Re: [apps-discuss] request for review: draft-bvandervalk-sadi-00.txt (individual submission)

[Graham Klyne <[e-mail]>]

I skimmed through this with some interest, but I feel (with some regret) that I 
must agree with Tim Bray that the IETF is probably not the right venue for this. 
  I have two main technical reasons for saying this:

(1) it does not really define any new protocol in the sense that IETF generally 
defines protocols; rather it defines a (quite reasonable) pattern of use for 
HTTP for a particular class of web services.

(2) the "protocol" described is quite specific in the content types it deals 
with (just RDF), which I feel means it is not the kind of common 
infrastructure-layer element that the IETF would normally consider.

None of this means that the work is without merit, or not worthy of wider 
review.  In particular, I note that it may find early applicability outside 
bioinformatics in areas such as open government data applications where there is 
significant use of RDF.

Suggestions I offer to present this to a wider audience are:
- I would expect that a proposal like this would likely gain more traction in 
W3C.  In the first instance, there might be a possibility to see if the W3C HCLS 
group would be interested to review this and publish it as a W3C Note.
- If you feel it really has applicability to communities beyond those who engage 
with W3C activities, it might be a candidate for individual submission as an 
informational RFC; though there no guarantee that it will get extensive review 
on that route.  (There is a "precedent" in BEEP (formerly BXXP) for 
specifications being first published by that route, and later becoming formally 
standardized as http://www.ietf.org/rfc/rfc3080.txt - though I don't know that 
it's widely used).

There may well be other, better ways to develop this work - those are just a 
couple of suggestions that occur to me.

Best,

#g
--

On 23/02/2012 01:37, Ben Vandervalk wrote:
> Hi all,
>
> I have recently submitted an I-D (as an individual) regarding RDF/OWL-based
> web services called SADI (Semantic Automated Discovery and Integration).
>
> My intent is to put this I-D on the standards track, and I have been
> advised that the first step is to solicit comments/reviews on this list, in
> order to guage interest.
>
> There are currently three implementations of SADI in Java, Perl, and
> Python, which are available from http://sadi.googlecode.com/. The Java and
> Perl implementations were created by the authors, whereas the Python
> implementation was created independently by Jim McCusker of MSU. The Python
> implementation does not yet implement all of the features in the draft, but
> we think it is likely that Jim will expand his implementation in time, or
> that other parties will create additional implementations of SADI.
>
> I am unfamiliar with IETF procedures regarding individual submissions and
> would appreciate any advice in that regard as well.
>
> Thanks for your consideration!
>
> ======
>
> A New Internet-Draft is available from the on-line Internet-Drafts
> directories. This is an Individual Submission.
>
>         Title : SADI: Semantic Automated Discovery and Integration
>         Author(s) : Ben Vandervalk, E. Luke McCarthy, Mark D. Wilkinson
>         Filename : draft-bvandervalk-sadi-00.txt
>         Pages : 31
>         Date : 2012-01-31
>
> This document describes Semantic Automated Discovery and Integration
> (SADI), a set of best practices for implementing stateless web
> services that consume RDF data as input and generate RDF data as
> output. The goal of SADI is to establish conventions that will
> enable a much higher level of interoperability between web services
> from independent providers than is currently possible under the
> widespread use of WSDL/XML and RESTful services. Under SADI,
> interoperability depends on the shared use of predicate vocabularies,
> rather than the shared use of particular XML schemas, JSON
> structures, or ad hoc data formats. Through the use of OWL to
> describe service input and output datatypes, SADI enables: i)
> automated discovery of services that provide data or computations of
> interest, and ii) automated matchmaking between local data and
> available services. By iterative application of the former two
> capabilities, SADI enables semi-automated construction of arbitrarily
> complex workflows across independent service providers.
>
> A URL for this Internet-Draft is:
> http://www.ietf.org/internet-drafts/draft-bvandervalk-sadi-00.txt
>
>
>
> -- Ben Vandervalk
> Wilkinson Lab, University of British Columbia
>
>
>
>
> _______________________________________________
> apps-discuss mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/apps-discuss