The end of the WS-GAF project

Dear all,

Happy New Year to all!

This message is to let you know of the end of the WS-GAF project. Friday is our last day and Jim Webber will no longer be working for us. I would like to thank Jim for all his excellent ideas and hard work throughout the last year. We are going to continue our collaboration of course and, in fact, we are already preparing something for the end of January that we hope you’ll find very interesting.

WS-GAF has come a long way. It was funded a year ago to investigate how “typical” Grid applications could be built only with the power of existing, standard Web Services technologies.

Here’s a very brief history…

In August 2003, Paul Watson, Jim Webber, Thomas Rischbeck, and I published a white paper expressing our concerns with the approach the Grid community was taking with OGSI. It started a series of discussions within the community; some constructive others not so. We now know that others had similar concerns with us. As a result, a process started to replace OGSI with what we know now as WSRF. WSRF adopted many of our recommendations:

– Unmodified use of existing WS technologies

– Clear separation between the concepts of a “service” and a “resource”

– Break the implicit 1-1 associated between a Grid Service Instance and a resource

– Factorisation of the spec

– Treat services as deployable entities and not as dynamically created instances

– Move away from Grid Service Instances

The WSRF authors chose not to move away from their conceptual model, one of our main points, but I guess we can’t have everything 🙂 We think that WSRF still has some object-oriented feel to it and the way in which it deals with state is not good for internet-scale applications. I guess only time will tell whether it can make it. WSRF is now being standardised in OASIS, still without the full support of the entire WS community!

In December 2004 we were funded to demonstrate that our approach to Grid community had some validity. We decided to build two applications within a year. One was going to be scientific in nature while the other was industry-oriented. Both of the applications were built using only existing specs and tools. The scientific application was about finding potential White Dwarfs in our galaxy by combining information from both the SDSS and SSA archives and visualising the results (Bob Mann’s contribution to this project was invaluable and I would like to thank him for that). The industry application was about the logistics of transferring bottles with chemicals between companies assuming a sensor-based network and infrastructure. Jim has done a great work on that.

We are still preparing the reports/papers for these two applications and as soon as they are ready I will send links to them.

So, after

– 3 journal publications (with more being prepared)

– 1 conference paper

– 1 invited MIT book chapter (under review)

– Co-authorship of the white paper-strategy for the UK e-Science community

– 1 industrial journal paper

– Workshop presentations

– >20 invited talks throughout the world

– >500 blog entries

– Lots of arguments with some members of the Grid community 🙂

we are coming to an end. Overall we are very pleased with the outcome of such a small project (it was funded for just 1 researcher for 1 year).

With Jim moving on, I am now moving beyond the arguments with the Grid community and back to focus on delivering scientific results and interesting applications. Paul has been great in allowing me to concentrate on WS-GAF but now it’s time to move all my energy back to the North-East Regional e-Science Centre and its activities (they are paying my salary after all). We are going to continue working in the area of Web Services and Internet-scale applications and I hope we’ll have much to report through some of the great projects that are under way in Newcastle.

I would like to thank Jim again for his work and Paul Watson for his great support. I would also like to thank all of you for the great discussions on this mailing list but also the face-to-face ones with some of you.

However, this is not an end to this mailing list. I am planning to continue sending links to interesting articles in the areas of Web Services and Internet-scale computing. I do hope that the discussions will continue.

–

Savas Parastatidis