[Insight-users] Reproducible results and semi-open science
Luis Ibanez
luis.ibanez at kitware.com
Mon Apr 14 18:38:20 EDT 2008
Hi Andriy,
Thanks for posting this very interesting question.
A similar situation has been identified in the Open Access
community, where there are different "degrees of openness",
and therefore they have come up with qualitative such as
Green Open Access
Golden Open Access
http://en.wikipedia.org/wiki/Open_access
------------
Before going into your original point, let me make a side
comment that is quite relevant here.
The terms:
* Academic
* Research
* Commercial
are usually brought up in to the discourse of Licensing for
Patents and Copyright, unfortunately, they are not always
defined properly.
It is common to assume that "research" is the disinterested
activity of intellectuals searching for knowledge, and we
tend to disregard that the largest institutions performing
research are indeed multi-national corporations.
Here are some examples of annual research budgets in 2007,
in billions of US dollars (billion = 1,000 million).
Toyota $ 7.48
Pfizer $ 7.42
Ford Motor $ 7.20
Johnson & Johnson $ 7.12
Microsoft $ 7.12
DaimlerChrysler $ 7.00
GlaxoSmithKline $ 6.61
Siemens $ 6.60
General Motors $ 6.60
Volkswagen $ 6.03
For more details, see IEEE ranking of R&D Top 100:
http://www.spectrum.ieee.org/dec07/5742
[Note that this is R&D Research and development budget,
that is, not limited to research]
It is unlikely that when authors say that their software can
be used for "research", they actually meant that they are fine
with multi-national corporations using the software at no cost.
On the other hand, you have government agencies investing on
research. Here are some examples of the US federal research budget:
Defense $ 80.68
National Institutes of Health $ 28.66
NASA $ 12.78
National Science Foundation $ 5.17
For more details on the US federal research budget, see the
following analysis by the American Association for the
Advancement of Science (AAAS):
http://www.aaas.org/spp/rd/prev09p.htm
A large fraction of the US Federal research budget is spent
in Universities in the forms of grants and research contracts.
However, even in the cases when research is done at Universities
and even when these Universities are non-profit organizations,
the "goodness" of the research activity is tainted by the
compulsion for patenting the results of the research.
---------
On the "Academic", side, academia is not necessarily an activity
outside of the commercial realm. There are private academic
institutions which operate as a business. That was actually one
of the reasons why the US Supreme Court stated that Duke University
was not excluded from complying with the terms of patents filed by
others.
Should we consider that a private for-profit institution is an
"academic" endeavor ? or a "commercial" one ?
How about a private non-for-profit organization ?
------------------------------
Let's now go back to your questions regarding open science and
your example about software licensing.
1) Your example states that there is a software "A" that is
distributed under a modified MIT license, where the
modification prohibits commercial applications.
Note that this prohibition make the license incompatible
with the definition of "open source" established by the
open source initiative.
You add that portions of software "A" are distributed
under the LGPL license, and some portions under the QPL
license.
2) The software "B", that you want to distribute under
an OSI open source license uses software "A".
3) You are interested in submitting to the Insight Journal
a paper that uses software B, and that therefore will
require software A as well.
This situation doesn't really raise too much of a concern,
because:
a) The Insight Journal is not a commercial endeavor
b) You are providing the source code of software "B"
and presumably the source code of software "A" as well.
By not being a commercial use of the software, you are still
satisfying the license of software "A", and since you are
providing the source code of both "A" and "B", you are satisfying
the LGPL and QPL licenses.
To summarize: as far as copyrights and licenses are concerned
this shouldn't be a problem.
You can simply submit software "B", along with a copy of
software "A", and your application + tests + data.
The concerns are rather of practical nature:
How big is software "A" ?
How easy is to automatically
build/install software "A" ?
How many other potential submissions
could benefit from software "A"
being available in the IJ server ?
The main interest of the Insight Journal is to enforce the
*reproducibility* of papers. The fact that some of those
papers have restrictions on what readers can do with them
after verifying their reproducibility, is a matter that
falls out of the scope of the Journal. It is however healthy
to make readers aware of those restrictions, so they don't
assume that all contributions are as freely available as
the ITK and VTK toolkits. (free as in freedom and free as
in price).
Please let us know if this answer your original question,
and whether you have any additional questions regarding
the Insight Journal.
Regards,
Luis
--------------------
Andriy Fedorov wrote:
> Hi,
>
> I am looking for some expert advice. Currently, there are pieces of
> software, which allow academic and research use royalty-free. With
> such software, reproducibility is still important, I believe, but you
> cannot call it "open science", because there are things you can and
> cannot do, thus the term "semi-open science"...
>
> Here is the situation. Let's say there is software "A", which is open
> for academic and research use without any additional arrangements, but
> not for commercial use. Let's say someone produced software "B", which
> they want to release under non-restrictive http://www.opensource.org/
> license, but that software "B" is using software "A" (without any
> modifications -- header files/linked libraries). To be more specific,
> consider two situations:
>
> 1) software "A" is released under the open source MIT license with the
> following explicit exceptions:
>
> "Distribution of modified versions of this code is permissible UNDER
> THE CONDITION THAT THIS CODE AND ANY MODIFICATIONS MADE TO IT IN THE
> SAME SOURCE FILES *****.h AND ****.c REMAIN UNDER COPYRIGHT OF
> THE ORIGINAL AUTHOR, BOTH SOURCE AND OBJECT CODE ARE MADE FREELY
> AVAILABLE WITHOUT CHARGE, AND CLEAR NOTICE IS GIVEN OF THE
> MODIFICATIONS.
>
> Distribution of this code for any commercial purpose is permissible
> ONLY BY DIRECT ARRANGEMENT WITH THE COPYRIGHT OWNER."
>
> 2) software "A" has some pieces released under LGPL, and some pieces
> under QPL. Software "B" uses pieces from both groups.
>
> I believe, that in both cases there is no violation of either of the
> three licenses by distributing software "B" under open source license.
> Although, under LGPL, the responsibility of following the license by
> the act of linking software "B" with software "A" is the
> responsibility of the end user. Am I correct?
>
> Practical question: what is the proper way to submit software "B" to
> Insight Journal? Software "A" is not included in the default
> configuration of the submission check system, so I am not sure how
> software "B", which depends on "A", could be tested.
>
> Finally, is there an interest for the Insight Journal to receive a
> paper about open software "B", which is using semi-open software "A"?
>
> Thanks in advance for your help!
>
> Andriy Fedorov
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
>
More information about the Insight-users
mailing list