Dear FSL users,
[please pardon the hijacking of this mailing list]
The NeuroDebian [1] Team is asking for your support. We are hoping to
obtain funding for continued maintenance, development and expansion of
the project that, among other activities, produces binary packages of FSL
for Debian and Ubuntu. An initial grant proposal has been previously
reviewed and we are about to resubmit it to address reviewer comments
(PI Dr. James V. Haxby; NIH program announcement PAR-08-010: Continued
Development and Maintenance of Software (R01)).
Please see the abstract and specific aims at the end for a more
detailed description of the updated project proposal.
We need to address two main reviewer concerns:
1. Proof of the state of the project
We previously failed to convince the reviewers that our efforts
_already_ help researchers to maintain a productive research software
environment with minimal effort. Therefore, if you are using
NeuroDebian, especially its FSL packages, and you feel that it is
beneficial for your research activities, we would appreciate your
letter of support describing: Why did you start using NeuroDebian?
What do you use it for?
2. Feasibility of virtual environments for software deployment
The reviewers argued that using a virtual environment (i.e. a virtual
machine, VM) is not a feasible solution to the problem of deploying
an integrated platform, like NeuroDebian, on the two major
non-GNU/Linux operating systems (Windows and Mac OS). Therefore, we
would appreciate your letter of support, if you rely on a VM to run
or evaluate research software (maybe to run FSL on Windows). Such
letter would preferably describe why you use a VM, and could offer a
short summary of the VM experience in your research activities.
We also appreciate letters on other aspects of the proposal, and would be
delighted to see requests for any particular functionality included in them.
If you would like to see the NeuroDebian project to continue its
development, we would be thankful if you send your "Letter of Support" via
email [4] (preferably a PDF) or fax [5] to provide additional weight for our
application. For your convenience, we have composed a generic letter
template [6].
If you have previously provided us a letter of support, and either
want to retract or alter it, based on the updated project description,
please email [4] us.
We would appreciate if we receive your letter of support within a
week, so we are still on time with the resubmission and ready to
dedicate ourselves to HBM 2011 (visit us at booth #108).
Thank you very much in advance for your support,
the NeuroDebian team
About NeuroDebian
-----------------
If you are using a Debian or Ubuntu operating system for your neuroscience
research you might already benefit from our efforts of integrating
neuroscientific software into these platforms. For the past 6 years our team
has provided Debian/Ubuntu packaging, maintenance and troubleshooting for open
source software, such as AFNI, ANTs, BiosSig4C++, Brian, FSL, Caret, Lipsia,
MRIcron, NiPy(PE), PsychoPy, PyMVPA, PyNN, and Voxbo (see [2] and [3] for more
complete references). Having research software developed by different groups
with different technologies properly integrated into a uniform environment
allows scientists to easily maintain a versatile up-to-date research
environment with just of few commands/mouse-clicks and focus on actual research
instead of tedious system administration tasks.
[1] http://neuro.debian.net
[2] http://neuro.debian.net/pkgs.html
[3] [log in to unmask]" target="_blank">http:[log in to unmask]
[4] mailto: NeuroDebian Team <[log in to unmask]>
[5] Fax: +1 (603) 646-1419
[6] http://neuro.debian.net/_files/letter_of_support_template.txt
-----------------------------------------------------------------
Abstract
========
Complex software systems play a more and more important role in
neuroscience research and managing an appropriate research environment
is becoming increasingly difficult. NeuroDebian
(http://neuro.debian.net) is a turnkey research software platform for
all aspects of the neuroscientific research process. It takes the
ideas of the Neuroimaging Tools and Resources Clearinghouse (NITRC,
http://www.nitrc.org), on maximizing research transparency and methods
sharing, one step further, by providing a comprehensive suite of
readily usable and fully integrated software with a robust testing and
deployment infrastructure. Consequently, it improves interoperability
among the tools and frees researchers from the burden of tedious
installation or upgrade procedures. That, in turn, positively affects
their availability for actual research activities, as well as their
motivation to test new analysis tools and stay connected with the
latest methodological developments in the field.
Over the past six years, NeuroDebian has integrated dozens of
neuroscience software tools into the Debian operating system
(http://www.debian.org), making its current version, Debian 6.0, the
first operating system world-wide with comprehensive built-in support
for MRI-based neuro-imaging research. In close collaboration with the
Debian community and all involved neuroscience research groups we have
provided middleware support for users and developers – consulting
developers regarding release practices and legal aspects and
streamlining technical support of NeuroDebian users. This joint effort
has been well received by the research community, and, according to a
recent survey, GNU/Linux-based systems are now the most common
computing platform in neuroscience, and NeuroDebian is the most
popular software resource dedicated to neuroscience.
