[comp.windows.x] talk, beeps

bob@MorningStar.Com (Bob Sutterfield) (12/14/89)

In article <8912131434.AA19336@crdgw1.ge.com> JARMOLOWSKI@esdsdf.dnet.ge.COM writes:
   If not, can anyone tell me where I can find the protocol used by
   the talk deamon so that I can write my own?

The protocol isn't documented - "Use the Source, Luke!"  See
uunet.uu.net:bsd-sources/include/protocols/talkd.h and
bsd-sources/src/etc/talkd.tar.Z.

   I generally keep my console window iconified and often fail to
   notice when someone wants to talk to me.  There are already too
   many applications that beep at me besides "talk".

Hmmm, perhaps we need an X protocol extension for stereo workstation
speakers?  Or for quadrophonic, so we could locate the beep in the
plane of the window?  Or maybe DTMF (i.e. Touch-Tone(tm)) encodings,
controllable from the resource database?

(only .5 :-)

rodney@dali.ipl.rpi.edu (Rodney Peck II) (12/14/89)

>>>>> On 13 Dec 89 19:29:40 GMT, bob@MorningStar.Com (Bob Sutterfield) said:

Bob>    I generally keep my console window iconified and often fail to
Bob>    notice when someone wants to talk to me.  There are already too
Bob>    many applications that beep at me besides "talk".

Bob> Hmmm, perhaps we need an X protocol extension for stereo workstation
Bob> speakers?  Or for quadrophonic, so we could locate the beep in the
Bob> plane of the window?  Or maybe DTMF (i.e. Touch-Tone(tm)) encodings,
Bob> controllable from the resource database?

Yeah, all the sparc-station 1's can make sounds and play digitized things.
I think an X extension for the larger type (stereo, high bandwidth, etc.

neat.
--
Rodney

dpg@citi.umich.edu (David Gorgen) (12/14/89)

In article <BOB.89Dec13142940@volitans.MorningStar.Com> bob@MorningStar.Com (Bob Sutterfield) writes:
>    I generally keep my console window iconified and often fail to
>    notice when someone wants to talk to me.  There are already too
>    many applications that beep at me besides "talk".
> 
> Hmmm, perhaps we need an X protocol extension for stereo workstation
> speakers?  Or for quadrophonic, so we could locate the beep in the
> plane of the window?  Or maybe DTMF (i.e. Touch-Tone(tm)) encodings,
> controllable from the resource database?
> 
> (only .5 :-)

My Master's thesis, done in 1977 at MIT's Architecture Machine Group,
involved getting sound images to emanate from any particular spot on
a 6x8 foot rear-projection video screen in a "Media Room", using four
speakers mounted at the corners of the screen.

One of my conclusions was that this "quadraphonic" technique cannot
deliver good enough cues to allow people to localize the sound very
well, unless a visual reinforcing cue is also provided.  Headphones
and some fancy signal processing might do much better, but have the
disadvantage that the sound image would be relative to your head
position, not to the outside world.  They are also physically awkward.

You could have the icon flash while beeping, but this is really a
regression to a "visual bell"; these are not very popular.

The people at the Media Lab (the descendant of the Architecture
Machine Group) would probably tell you today that the "talk" icon
should actually TALK to you, telling you to look at it.  This would
mean we should be talking about a speech output extension to X.

Have we gone far enough overboard yet?

--------------------------------------------------------------------------
Dave Gorgen / GTD-East (formerly Apollo Computer), Hewlett-Packard Company
located at:  University of Michigan, CITI          dpg@citi.umich.edu
(Center for Information Technology Integration)    313-998-7482 or -7479
--
Dave Gorgen / GTD-East (formerly Apollo Computer), Hewlett-Packard Company
located at:  University of Michigan, CITI          dpg@citi.umich.edu
(Center for Information Technology Integration)    313-998-7482 or -7479

ral@ARISTOTLE-GW.JPL.NASA.GOV (Roger Lighty) (12/14/89)

References: <1989Dec13.221405.3283@terminator.cc.umich.edu>
            <8912131434.AA19336@crdgw1.ge.com>
            <BOB.89Dec13142940@volitans.MorningStar.Com>

In in responing to article <BOB.89Dec13142940@volitans.MorningStar.Com> 
     bob@MorningStar.Com (Bob Sutterfield) writing:
>>    I generally keep my console window iconified and often fail to
>>    notice when someone wants to talk to me.  There are already too
>>    many applications that beep at me besides "talk".
>> 
>> Hmmm, perhaps we need an X protocol extension for stereo workstation
>> speakers? ...

