httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From pit...@cc.gatech.edu (James Pitkow)
Subject Re: for the wish list...
Date Fri, 11 Aug 1995 00:36:41 GMT

> I think he was thinking of server-wide headers rather than document-specific,
> but it should be noted that one can add document-specific headers now using
> the .asis type.  I would recommend keeping this to an absolute minimum, and
> if there's actual metainformation available the client should find it out
> using the URL2URC method on the server as a separate access.  Just 
> thinking out loud.

Yup, I agree.  I prefer to envision this as merely asking an object,
based upon the list of methods it supports, for certain data within
it's address space.  An example of this is meta-information, which has
many dimensions and degrees of complexity.  In this way, the process of
information discovery can be viewed more so as an iterative process and
less as document retrieval.  The use of meta-information outside of the
traditional-ish urc author, publisher, etc. fuels much of our more interesting
research, which also makes navigation ...  So, I tend to feel a bit 
constrained with .asis though recognize that it or some variant will 
most likely actualize sooner than later.  

Jim.

Mime
View raw message