[638] in athena10
Re: set skip_foo; window managers
daemon@ATHENA.MIT.EDU (William Cattey)
Tue Dec 2 14:36:37 2008
In-Reply-To: <ACDEC213-34DC-4DE6-A1E5-18F0E43568BD@mit.edu>
Mime-Version: 1.0 (Apple Message framework v753.1)
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <211EB2B7-F7A3-4056-A0A0-7DEF1A2204F4@mit.edu>
Cc: Jonathan Reed <jdreed@mit.edu>
Content-Transfer-Encoding: 7bit
From: William Cattey <wdc@MIT.EDU>
Date: Tue, 2 Dec 2008 14:34:33 -0500
To: athena10@mit.edu
The current Athena 10 implementation of xsession.tcsh and
xsession.bash seems to support:
skip_x_startup
skip_quotawarn
skip_authwatch
skip_lert
With regards to skip_xlogout and skip_dash, my thinking was that
we've completely de-supported dash, and so these appropriately go away.
One more skip_ foo, not already mentioned:
skip_panel
Like skip_capplets, works around funky window manager decorating:
mwm decorates nautilus.
wm decorates nautilus, panel, and the background window.
I also discussed this situation with geofft yesterday. Back then I
was still thinking that supporting the WINDOW_MANAGER variable was
ok, because I thought twm, and mwm might be reasonably popular, and
that they behaved reasonably well under modern gdm.
Today I see that mwm and twm never fully caught up with common
conventions. At the moment I'm inclined to take the position:
If you want your own window manager, you should bite the bullet and
use a custom
.xsession file.
That position further reduces the scope of the stuff we end up
maintaining and testing, and nudges us further in the direction of,
"step out of the way and let the upstream defaults operate."
-Bill
----
Important: IS&T IT staff will *NEVER* ask you for your password, nor
will MIT send you email requesting your password information. Please
continue to ignore any email messages that claim to require you to
provide such information.
----
William Cattey
Linux Platform Coordinator
MIT Information Services & Technology
N42-040M, 617-253-0140, wdc@mit.edu
http://web.mit.edu/wdc/www/
On Dec 2, 2008, at 2:31 PM, Jonathan Reed wrote:
> There's also skip_dash and skip_xlogout. I can't remember what we
> decided about Dash - are we completely de-supporting it? If so, we
> should probably make some outreach to current Dash users. I will
> attempt to crawl AFS and get a list of people who
> have .athena_dash_interface files.
>
> Unless someone already did this and has such a list.
>
> -Jon
>
>
> On Dec 2, 2008, at 2:02 PM, Jonathan Reed wrote:
>
>> Bill and I were discussing the various "set skip_foo" things we
>> tell people to set, and what we should do with them in Athena 10.
>> I see two possible approaches. The first, a slash-and-burn
>> method, is to desupport all of them, and say "If skipping this one
>> feature is important to you, then go write your own .xsession".
>>
>> The second approach would be the following:
>>
>> Desupport:
>> skip_capplets: does this even work anymore?
>> skip_quotawarn: Quota is sufficiently large these days that I
>> don't think we care
>> skip_lert: There are plenty of ways to tell lert to be quiet, why
>> do we need to support this?
>> skip_athneteventd: Will we even be using this in Athena 10?
>>
>> Support:
>> skip_xss
>> skip_tty_startup
>> skip_x_startup
>>
>> Unsure:
>> skip_authwatch: Do people use this?
>> skip_initial_xterm: Is this really a big deal?
>> skip_nautilus: This is the big one. The primary reason, IIRC,
>> for having this was that nautilus failed miserably with window
>> managers that were not Sawfish/Metacity. It still does. However,
>> nautilus will be more useful under Athena 10, now that things like
>> flash drives and CDs will "just work". So arguably, I say we punt
>> it, and if you want to run a window manager that doesn't play nice
>> with nautilus, you get to figure it out on your own.
>>
>> That also brings us to the question of how we want to deal with
>> window managers. Neither twm nor mwm are installed by default
>> under Athena 10, and they're not in the windowmangers locker
>> because they were always in the release. Should we install twm
>> and motif-clients for debathena-workstation and debathena-
>> cluster? Or, should we desupport the WINDOW_MANAGER variable
>> entirely, and people who want a custom window manager get to run
>> their own xsession?
>>
>> Thoughts?
>>
>> -Jon
>>
>>
>
