Monday, 12 March 2012

Open source and research: the case of GAIA

How important is the open-source software as part of the scientific research nowadays? Since I am an astronomical researcher, a lover and a user of open-source software, I’m very interested in trying to deepen this topic.

Well, an opportunity to talk about it now comes from the observation of software tools that are utilized in a big project to which I am also taking part, that is the definition of procedures for processing and analyzing the photometric data that will be generated by the ESA (European Space Agency) probe called GAIA (curiously, one of the first articles appeared on my italian blog was just about GAIA, in 2002). The probe will be launched in 2013, but the work for the definition of the appropriate procedures is already running at full capacity.

An artistic image of GAIA
(Credit: ESA/Medialab)

In my opinion, even a simple, brief list of software tools used by the different teams of Gaia - coordinated through a European network of scientific institutes - would probably be enough to understand that the open-source software is doing great or - to put it in a more technical way – that it now can count on its own defined space essential for applications and fields, at least in scientific research.

To proof this, I’ve written down a list (incomplete) of the open source software currently used in the development of the Gaia data reduction procedures, made by simply thinking about the tools that are used, by me or my colleagues, for the daily work within the project itself...

So, this is the catalog:
  • Java: is the language of analysis software and data reduction. Following a decision of ESA, all procedures need to be written in Java. This involves a series of remarkable benefits in terms of independence from the hardware, portability, modularity, etc. ... too long to be fully explained here.
  • Eclipse: is the highly recommended development environment  (which is to say, do what you like, but you don't expect support with other environments…)
  • SVN: all the code is put under revision control, using subversion
  • MediaWiki: there is a wiki with restricted access, very large, in which is shown all the project documentation, the meetings and seminars for the various teams, the documentation. Briefly, a sort of mini thematic Wikipedia, devoted to people working on the project.
  • Hudson: a tool to automatically test the codes, at scheduled intervals, and submit reports on webpage
  • Cobertura: this tool is able to calculate the percentage of the code accessible to the test procedures
  • Mantis: is the chosen tool for controlling and managing bugs in the project
  • Grace: a useful tool to create graphics
  • Topcat: an interactive browser of tables and data editor
  • ant: a useful compilation tool in Java
  • Plastic (Platform for Astronomical Tool Interconnection) is a protocol of communication among different tools utilized mainly in astronomy (now is going replaced by SAMP)
  • And probably there's something more that I cannot recall right now... :-)

The interesting thing is that all this software is released under the GPL (General Public License) or similar, which makes it much easier to spread and use the software itself: there is no need to obtain proprietary and restrictive licenses (or to make our own institutes acquire them…): you can download the software and begin to use it immediately. That’s it. It's not bad, I’d say, both for the "personal scientific productivity” and for the undoubted advantage that this has as part of the real project. Can you imagine how much of the researcher’s time and of the taxpayer’s money should be spent if they had to obtain licenses (renewals, software keys…) for all these things?


(Originally published on the blog SegnaleRumore, kindly translated by Claudia Castellani).

LinkWithin

Related Posts with Thumbnails