Re: Status: -> Progress:

Rick Troth (
Wed, 11 May 1994 11:52:25 -0500 (CDT)

> Let me pitch an idea that would mean something to a new HTTP.
> ...
> Imagine a new httpd that doesn't write to a "filesystem", rather
> was written in the OO versions of tcl, perl(5?), or python.

This is already almost possible without a complete rewrite.
This is why I suggested a "parameters" syntax

GET /path/path/path/object,parms?args [HTTP/1.0]

(whether that delimiter is a semi-colon or a comma,
I don't care; maybe it could even be an ampersand) The "args"
are dynamic, while the "parms" are more (but not totally) static.
The reason I mention it in this context is that it's specifically
intended for *program* objects, not ordinary files.

> Attributes such as logging information, discretionary access, meta
> information such as indices and document owner would be object
> attributes in a "document class". ...
> Thus, could HTTP be extened to send any arbitrary "message" to
> the "document" object? This perl5 or tcl-dp httpd would
> recompose the document to send to the client, thus allowing
> conformance with the stream it should be sending, ...

This is really good stuff, Paul. But again, I don't think
we have to completely reinvent HTTPD to accomplish it. I think we're
almost there with some of the latest juicy CGI ideas. Any "object"
that's a program instead of a file can accomplish what you're after.
Just be vewy vewy kay-uh-ful about the security implications.

> /ramble off
> --Paul

Rick Troth <>, Rice University, Information Systems