Those `mailcrypt' "push OK" boxes

[Stainless Steel Rat]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

* François Pinard <pinard@iro.umontreal.ca>  on Sun, 25 Apr 1999
| Most error messages in Emacs do not require confirmation.

None of mailcrypt's messages require confirmation.  The reason for the 'OK'
buttons is because of the nature of dialog boxes, not because of any of
mailcrypt's requirements.

| So, if I interpret your answer, you are saying that you have no required
| confirmation that you read the error message?

Like I said: just like any other error message in Emacs.  If mailcrypt
fails, it is because the underlying PGP or GPG process failed.  Mailcrypt
tells you that -- and the buffer being worked on will be obvious in its
lack of being processed.

| It might then disappear at the next random keystroke.

Whoopie.  Fact is, if mailcrypt fails to encrypt or decrypt a buffer, it is
so obvious that no 'random keystrokes' will make any difference.  If that
were not the case, if the failure were subtle or inobvious, sure, I agree
that acknowledgement of some sort would be a Good Thing.  But not for this,
especially when n-1 times out of n (where n is a large number) it is
nothing more than an obnoxious inconveniece for the sake of the 1 time out
of n the human drops something on the keyboard or fat-fingers or the like.

I suggest trying an Emacs built without dialog boxes and see for yourself.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v0.9.5 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQE3I8zqgl+vIlSVSNkRAknpAJ92D11sgvkmH7gIUSnh82becldY+QCgmL/1
Kb+hwLnGHiauQFNkZjzYtzk=
=9ZjT
-----END PGP SIGNATURE-----

-- 
Rat <ratinox@peorth.gweep.net>    \ Warning: pregnant women, the elderly, and
Minion of Nathan - Nathan says Hi! \ children under 10 should avoid prolonged
PGP Key: at a key server near you!  \ exposure to Happy Fun Ball.

Re: Those `mailcrypt' "push OK" boxes

[François Pinard]

Stainless Steel Rat <ratinox@peorth.gweep.net> écrit:

> | So, if I interpret your answer, you are saying that you have no required
> | confirmation that you read the error message?  [...]  It might then
> | disappear at the next random keystroke.

> Whoopie.  Fact is, if mailcrypt fails to encrypt or decrypt a buffer, it is
> so obvious that no 'random keystrokes' will make any difference.

For encryption, I agree it will be obvious.  For a failed signature
verification, or for untrusted keys, will it be that obvious as well?

> I suggest trying an Emacs built without dialog boxes and see for yourself.

I would rather not go to the effort of rebuilding Emacs just to `see',
unless I have some good hint it will improve the situation.  I did suggest
a work around to ease the problems I saw.  In what way using the mini-buffer
would give me something better?

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard