Difference: SocialComputingTopics ( vs. 1)

Revision 12005-10-19 - Main.DenisPerretGallix

Line: 1 to 1
Added:
>
>
META TOPICPARENT name="EmergingTopics"

Social, Collaborative or Web Computing

After pointing out the necessity to increase strongly the computing power open to scientific research, let me address the next most important issue, namely: more and better collaborations. As said previously, our field progresses thanks to the expertise of computer scientists, physicists, theorists and mathematicians. But each group has different motivations, different thinking patterns, uses different concepts expressed in different languages. Moreover our community is large and spread out over distant places.

Therefore surveying, discussing and, possibly, developing tools to bridge these breaches has been a permanent concern in this workshop series and should remain high on the agenda. These questions have triggered both in the academic and commercial worlds the development of tools which include email, chat, net-conference, VoIP , groupware or blogs.

Some of these technologies rely on the development of additional middleware layers between the browser and the server like *AJAX* (Asynchronous JavaScript and XML) to speed up web applications. More and more ``office applications'' used to run on isolated computers can be ported to the web to create a new collaborative environment.

Simultaneously, a new revolutionary use of the web is taking place based on the [WikiTopics][*Wiki*]] concept. Even if the main ideas have matured for more than a decade, the Wiki technology is changing the way we use the web. It gives the possibility to everyone, would it be physicist, theorist or mathematician, to modify or add new text or file to any Wiki web page using simply a web browser.

This may not appear, at first sight, as a major innovation, but it brings solutions to many every day problems. A documentation or information web page can be updated or created by the very author of the news without relying on a webmaster. This is a road to an always up-to-date information. Everyone can talk to everyone without going through mailing list discussions. Scientific papers can be written quasi-simultaneously over the Internet without the cumbersome cycle of comments and corrections through email. A common global knowledge database is being progressively built.

The success of Wikipedia is here to show that this model works and provides a new communication technology that can be used both at the global level and within a limited collaboration. The selection of the most appropriate Wiki implementations, the development of specific tools like latex-ps-pdf interfaces or the implementation of collaborative code development are important issues.

Collaborative developments need also better code and project documentations. Automatic documentation systems digging comments and information from the code files themselves are a great improvement toward an on-line instantaneous project documentation. Javadoc, Doxygen or Root THtml class are good examples of such packages.

Collaborative projects are more easily managed if a common development platform is used by all contributors. Among them, the Eclipse integrated development environment, based on Java, runs on most OS and computer systems. Many computer languages, modeling and documentation tools have been implemented at various level of interoperability. Initiated by several major companies, it is now made available to the open source community. Comparisons with other potential packages like Xcode, Kdevelop or NetBeans as well as possible HEP specific plugins developments should be addressed.

Finally, software engineering has been, so to say, revisited and a ``new'' concept called by the fancy name of *``extreme programming''* or XP has made its way to the community. XP is ``a deliberate and disciplined approach to software development''. This approach precisely insists on: setting up short links between programmers and users, making short development cycles but with many iterations, promoting strong involvement of end-users and pair programming (one work-station for two persons). Is this model appropriate to physics research activities and to the new societal environment? In any case this is pushing to the extreme the obsessing need to develop and support collaborative work, trademark of this workshop series since its origin.

-- DenisPerretGallix - 19 Oct 2005
 
This site is powered by the TWiki collaboration platformCopyright &© by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback