W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > January to April 1996

Re: Compression of HTML documents over HTTP

From: Mirsad Todorovac <tm@rasips2.rasip.etf.hr>
Date: Tue, 5 Mar 1996 18:31:51 +0100 (MET)
Message-Id: <199603051731.SAA19029@rasips2.rasip.etf.hr>
To: Jeremy_Cath@reading.sterling.com
Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
> 

I'm not [the prime] authority on this subject, but some browsers handle
compress'd and gzip'ed content well.  However, they are not well-behaved
enough to indicate whether they do or not (concerning Mozilla and NCSA Mosaic),
so the server doesn't know about that.

It seems that Accept-Encoding: and Content-Encoding: serve this purpose well,
the browsers just do not seem to send them (although Content-Encoding is
recognized in browsers above, even Netscape2.0 proved not to send
Accept-Encoding, but IMHO this issue may better be discussed through
www-talk [in HTTP/1.0 draft it's covered well enough, just the current practice
stays behind]).

> Just a thought, but looking at my server there are a number of documents which if I PKZIP them, reduce in side anywhere from 20-60%
> As bandwidth becomes more of a premium, and response times more critical as the number of users increase would it be appropriate to implement some form of compression of the raw text in the .HTM / .HTML documents ?
> 
> Obviously this would require support at two places - compression (on the fly - as some documents need to refresh frequently, or cached) at the server, and at the browser.
> As long as the (de)compression algorithm was fast enought to reduce the overall transmission time there are benefits.

gzip/ungzip proved to be fast enough to speed up transfer, even when fetching
from local server (from the same host).  on-the-fly-compression overhead on
server side may be considerable enough to advise caching of compressed files
(on server), but again we go off the HTTP issues.

[1]	T. Berners-Lee, R. T. Fielding, H. Frystyk Nielsen,
	"Hypertext Transfer Protocol -- HTTP/1.0", 02/19/1996.  URL :
	http://www.rasip.fer.hr/cgi-bin/draft/draft-ietf-http-v10-spec-05.txt
	(or your nearest site), chapter 10.3, D.2.3

> 
> Jeremy E Cath
> Jeremy_Cath@Sterling.Com
> http://193.130.124.60/rufan-redi
> 

-- 
    | Mirsad Todorovac					|
    | Faculty of Electrical Engineering and Computing	|
    | University of Zagreb				|
    | Unska 3, Zagreb, Croatia 10000			|
    |							|
    | e-mail: mirsad.todorovac@fer.hr			|
Received on Tuesday, 5 March 1996 09:40:52 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:48 EDT