From two PhD students trying to get a few neuroimaging research software packages to work smoothly in their own labs, to a broad neuroscience software initiative with thousands of unique users and an even larger group of unwitting beneficiaries. In five years, without almost any external funding. The quick growth of NeuroDebian shows the value the research community places on neuroscience tools that work well together.
The two principal NeuroDebian leaders, Michael Hanke and Yaroslav Halchenko, started out as PhD students working on neuroimaging data, each trying to make it easier to deploy research software in their respective labs. In 2006 they met for the first time, and as a result founded the Experimental Psychology in Debian (ExpPsy) project which a few years later was renamed into the more general 'NeuroDebian' to reflect its widening scope. Currently, NeuroDebian is an active collaboration with dozens of research groups and more than 50 integrated research tools and datasets for neuroimaging, electrophysiology, psychophysics, and distributed computing.
Michael Hanke (MH):The goal of NeuroDebian is not to create a new, separate environment just for neuroscientific research. Instead, it aims to add the facet of neuroscience research to the abilities of the "universal operating systems" that Debian strives to be. This approach has many advantages over a separate project: longevity, man-power, and reduced cost to name just a few.
INCF: Who is it for - who can use it, and who benefits especially?
It is evident from the testimonials from individuals and research institutions that an integrated platform like NeuroDebian saves a lot of time and frustration for individual researchers and labs. What might be easier to miss for the community at large is the benefit to developers, who get help and support to make their software available to user, and can do so with less time and effort wasted.
Yaroslav Halchenko (YH): Debian is really for everyone, and is great for many scientists due to the outstanding work of other teams such as Debian Med, Debian Chem, and Debian Science. With such joint effort of enthusiasts from different fields of endeavor Debian provides a universal turnkey platform where general purpose and research software is well tested, integrated and made conveniently available to users.
MH: NeuroDebian is for anyone who wants things to just work and who would rather do science than wandering the web to get some software installed. It is also for researchers who need complex research software tools, but cannot afford the IT-support staff that their maintenance seems to require. It is for teachers that want to provide their students with a complete training environment that they can legally share and experiment with at home. Finally, NeuroDebian is for people that want to share their achievements with the community -- researchers that appreciate a channel for efficient dissemination of readily usable software.
A survey on tool use, environments and software requirements sent out by NeuroDebian to the neuroscience community earlier in 2011 shows that neuroscientists in general appear to like and use Linux; a majority of the respondents were using Linux as part of either their personal computing environment or their institution's.
INCF: how did your survey fall out? And what would you say about the result?
YH: The results were striking, heart-warming and informative. It was
great to get solid evidence that formerly known as "for-geeks-only"
Linux platforms are actually a flagman of scientific computing nowadays and
that majority of the users already familiar with virtualization principles.
MH: We already knew
that Linux would be strong
as a computing environment, but we didn't expect to see that it was basically the computing environment
of choice for most researchers, both on the big managed systems (e.g. clusters) and on the
desktop.
For many of these Linux users, NeuroDebian is only a short further step away - and for the people using WIndows or Mac, there is a possibility to set up NeuroDebian on a virtual machine. There are several indications that the community is catching on to the possibilities; usage of NeuroDebian has been growing quickly during the past two years.
YH: According to the statistics of the main repository alone (excluding 4 mirrors we have) there is a steady growth of users: if in Jan 2010 we got 963 unique visitors and 86,466 hits, in Jan 2011 3,500 visitors and 186,000 hits and this Sep 2011 it was 5,624 unique visitors and 356,574 hits.
INCF: Having come this far already, what are your future goals?
MH: We need to find a way to rest NeuroDebian on more shoulders. We are actively looking for interested people and offer Debian packaging training. We help scientific developers to acquire the skills to maintain their own software in Debian themselves. Financial support would help to address some items that are very high on our todo list, but so far we haven't been able to convince funding agencies that there is scientific value in Debian.