To further contribute to the dissemination of new methods, the
NeuroDebian project aims to expand its coverage of software and to
assure robust operation across a wide variety of deployment
scenarios. Developing an environment with a large number of tightly
integrated neuroscience software tools will allow for testing efforts
that continuously verify software interoperability. We will develop a
framework to derive a comprehensive description of a NeuroDebian
analysis environment, and offer anyone the building blocks to, later
on, reincarnate an identical copy, thus addressing an essential aspect
of reproducible research. By means of virtualization solutions we will
offer researchers the tools to take advantage of NeuroDebian on
non-GNU/Linux operating systems, and advanced computing platforms
(e.g., distributed and cloud computing) for efficient large-scale data
analysis and modeling.
By fostering proven and efficient practices of the free and
open-source software community in neuroscience, NeuroDebian will help
to assure the availability and continued usefulness of existing
software.
Specific aims
=============
This project aims to further improve integration of neuroscience
software into the larger free and open source software community by
adopting standards and practices that have proven to yield a maximum
of quality and productivity. To this end, we will keep working closely
with a large number of neuroscience software developers, as well as
the Debian community. In particular we aim to achieve:
Aim 1 Ongoing maintenance of neuroscientific software in (Neuro)Debian
NeuroDebian currently maintains over 30 software projects, from
single-purpose tools to complex analysis suites. All integrated
software requires timely response to bug reports, and software
updates. We aim to continue to offer reliable and prompt service in
providing an efficient research environment.
Aim 2 Increased coverage of neuroscientific research tools
To enhance the utility of NeuroDebian for a wide range of research
applications we will
a extend software coverage beyond (f)MRI/DTI-based neuroimaging to
tools for intra/extra-cellular recording and modeling, EEG/MEG,
and data management: e.g., BrainVisa/Anatomist, Camino, DTI-TK,
FreeSurfer, NEURON, XNAT, and other software that becomes
available during the project lifetime;
b integrate essential Matlab-based open-source software: e.g.,
BrainStorm, EEGLAB, Fieldtrip, PsychToolbox, SPM;
c facilitate work on increasing the compatibility of Matlab-based
neuroscience tools with alternative open-source computing
platforms – such as Octave – to improve their availability in
high-throughput, and cloud computing environments and loosen
dependencies on proprietary systems;
d mentor interested developers in maintaining their software in
Debian by themselves.
Aim 3 Quality and interoperability assurance
Independent research software tools evolve at their own pace. This
poses a challenge for heterogeneous computing environments. To
assure reliability and interoperability without stagnation we will
a exercise available test batteries on recent and upcoming releases
of Debian and Ubuntu to assure robust performance and inform
developers about upcoming changes before researchers are affected;
b develop new test suites for common heterogeneous analysis
pipelines and run them routinely to assure proper functioning and
ongoing compatibility of all involved tools;
c make developed test suites readily available to users so they can
verify correct operation of their particular research
environments.
Aim 4 Sustained availability of software and precise re-creation of
complete research environments
The scientific workflow frequently requires re-analyses of data with
particular versions of software, for example, to revise a manuscript
or to reproduce a study. We will
a employ Debian’s existing software archive snapshotting framework
to preserve and distribute all previous and current versions of
supported software in NeuroDebian;
b build on Debian’s package management systems, to develop tools to
describe a particular analysis environment (with all versioned
dependencies) to be able to reconstruct it at any later point in
time – by anyone – given access to the specification and to the
software archive snapshots.
Aim 5 Broad availability of NeuroDebian on common and advanced
computing platforms
A NeuroDebian-based system is not bound to computers solely running
Debian. We will
a provide binary packages for Debian-derived operating systems
(e.g., Ubuntu);
b provide a virtual appliance allowing deployment of NeuroDebian in
a virtualized environment on proprietary operating systems
(e.g., Microsoft Windows and Mac OS X), as well as on other
non-Debian GNU/Linux distributions;
c provide NeuroDebian system images for cloud and high-throughput
computing that are compatible with popular service providers and
environments, such as Amazon EC2, and Condor.
|
