Newsgroups: sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!uhog.mit.edu!bloom-beacon.mit.edu!gatech!howland.reston.ans.net!ix.netcom.com!netcom.com!olea
From: olea@netcom.com (Michael Olea)
Subject: Re: Wanna talk GIF??
Message-ID: <oleaD1yEGE.71B@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <3eciaf$h5u@netaxs.com> <usenet-0301951950220001@lowry.eche.ualberta.ca> <TGL.95Jan3204135@netcom19.netcom.com>
Date: Thu, 5 Jan 1995 22:29:02 GMT
Lines: 30

tgl@netcom.com (Tom Lane) writes:

>usenet@lowry.eche.ualberta.ca (Brian Lowry) writes:
>> ... TIFF is, and has
>> always been, superior to GIF, and there's no lack of support for it.
>> [TIFF supports LZW, naturally... as well as a whole slew of bit depths]

>Stop to think half a second, man.  If Unisys succeeds in enforcing their
>patent claim against LZW compression in GIF, whose doorstep will they be
>on next?  That's right, TIFF users.  Better get your head out of the
>clouds and pay attention to what's really going on here.

>I happen to think there are three or four reasons why Unisys' claim is
>invalid and immoral, but no court of law has ever asked me for advice.
>In the current legal climate, they probably have a very good chance of
>winning a suit.

>			regards, tom lane
>			organizer, Independent JPEG Group


	The TIFF 6.0 spec. discourages the use of LZW specificaly because
of the Unisys (and IBM) claims.  It is entirely possible to use TIFF and
not use LZW - the only problem is that no alternative lossless compression
scheme of color data was suggested in the spec.  One of the strengths of
TIFF, in my opinion, is the clear separation of compression algorithm and
file format (though I still hear people say "TIFF compression").  Why not
draw on that strength and define a replacement for LZW in TIFF?  GZIP comes
to mind...

