[19886] in Athena Bugs
Re: netscape and sawfish continue downward spiral of death
daemon@ATHENA.MIT.EDU (John Hawkinson)
Thu Oct 11 10:31:42 2001
Date: Thu, 11 Oct 2001 10:31:40 -0400
From: John Hawkinson <jhawk@MIT.EDU>
To: "t. belton" <tbelton@MIT.EDU>
Cc: bugs@MIT.EDU
Message-ID: <20011011103140.V23491@multics.mit.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Pine.GSO.4.30L.0110110947110.12668-100000@iphigenia.mit.edu>
t. belton <tbelton@MIT.EDU> wrote on Thu, 11 Oct 2001
at 09:53:59 -0400 in <Pine.GSO.4.30L.0110110947110.12668-100000@iphigenia.mit.edu>:
> > Maybe it would be interested to have a wrapper around netscape
> > that let us know how many netscapes that were cleanly started were
> > not cleanly exitted. Since there seems to be precedent for such tracking
> > in 3partysw land.
>
> It'd slow things down, wouldn't it?
I don't think it would slow things down appreciably, no.
If we "wrapped" netscape there would be a slowdown, but if we just
ran the logging program that's part of slw (or a variant), it
wouldn't be any slower than the halfadozen other things we do in the
netscape startup script.
> And it'd just tell us that Netscape doesn't exit cleanly fifty
> percent of the time (or thereabouts). At least that'd be my
> estimate, based on anecdotal evidence from the MIT user community
> over three years.
If we aren't going to do anything with the data, then obviously
it's useless to collect it. And likewise, if it doesn't seem
that the effort of collecting it is worth anyone's time, the same.
It occurs to me that it might tell us:
Are there platform dependancies?
Does it happen more often with 9.0 than 8.4 (well, 9.1 versus 9.0,
since we cannot get 8.4 data)?
Does it happen to some users more than others (if a single user
can easily reproduce the problem, or it happens to them all the
time, then there's a chance of debugging the problem)?
Is it time-of-day or network-load related?
But I've always been a fan of having piles and piles of data, even if
you never know when you're going to use it, and I understand that
other people's predilictions may not be in line with my biases. See
above with respect to "worth anyone's time."
> I'd rather focus on getting away from Netscape as quickly as possible.
> Unfortunately none of the alternatives is QUITE ready for prime time yet.
> But that time is getting closer.
This strikes me as pie-in-the-sky. Is there any concrete reason to believe
we will be using anything other than Netscape as the supported browser
within 18 months? If not, then I don't think it's acceptable to simply
defer problems.
Do you have any serious on-the-horizon candidates?
--jhawk
