Re: Common Gateway Interface

>>>>>> "SL" == "Scott Lawrence" <lawrence@agranat.com> writes:
>>>>>> "DB" == "Dylan Barrell" <dbarrell@hotmail.com> writes:
>
>DB> If we beleive that standards are a good thing (and I certainly
>DB> hope we do) then we should also believe in standards which will
>DB> standardise the interoperability of HTTP daemons and
>DB> applications. It will benefit the industry.
>
SL>  Perhaps so, but it really isn't part of the protocol between servers
SL>  and clients.

That depends on how you define the server. If the server constitutes all the 
pieces which integrate with the daemon then this does form part of the protocol.



SL>  There are organizations that focus on API definitions;
SL>  POSIX being the best example that comes to mind.

These organisations are not active in this arena. IETF and W3C have the power 
and the momentum to standardise here and should utilise it to best effect.

>
>DB> Ultimately I would like to see the interface between daemons and
>DB> applications standardised [...]
>
SL>  HTTP servers operate in a wide variety of environments; some are
SL>  the sort of daemons you refer too, others are tightly integrated
SL>  into database products, specialized applications, or special purpose
SL>  hardware such as printers, routers, and coffepots.
SL>
SL>  Getting the protocol agreed upon is quite difficult enough...

I agree that the protocol is difficult enough but I also think that some simple 
statement of the form of "All header information must be made available to all 
executable elements on the server side" with acceptable definitions for the 
terminology would suffice for this body. Let the other bodies worry about the 
way these elements are made available to the various elements be they CORBA 
objects, JavaBeans, Perl scripts, DLLs or anything else.

Cheers
Dylan


---------------------------------------------------------
Get Your *Web-Based* Free Email at http://www.hotmail.com
---------------------------------------------------------

Received on Friday, 30 May 1997 09:03:17 UTC