Dear all,

I am very pleased to announce that we now have a particular scientific application in mind that we will attempt to tackle...

We are going to search for white dwarfs (for an introduction to white dwarfs, please visit http://imagine.gsfc.nasa.gov/docs/science/know_l1/dwarfs.html).

Bob Mann, Nigel Hambly and the other people in the Royal Observatory, Edinburgh have been very helpful in educating me about the problem and what needs to be done. Here's a very brief description (trying to avoid getting it wrong :-) )

Our application will need to provide a way for astronomers to find information about objects in the sky from a number of databases. It will utilise the SkyServer services and it will combine the results with information from other databases (mainly the SSA in Edinburgh). Our aim is to identify objects in the sky that are moving and, hence, are candidates for being white dwarfs. Since we are looking for moving objects, the database queries and comparisons cannot be based only on their recorded location (spatial search).

Our goal is to search for white dwarfs in our galaxy through analysis of the returned data. When (and if:-) we discover a white dwarf, we would like to be able to search for more information about it, then use the returned data related to the white dwarf's composition and images to create 3D visualisations of it. We could even locate other nearby objects and create a 3D representation of that portion of the galaxy. The auto-stereoscopic 3D Displays (part of the Durham e-Demand project, http://www.dur.ac.uk/n.s.holliman/3ddisplays.html), our newly acquired SGI CAVE, or other technologies elsewhere (you may have interesting technologies and ideas?) could be the targets for our visualisations.

The challenges ahead and what we would like to demonstrate:

The scientific problem we have is real and extremely interesting. We are looking forward to building a solution for the astronomers that could be used in their attempts to search for white dwarfs. However, we need to bear in mind that through this process our main goal is to demonstrate how service-orientation and the WS-GAF (http://www.neresc.ac.uk/ws-gaf) ideas can be applied in building global-scale, distributed, Grid applications.

Some of the features (but not the only) we would like our application to incorporate are:

- Distributed information access and queries for the astronomical data

- Computational resources for the models, and

- 3D visualisations

- Security

We think that the above features can be considered as "typical" of a Grid application. Through their incorporation in our global-scale application, we hope to demonstrate that it is possible to build Grid applications utilising the existing Web Services infrastructure. However, if during the process we identify a need for additional technologies we will document it and report our findings to the community.

This message effectively marks the beginning of the work on the WS-GAF application. Thanks to funding from the DTI and JISC, Jim Webber will begin work on the project on Monday (contract signing permitting). Jim was one of the authors of the GAF document, and also of a recent book on Developing Enterprise Web Services - http://www.amazon.co.uk/exec/obidos/ASIN/0131401602/ref=sr_aps_books_1_1/026-4499569-0816413).

Our plan for the first few months of the project is as follows:

1) Document the overall design, identifying the services and their interactions. For new services, define the information that will be exchanged. We will work with all interested parties in defining the appropriate document formats where existing specifications do not exist.

2) Design and implement any new services (we hope to keep these to a minimum).

3) Construct an initial prototype and evaluate the results.

Our aim is to have something to report on this at the March GGF in Berlin. In order to build the application we will implement WS-Context: the one part of WS-GAF for which we currently have no implementation.

After Berlin, the application can be enhanced to add features and so expand the scope of the evaluation.

We are very excited about the coming months. We hope to work with all of you throughout the process and get your advice on the architecture and implementation. If you have any ideas, comments or existing services that may be appropriate then please do get in touch.

Best regards,

--
Savas Parastatidis
http://savas.parastatidis.name (now blogging)