Re: comments on HTTP draft of 5 Nov 93

Tony Sanders <sanders@bsdi.com>
Errors-To: sanders@bsdi.com
Errors-To: sanders@bsdi.com
Message-id: <199312090031.SAA03435@austin.BSDI.COM>
To: Jim Davis <davis@dri.cornell.edu>
Cc: www-talk@nxoc01.cern.ch, Tim Berners-Lee <timbl@www0.cern.ch>
Subject: Re: comments on HTTP draft of 5 Nov 93 
In-Reply-To: Jim Davis's message of Wed, 08 Dec 1993 17:58:27 EST.
Errors-To: sanders@bsdi.com
Reply-To: www-talk@nxoc01.cern.ch
Organization: Berkeley Software Design, Inc.
Date: Wed, 08 Dec 1993 18:31:09 -0600
From: Tony Sanders <sanders@bsdi.com>
> > so the user also knows this fact ... .  Not all objects returned
> > by HTTP are HTML.  This information *MUST* be allowed in the object
> > meta-information (aka the HTTP header) because not all data types allow
> > ISINDEX and ISMAP!
> 
> Oh!  I get it.  In the old days, the browser saw an ISINDEX
bingo.

> If that's the real and true reason, then the RFC should explain that.
It does, just not very well :-)

> But my argument about ISMAP still makes sense doesn't it?  It's
> one thing to "blindly" search a document with text keywords,
> it's another to submit a pair of X Y coordinates.
You aren't searching the document though.  You are using the URL to initate
a search of "something" and that "something" is defined by the server not
the client.  It might be a WAIS database, like:
    http://httptest.bsdi.com/plexus-3.0/plexus.html
or it might be a graphical index into some other data space, etc.

> And I certainly see the need for this information now.  But
> tell me, do the existing browsers check for these methods in
> the header?
Nope, not yet.

> My server certainly is not returning them at present,
> and things still work, but perhaps that's because they are not
> needed when the data is HTML.
Right.   Also note that SPACEJUMP would not be an attribute of the
document, but rather would be returned when you retrieved the
URL that points to the image itself (indicating that the URL can
be used as a spatial index vs. a textual index for TEXTSEARCH).
Thus the names.

Pretty neat eh?

I still agree that there should be another way to indicate the attributes
of an object instead of them being methods.  It's also highly likely that
they should allow options, e.g.,

    WWW-object-attr: SPACEJUMP;dimensions=3, TEXTSEARCH;style="wais"

--sanders