Re: Netscape 4.5 and HTTP/1.1 Accept-Encoding

> From: Chris Newman <Chris.Newman@innosoft.com>
> Date: Tue, 03 Nov 1998 14:44:39 -0800 (PST)
> To: Jim Gettys <jg@pa.dec.com>
> Cc: http-wg@hplb.hpl.hp.com
> Subject: Re: Netscape 4.5 and HTTP/1.1 Accept-Encoding
> -----
> On Tue, 3 Nov 1998, Jim Gettys wrote:
> > compress 	The encoding format produced by the common UNIX file compression
> > 	program ``compress''. This format is an adaptive Lempel-Ziv-Welch coding
> > 	(LZW).
> 
> Yikes!
> 
> If you reference LZW in the HTTP spec, then RFC 2026 section 10.3.2 comes
> into effect. That means you have to document the Unisys patent in the spec
> and the IETF executive director will have to contact Unisys with a request
> for openly specified, reasonable, non-discriminatory licensing terms.
> 
>                 - Chris

I don't think we have a problem here: compress is not required for 
interoperability, is essentially deprecated (by deflate and gzip), and 
it is not a normative reference (beyond saying it is what a program does).

And it is documenting existing practice (i.e. was in even RFC 1945, I think).
			- Jim