Xref: utzoo news.groups:33323 news.software.b:8522
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!zephyr.ens.tek.com!gvgpsa!gold.gvg.tek.com!shaunc
From: shaunc@gold.gvg.tek.com (Shaun Case)
Newsgroups: news.groups,news.software.b
Subject: Re: Hypertext style context for news articles: example
Message-ID: <2676@gold.gvg.tek.com>
Date: 30 Jun 91 00:40:09 GMT
References: <1991Jun28.002447.19000@zorch.SF-Bay.ORG> <1991Jun29.012659.12903@unlinfo.unl.edu> <1991Jun29.104431.20366@zorch.SF-Bay.ORG>
Followup-To: news.groups
Organization: Grass Valley Group, Grass Valley, CA
Lines: 26

Something else to ponder is how to handle nested quotes.
That makes expiry a bit more interesting as a programming
problem.


I like the idea of expiring all non-referenced parts
of an article, and moving the referenced parts into
a single large indexed file.  (Large is probably equal
to "real big.")  As long as it is indexed, each section
could be compressed using something simple like backpointers
or Huffman, or even mappings to /usr/dict/words (if you have
it.)  This would make following links slower, but that's ok,
since it would probably not be the common case (IMHO.)

You couldn't grep expired articles, but you can't now,
either.  grepping new articles would be faster, due to
reduced size.




-- 
shaunc@gold.gvg.tek.com  
-- 100,000, perhaps 200,000 or more Iraqis died in a "Turkey Shoot"
   inappropriately called a "war."  -- Michael Albert
The above work is in the public domain, unless it is a piece of email.