Re: [OT - or may be not] The case for open computer programs

[Luis Ibanez <luis.ibanez@kitware.com>]

What is more interesting is the reaction that 
followed in serious scientific journals:


1)  PLoS ONE (the Open Access Mega Journal that currently
     publishes 3% of all the STM literature) now requires 
     software papers to include the source code under an 
     Open Source license:

http://www.plosone.org/static/guidelines.action#software
http://www.plosone.org/static/policies.action


<quote>

Software. PLoS supports the development of open source software and believes that, for submissions in which software is the central part of the paper, adherence to appropriate open source standards will ensure that the submission conforms to (1) our requirements that methods be described in sufficient detail that another researcher can reproduce the experiments described, (2) our aim to promote openness in research, and (3) our intention that all work published in PLoS journals can be built upon by future researchers. Therefore, if new software or a new algorithm is central to a PLoS paper, the authors must confirm that the software conforms to the Open Source Definition, have deposited the following three items in an open software archive, and included in the submission as Supporting Information:

• The associated source code of the software described by the paper. This should, as far as possible, follow accepted community standards and be licensed under a suitable license such as BSD, LGPL, or MIT (see http://www.opensource.org/licenses/alphabetical for a full list). Dependency on commercial software such as Mathematica and MATLAB does not preclude a paper from consideration, although complete open source solutions are preferred.
• Documentation for running and installing the software. For end-user applications, instructions for installing and using the software are prerequisite; for software libraries, instructions for using the application program interface are prerequisite.
• A test dataset with associated control parameter settings. Where feasible, results from standard test sets should be included. Where possible, test data should not have any dependencies — for example, a database dump.""

</quote>

One of the new Journals in BiomedCentral,

   http://www.openresearchcomputation.com/about

which is also Open Access,

developed a similar policy:

<quote>

• Reproducibility Verification
  • Software: "Open Research Computation differs from other journals with a software focus in its requirement for the software source code to be made available under an Open Source Initiative compliant license, and in its assessment of the quality of documentation and testing of the software."
  • Data: Open Research Computation has very high standards for data availability and reproducibility. It is expected that all the data, code, and software required to reproduce any examples in the paper will be made freely available for download from an appropriate recognized repository or the journal website.
• Review criteria for Source Code
  
  • Code and License
    • Is the source code as well as executables and/or an instance of the service (of a clearly defined version) available on appropriate public repository?
    • Is the source code made available under an Open Source Initiative (http://www.opensource.org/licenses/category) compliant license? Specifically users must have the right to examine, compile, run and modify the code for any purpose.
    • Are project authors and contributors clearly defined, ideally through a Description of a Project [http://en.wikipedia.org/wiki/DOAP, http://trac.usefulinc.com/doap] document? We recommend the use of the automatic DOAP generator such as those linked here: http://trac.usefulinc.com/doap/wiki/Generator

</quote>



It has been said that Open Source was the application
of the Scientific Method to the process of Software 
development.  These recent developments show that
Open Source has a lot to give back to the scientific
community where the practice of Reproducibility 
Verification has been lost and substituted by the 
inferior and quite defective practice of peer-reviews
based on simple opinions instead of reproducible 
experiments.

For one thing, the simple practice of doing revision
control, and implementing unit testing frameworks
that can be executed over and over again, will 
already revolutionize the way software is managed
in many research institutions. It is sadly too common
that nobody in a lab can replicate a computational 
experiment even days after it has been performed.


More on this by Victoria Stodden:
http://www.stanford.edu/~vcs/talks/CaltechMay122011-STODDEN.pdf
Technology and the Scientific Method:
Tools and Policies for Addressing the Credibility Crisis in Computational Science."


For more in the story behind that Nature paper:
http://videolectures.net/cancerbioinformatics2010_baggerly_irrh/


and since the topic of Open Access came up,
there is still time to sign this petition to the White House:

                        http://wh.gov/6TH

to make all US Federally Funded research 
available as Open Access to the public.

More details in:

http://blog.wikimedia.org/2012/05/25/wikimedia-foundation-endorses-mandates-for-free-access-to-publicly-funded-research/
http://lj.libraryjournal.com/2012/05/legislation/acrl-urges-librarians-to-sign-research-access-petition/



     Luis



---------------------------------------------------------

On Tue, May 29, 2012 at 3:20 AM, lina <lina.lastname@gmail.com> wrote:

On Tue, May 29, 2012 at 2:49 PM, Oz Nahum Tiram <nahumoz@gmail.com> wrote:
> Hi All,
> Indeed strong words, published in Nature where you need to pay $32 to
> read what we all know already.

There is a link, probably you may access (temporarily) to read if you
are interested.

https://docs.google.com/open?id=0B93SVRfpVVg3aW0tX3RLanRKdUE


Best regards,

>
> Regards,
> Oz
>
> On Tue, May 29, 2012 at 8:00 AM, Andreas Tille <andreas@an3as.eu> wrote:
>> Hi,
>>
>> you might like to read:
>>
>> The case for open computer programs
>> Darrel C. Ince, Leslie Hatton & John Graham-Cumming
>>
>> Scientific communication relies on evidence that cannot be entirely
>> included in publications, but the rise of computational science has
>> added a new layer of inaccessibility. Although it is now accepted that
>> data should be made available on request, the current regulations
>> regarding the availability of software are inconsistent. We argue that,
>> with some exceptions, anything less than the release of source programs
>> is intolerable for results that depend on computation. The vagaries of
>> hardware, software and natural language will always ensure that exact
>> reproducibility remains uncertain, but withholding code increases the
>> chances that efforts to reproduce results will fail.
>>
>> http://www.nature.com/nature/journal/v482/n7386/full/nature10836.html
>>
>>
>> Kind regards
>>
>>       Andreas.
>>
>> --
>> http://fam-tille.de
>>
>>
>> --
>> To UNSUBSCRIBE, email to debian-science-request@lists.debian.org
>> with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
>> Archive: [🔎] 20120529060057.GC18760@an3as.eu" target="_blank">http://lists.debian.org/[🔎] 20120529060057.GC18760@an3as.eu
>>
>
>
> --
> To UNSUBSCRIBE, email to debian-science-request@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

> Archive: [🔎] CADKMUwm_iJLK15dLaK142u8Si2Fpqe4PtrX6oUmUTztHkR7yZg@mail.gmail.com" target="_blank">http://lists.debian.org/[🔎] CADKMUwm_iJLK15dLaK142u8Si2Fpqe4PtrX6oUmUTztHkR7yZg@mail.gmail.com

>


--
To UNSUBSCRIBE, email to debian-med-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Archive: http://lists.debian.org/CAG9cJmk0LPYhsWnZQ0u2OFdXaTAsXF5K1H+kWsaBi4WxqEuZQ@mail.gmail.com