Re: R2RML Implementation

Forwarding to the WG

Hi Mykel:
I am glad R2RML is working for you.  What we really need right now is for folks to run the
test suite and tell us how they did.  This will help us get to Recommendation status.

I am sure many in the WG will be happy to hear about your experiences with R2RML and
this will be valuable input when we get to discussing the future of the WG.  Right now, I
suspect most folks are heads down implementing and testing but let me ask the WG and
maybe we can have a call.

Another thing you could do is to write a blog about how you use R2RML.  That publicity
will help all of us.

All the best, Ashok

On 6/8/2012 1:46 PM, Mykel Kulkarni wrote:
> Hi Ashok et al.,
>
> My colleagues and I will be doing a debrief of the conference on Monday and can follow up shortly thereafter. During that, I will certainly mention the use of your test suite, as well as to what extent we can share details on our more recent implementation information on R2RML.
>
> It is hard not to write a book on the details of everything we did pertaining to our use of R2RML over the last few years. We could probably set up a call, or GoTo meeting if you are interested in discussing, and/or seeing a demo of any variety of things we have accomplished using R2RML, from federating across multiple disparate RDB sources, to exposing excel flat files as SPARQL endpoints. If there are certain specifics you are interested in, please let me know, and we can tailor the writeup regarding our implementation appropriately. However, if you intend to post, publish, or cite, any information pertaining to Spry's efforts in implementing R2RML, and/or the systems and information, we ask that you credit Spry Inc.
>
> Currently, we are preparing one of our R2RML mappings for live systems testing in the next following weeks. This, along with some IA paperwork (unrelated to the mappings) are among the last few things we need to complete in order to push R2RML into an enterprise-wide production environment.
> I have also been working with one of our new customers in a commercial space, and have found myself discovering quite a few interesting things when faced with federating, sources exposed as R2RML, D2RQ, excel flat files, and raw triple stores. Source details on this I am likely unable to share, but the methodology and approach I can go into if there is interest.
>
> We can also go into how, and why, we took the R2RML mapping approach that we did, performance, using the mappings to query for provenance of the data, and the struggles/success that we had over the past few years.
>
> I would be happy to answer additional questions any of you have, and look forward to talking with you soon,
>
> Mykel
>
>
> On Wed, Jun 6, 2012 at 6:20 PM, ashok malhotra <ashok.malhotra@oracle.com <mailto:ashok.malhotra@oracle.com>> wrote:
>
>     Hello Mykel:
>     We spoke briefly this morning.
>     I am the co-chair of the W3C RDB2RDF WG that developed the R2RML and the DM specs.
>     We are trying to move these spec to W3C Recommendation and we need implementation
>     experience to do that.
>
>     What would be great is if you or one of your colleagues could run our test suite and create
>     an implementation report.  We will give you all the help you need.  If that is too much to ask
>     could send us mail saying that you have implemented R2RML and you are using it in your
>     systems and any other usage and implementation information you care you share with us.
>
>     Many thanks!
>     -- 
>     All the best, Ashok
>
>
>
>
> -- 
> Mykel Kulkarni
>
> Spry Inc.
> Office: 443-212-5072
> Cell: 609-240-6624