Re: FW: Review request

Hmm, the paper claims;

"We show how to eliminate parts of HTTP/1.1, making it smaller and
easier to implement."

Yet their suggestions are to;

(1) Remove OPTIONS, PUT, DELETE, and TRACE messages
from the standard. The Allow header should also
be removed.
(2) Retain the HEAD message, but clarify its purpose.
(3) Make the Via header mandatory in all messages. This
suggestion is consistent with the HTTP/1.1 errata [16].

Which would perhaps shrink the spec by a couple of pages, but AFAICT,
not make it any easier to implement.  Plus, as Joe observes,
standardizing on what's used, rather than on what works, seems a bit
silly.

I think the paper would have been much better if the work that was
done was recast simply as an examination of current practice with
HTTP.

Also, I felt the SOAP based Web services stuff was out of place.

Mark.

On 8/18/06, Larry Masinter <LMM@acm.org> wrote:
>
> I had trouble writing what I thought about this
> paper, so I thought I would 'forward this link
> to other people that may be interested in this
> topic'.
>
> Larry
>
>
> -----Original Message-----
> From: Paul Adamczyk [mailto:padamczy@uiuc.edu]
> Sent: Wednesday, August 16, 2006 8:22 PM
> To: LMM@acm.org
> Subject: Review request
>
> Dr. Masinter,
>
> I am contacting you, an author of the HTTP/1.1 standard, to request a review
> of my paper, which discusses the relationship between HTTP and Web services.
> I am very interested in your opinion about this topic.
>
> Please visit: pinky.cs.uiuc.edu/~padamczy/docs/http_ws_draft.pdf
>
> Feel free to forward this link to other people that may be interested in
> this topic.
>
> Thank you,
>
>
> Paul Adamczyk
> PhD student
> University of Illinois
>
>
>

Received on Friday, 18 August 2006 18:14:39 UTC