Dave Gorgen / GTD-East (formerly Apollo Computer), Hewlett-Packard Company
answers:
 ...

> The people at the Media Lab (the descendant of the Architecture
> Machine Group) would probably tell you today that the "talk" icon
> should actually TALK to you, telling you to look at it.  This would
> mean we should be talking about a speech output extension to X.
>
> Have we gone far enough overboard yet?

Not quite!  Take a look at SERPENT.  (There will be a tutorial at the
X technical conference):

	TUTORIAL #7: The Serpent UIMS
	Len Bass, Erik Hardy, Rick Kazman, Dan Klein
	Software Engineering Institute, Carnegie Mellon University

Serpent has the hooks for X and sound.

-- roger

casey@gauss.llnl.gov (Casey Leedom) (12/14/89)

  Sure.  Why not.  I should follow this one up ...  I mean, what is the
infinite net bandwidth for but to provide an infinite number of $.02
slots ... :-)

| From: dpg@citi.umich.edu (David Gorgen)
| 
| You could have the icon flash while beeping, but this is really a
| regression to a "visual bell"; these are not very popular.

  I've actually asked that it be possible to have icons reverse when
associated windows change state (output, etc.) so you could tell that
something had happened to a window since it was iconified.  Jim Fulton
mumbled something about ``sounds like a nice idea, but it would depend on
mumble mumble glech foo-bar *sounds of eagles eating chetos* *plaid smoke
rising from primitive fires* *reams of envelope backs whose reverse sides
were napkins* ...'' (I.e.  he got beyond me real quick ...  :-))

Casey

envbvs@epb2.lbl.gov (Brian V. Smith) (12/15/89)

In article <41512@lll-winken.LLNL.GOV>, casey@gauss.llnl.gov (Casey
Leedom) writes:
>   I've actually asked that it be possible to have icons reverse when
> associated windows change state (output, etc.) so you could tell that
> something had happened to a window since it was iconified.  Jim Fulton
> mumbled something about ``sounds like a nice idea, but it would depend on
> mumble mumble glech foo-bar *sounds of eagles eating chetos* *plaid smoke
> rising from primitive fires* *reams of envelope backs whose reverse sides
> were napkins* ...'' (I.e.  he got beyond me real quick ...  :-))

A version of xterm in X10R4 had an option for icons that would 
change the icon when output had gone to the window while it was iconified.

I think, but am not positive that Ed Moy was the persion who made this
change (and several others) to xterm to do this.

Is this possible under X11, or would it require a change in the protocol
to do it?
_____________________________________
Brian V. Smith    (bvsmith@lbl.gov)
Lawrence Berkeley Laboratory
I don't speak for LBL, these non-opinions are all mine.

bob@MorningStar.Com (Bob Sutterfield) (12/15/89)

In article <41512@lll-winken.LLNL.GOV> casey@gauss.llnl.gov (Casey Leedom) writes:
      From: dpg@citi.umich.edu (David Gorgen)

      You could have the icon flash while beeping, but this is really
      a regression to a "visual bell"; these are not very popular.

   I've actually asked that it be possible to have icons reverse when
   associated windows change state (output, etc.) so you could tell
   that something had happened to a window since it was iconified.

There was a user at OSU CIS who steadfastly refused to move from X10
to X11.  Transition difficulty?  Not really, he is one of the better
programmers/systems folks I know, adept at environmental flexibility.
Performance problems?  Not really, he has an extended-memory Sun-3/50
and X11's performance is at least survivable.  Compatibility?  No, by
now more machines there run X11 than X10.  Features/toys?  No, all the
important desk accessories (post-its, calendars, etc.)  have
analogies, if not direct ports, and there are lots of new extras
besides.  And he acknowledges that the resource database and improved
window manager functionality are big wins.

