Classification:
Prologue:
Epilogue:
Thanks Angelo, it is good to get input from real developers. However, given
your observation, I still don't know which side of the argument (use http vs.
develop a brand-new protocol) you would come down on.
---------------------- Forwarded by Roger K Debry/Boulder/IBM on 01/06/97 10:54
AM ---------------------------
ipp-owner @ pwg.org
01/06/97 10:35 AM
To: Angelo_Caruso @ wb.xerox.com at internet
cc: ipp @ pwg.org at internet
Subject: Re: IPP> What is it we really need?
Angelo,
Outstanding comments, indeed. As a real (protocol) implementor, I think
some folks may be sweeping a LOT of implementation realities under the
carpet in trying to leverage HTTP support.
...jay
> Since I'm a real (printer) implementor I guess it's time I chimed in. I
> believe there are many among us who are currently involved in web
> based application development. As such, we are predisposed to think
> that developing on top of HTTP is the way to go since we are currently
> more comfortable with it. However, so called "low level sockets
> programming" is not as bad as it is being made out to be. There are
> well defined APIs for TCP level programming on EVERY major OS (what do
> you think web browsers/servers use anyway?). Furthermore, having
> recently implemented an embedded web server, I can tell you that there
> is no such thing as CGI or ASAPI for low end embedded systems. For
> these systems (and I mean real printers here) it's all C code and
> there is no advantage to building on HTTP rather than TCP.
>> Angelo Caruso
> Xerox Corp.