[3141] in Release_7.7_team

home help back first fref pref prev next nref lref last post

Re: Need to know if migration to Netscape 6 is feasible.

daemon@ATHENA.MIT.EDU (t. belton)
Thu Feb 14 14:04:34 2002

Date: Thu, 14 Feb 2002 14:03:32 -0500 (EST)
From: "t. belton" <tbelton@MIT.EDU>
To: Bill Cattey <wdc@mit.edu>
cc: <release-team@mit.edu>, <owls@mit.edu>
In-Reply-To: <QwOOkqNz0001Jgrp4J@mit.edu>
Message-ID: <Pine.GSO.4.30L.0202141348340.15339-100000@iphigenia.mit.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

Aaron has already noted some of this:

- There is a 6.x for Linux. It depends on certain libraries which may or
may not sit well with the way we do things; I have not checked for
conflicts yet. We have test copies running, so presumably it can be done.

- There is a 6.x for Solaris which is kept and maintained by Sun. I have
installed it and played with it. Moving this to the Athena methodology
will be a pain because Sun assumes you are installing single-user and can
play with your local system packages as necessary. But it can be done.
There are library complications here as well, which is why you need to
mess with so many Sun packages to begin with.

- There is no 6.x for Irix that I know of.

In general 6.1 depends on a particular graphics approach to *nix that
Netscape didn't use before. This is why I keep stressing the libraries.
It's been many months since I had the details in short-term memory, but I
remember poking around and thinking, "Oh, dear, this could get messy."

Netscape 6 has not been very well received by the *nix community here in
its limited tests. The consensus seems to be that it is is ugly, unwieldy,
and slow. There, are, however, two compelling reasons for going through
the pain of the upgrade from 4.x:

1. For compatibility with stupid plugins and sites which look at the
browser's self-identification string, say, "Sorry, I don't play with 4.x,"
and shut down. These are increasing every day.

2. If 6.x is substantially less buggy than 4.x. I have no information save
hearsay on this.

But ... if the Mozilla/Alternate Browser idea has been temporarily
kiboshed, then we may have to look into installing 6.x just to keep people
quiet. We could probably get it all up and running in infoagents in a
fairly quick amount of time - barring Irix, of course, and with some heavy
quota increases because 6.x makes even the recent 4.x binary sizes look
tiny.

Bill, I had no idea people were automatically assuming that *nix/Athena
was going to upgrade to 6.x eventually. I have, in fact, been staying out
of the 6.x upgrade discussions on the other platforms, working on the idea
that 6.x had been tried for *nix, found wanting, and that we were looking
for something better.

- Todd


home help back first fref pref prev next nref lref last post