archie wais server - WWW access directly to files

Brewster Kahle <brewster@think.com>
Date: Tue, 18 Feb 92 05:58:16 PST
Message-id: <9202181358.AA28643@quake.think.com>
From: Brewster Kahle <brewster@think.com>
Sender: brewster@quake.think.com
To: timbl@nxoc01.cern.ch
Cc: kean@talon.ucs.orst.edu, wais-talk@think.com, www-talk@nxoc01.cern.ch
In-reply-to: Tim Berners-Lee's message of Tue, 18 Feb 92 14:25:04 GMT+0100 <9202181325.AA19967@ nxoc01.cern.ch >
Subject: archie wais server - WWW access directly to files
   Date: Tue, 18 Feb 92 14:25:04 GMT+0100
   From: timbl@nxoc01.cern.ch (Tim Berners-Lee)

   ...
   Magic -- I really like it.  I have modified the WAIS-WWW gateway so that is the  
   database is an archie one (i.e.database name contains "archie" :-| ) a W3 search of  
   the database will give links directly to the files.  This relies on the headline  
   starting with host:/filename.

I wish we used /filename@hostname so that waisretrieve could handle it (and
be compatible with the WAIS doc-id.

   Now the things which need neatening up. The gateway I was using is here in CERN,  
   and the server is at ordst.edu. This implies a transatlantic link for every search.

   - Would you or anyone like to run the gateway locally? Volunteers?
WAIS packets are small, but the distance makes reduces reliability.

   - Perhaps we should run the WAIS archie here, mirroring files.

   - One of us could run a direct W3 archie server rather than using a gateway
	process.

   To use www as an archie client,

	   setenv  WWW_wais_GATE  http://info.cern.ch:8001/

   (or point to a more local gateway!) and 


	   alias archie  www wais://archive.orst.edu:9000/archie-orst.edu

   I've put a pointer to the archie index in our CERN home page, but not publicised it  
   yet.

   Tim BL

This is really cool.

-brewster