From: dlv@bwalk.dm.com (Dr. Dimitri Vulis)
Date: Wed, 6 Dec 95 15:29:40 PST
To: cypherpunks@toad.com
Subject: Re: Note on "Barring Netscape"
In-Reply-To: <9512061547.ZM8340@glacius.alias.com>
Message-ID: <yJJPFD9w165w@bwalk.dm.com>
MIME-Version: 1.0
Content-Type: text/plain


"Richard Martin" <rmartin@aw.sgi.com> writes:
> On Dec 6, 11:57am, Richard Charles Graves wrote:
> > The Microsoft Internet Explorer sends the user-agent "Mozilla 1.22
> (compatible"
> > to hte server, which triggers an incorrect response from, e.g., www.c2.org.
> > This little fraud has the potential to make you look silly if left
> unexplained.
>
> this *is* fraud, in a way. Microsoft is shipping a product which in a
> certain exchange claims to be a product of another company. Microsoft's
> software is being treated better around the net because it is recognised
> as Netscape, which it isn't.

This reminds me of how many many years ago, when IBM and Microsoft were
good friends, earlier version of MS Windows(?) video drivers were
hard-coded not to recognize a video card as being VGA-compatible (or
even EGA-?) unless its ROM had the 3 letters "IBM" at offset 0x1e. The
genuine vanilla IBM card had the words "COPYRIGHT IBM" at that address.
Various clone makers had to put "IBM" at that location to make their
cards work with Windows(?). E.g., I have a Trident VGA card whose ROM
says "RESERVED FOR IBM COMPATIBILITY" positioned so that "IBM" is
exactly at 0x1E. Some other cards just say "IBM" there with no
explanation in adjascent memory.

I think that if some servers refuse to talk to clients unless they see
"Mozilla", then I can't blame Microsoft for impersonating Netscape.

I think the whole protocol is way stupid. Instead of asking the client
for its name, and then looking up in some database what a client with
that name is capable of, a server should ask the client only about the
capabilities that the server is planning to use.

---

Dr. Dimitri Vulis
Brighton Beach Boardwalk BBS, Forest Hills, N.Y.: +1-718-261-2013, 14.4Kbps