He won't move to X11 until xterm has active icons once again.  You
remember the kind - when an X10 xterm is closed, its icon is a
miniature version of the window with 1-pixel-high fonts.  It lets you
keep track of what's happening in a closed window, really very handy.
I weaned myself of them soon enough after R1 came out, but some folks
find that feature too attractive to give up.

Oh well...

root@johnbob.uucp (12/15/89)

In article <4463@helios.ee.lbl.gov> envbvs@epb2.lbl.gov (Brian V. Smith) writes:
...
>A version of xterm in X10R4 had an option for icons that would 
>change the icon when output had gone to the window while it was iconified.
>
...
>Is this possible under X11, or would it require a change in the protocol
 ^^^^^^^^^^^^^^^^
Yes, if the window manager you are using will let you have your own icon window.


>Is this possible under X11, or would it require a change in the protocol
                                ^^^^^^^^^^^^^^^^^^^^^^^^^
I don't think so...   You might have to do something to a window manager if it
won't let you have your own icon window.

>Brian V. Smith    (bvsmith@lbl.gov)
>Lawrence Berkeley Laboratory
>I don't speak for LBL, these non-opinions are all mine.


You could try changing the volume of the bell, ring it more than once, change the pitch
or duration.  (or is XChangeKeyboardControl() a localism?)

-john harvey

John Harvey
AT HOME  ...!sequoia!johnbob!jph  OR
..!uunet!cs.utexas.edu!execu!sequoia!johnbob!jph
AT WORK  john@johnbob  OR
         @cs.utexas.edu:ibmchs!auschs!johnbob.austin.ibm.com!john  OR
..!uunet!cs.utexas.edu!ibmaus!auschs!johnbob.austin.ibm.com!john
I don't speak for anybody.  Not even myself.

swick@ATHENA.MIT.EDU (Ralph R. Swick) (12/15/89)

> He won't move to X11 until xterm has active icons once again.

Sigh.  The long-range consequences of one's hacks are sometimes
frightening, expecially this early in the morning ;-|

rws@EXPO.LCS.MIT.EDU (Bob Scheifler) (12/15/89)

    Not really, he is one of the better
    programmers/systems folks I know, adept at environmental flexibility.

Then he should have been adept at putting an active icon into xterm by now,
and feeding the sources back to MIT.  Come on ...

jim@EXPO.LCS.MIT.EDU (Jim Fulton) (12/15/89)

	He won't move to X11 until xterm has active icons once again.

R4 will support changing the font on-the-fly and will have a new menu just for
this purpose.  One of the entries will be "tiny" for nil2....  Oughta make 
active icon windows relatively easy to implement although window managers
are allowed to ignore them (which is why the R4 method is more useful; instead
of iconifying, just bind the set-font action to function key and use that).

toml@ninja.Solbourne.COM (Tom LaStrange) (12/15/89)

>>A version of xterm in X10R4 had an option for icons that would 
>>change the icon when output had gone to the window while it was iconified.
>>
>>Is this possible under X11, or would it require a change in the protocol
>^^^^^^^^^^^^^^^^
>Yes, if the window manager you are using will let you have your own icon window
>

xterm would also have to modified to provide such an icon window, it 
currently does not.

--
Tom LaStrange

Solbourne Computer Inc.    ARPA: toml@Solbourne.COM
1900 Pike Rd.              UUCP: ...!{boulder,sun}!stan!toml
Longmont, CO  80501

D. Allen [CGL]) (12/19/89)

Last August I changed our R3 xterm to beep and shove "***" into the
beginning of the icon name of the iconified xterm when output appeared,
so you'd know when something was happening.  Our local X Master sent
these changes to MIT.  Not as nice as the little box that X10 xterm would
draw for me, but much better than nothing.  I also had to fix awm to send
map notify events to the real client as well as to the client frame, so
that xterm would notice.  RTL seems to work fine.  UWM does not.
-- 
-IAN! (Ian! D. Allen) idallen@watcgl.uwaterloo.ca idallen@watcgl.waterloo.edu
 [129.97.128.64]  Computer Graphics Lab/University of Waterloo/Ontario/Canada