 |
Peer-Reviewed Independent Software |  |
Mission |
Components |
Concepts |
Licensing |
Philosophy |
|
Peer-Reviewed Independent Software | |
Mission |
Components |
Concepts |
Licensing |
Philosophy |
Prism is a vision, a mission if you will, to encourage the evolution of peer-reviewed, independent software for academic and clinical neonatology. Our purpose is not to compete with proprietary clinical systems, which will no doubt continue to serve an invaluable role in neonatal care far into the future. Rather, our goal is to be able to incorporate new practices, improve work flow, add features, and fix newly discovered problems as healthcare evolves, promoting reliability and quality through independent peer review. Whether or not our particular experiment succeeds, we believe this approach is a good fit philosophically for academic neonatology.
By "peer-reviewed" we mean open-source, and by "independent" we mean free. (Free in the sense of liberty, not necessarily in the sense of price. You may, for example, charge a fee for distributing free software, but you cannot prevent anyone else from giving away, without charge, as many copies as they want.) For those unfamiliar with the concepts of open-source or free software, please see the overview of essential concepts below. In a nutshell:
The basic idea behind open source is very simple: When [developers and users] can read, redistribute, and modify the source code for a piece of software, the software evolves. People improve it, people adapt it, people fix bugs. And this can happen at a speed that, if one is used to the slow pace of conventional software development, seems astonishing. [OSI]
Such an approach to software development encourages:
Mindwrapper, a software development framework, written in Python with wxPython, designed to make it easier for clinicians and other non-programmers to participate in software evolution.
The Prism Library (The Library) is a suite of self-contained custom or customizable clinical applications written using Mindwrapper.
The Prism Application (or just Prism) is a larger standalone application that integrates a number of components from The Library (also written using Mindwrapper).
We have chosen the terms peer-reviewed and independent to describe our collaborative approach to software development because: (1) these terms are possibly more familiar to academicians, and (2) they are relatively neutral terms corresponding to the two contrasting movements called "open-source" and "free software". The following links are useful in comparing the two movements.
| peer-reviewed | independent |
|---|---|
| "pragmatic" open-source |
"ideological" free software |
| the initiative, (home page) |
the movement, (home page) |
| official definition (summary) |
official definition (summary) |
To the degree that these alternatives represent opposing positions, we do not intend to pick one side over the other. Rather, we recognize important contributions from both sides of this conversation.
A central (and often devisive) concept in open-source software licensing is copyleft, also called reciprocity or share-alike. The basic idea is that if you take free software and make improvements to it, you cannot distribute your modified versions (for money or otherwise) unless you do so under the same terms as the original software. An excellent summary of this concept and the issues surrounding it can be found at the wordiQ site, from which the following quote is taken.
The word copyleft started out as a pun on copyright. Where copyright law is seen by the original proponents of copyleft as a way to restrict the right to make and redistribute copies of a particular work, a copyleft license uses copyright law in order to ensure that every person who receives a copy or derived version of a work, can use, modify, and also redistribute both the work, and derived versions of the work. Thus, in a non-legal sense, copyleft is the opposite of copyright. [wordiQ]
Strong proponents of copyleft see it as a way "to create the most favorable conditions for a wide range of people to feel invited to contribute improvements and/or elaborations to this work, in a continuing process." On the other hand, others recognize a beneficial role for proprietary software derived from open-source software, and prefer licenses that are more permissive in this regard. Such individuals, even in advocating free software, often see copyleft licenses as a deterrent to participation in open-source development. Which position is "correct" obviously depends upon the situation, although the pros and cons may not always be entirely clear up front.
Over fifty different licenses are listed and discussed on the OSI and FSF Web sites. Of these licenses, four may be considered cannonical in the sense of being widely accepted as best practice.
| open-source | free software | |
|---|---|---|
| copyleft | OSL | GPL |
| permissive | AFL | LGPL |
The Open Source License (OSL) and the Academic Free License (AFL) are similar to their counterparts, the GNU General Public License (GPL) and Lesser General Public License (LGPL), with the following differences.
With regard to Prism, we believe that these three points are of minor significance. Furthermore, there may be advantages in keeping this project compatible with the GNU General Public License (GPL), so we will select from the right-hand column. With regard to the copyleft issue, we see little or no reason to be permissive in allowing proprietary derivatives to be distributed, so our choice falls to the copyleft GPL.
Prism. We find no compelling reason to allow derivatives of Prism (the primary application) to be made proprietary. The license for the Prism application should therefore be a copyleft style license, and we have chosen the GNU General Public License (GPL). We believe the GPL will invite clinicians and others to feel comfortable contributing improvements and additions to this system.
The Library. Likewise we see no advantage in allowing proprietary derivatives of contributions to the Library. We cannot (and do not particularly wish to) prohibit others from writing and using their own custom applications with Mindwrapper. If the authors distribute such programs, however, we urge them to do so under the GPL, and if they want to contribute them to the Library, they must do so under the GPL.
Incidentally, the authors of such works retain the copyright to their work. If a cohesive community grows up around this project, at some point in the future we may ask contributors to assign the copyright to a consortium, but for now, individuals keep their own copyrights under the GPL.
Mindwrapper. We recognize the possibility that a more permissive license for Mindwrapper might encourage a broader audience of programmers to contribute to its ongoing improvement. It is not absolutely clear that this is the case, however, so for the time being we will use the GPL for Mindwrapper, as well. By the nature of the framework, any application written in the Mindwrapper notation is "a derivative" thereof. If you come up with a good reason for wanting to distribute such a derivative under something other than the GPL, contact us and we will consider dual licensing to include the LGPL or possibly the AFL. Incidentally, the licenses for Python and wxPython are OSI-approved and GPL-compatible.
In a brilliant on-line article, Open Source and Academia, Taylor and Riley conclude that academic research and pedagogy should make the strongest use of Open Source as a model for publication and distribution of scholarly work, and its production in the academic environment. Quoting Feller and Fizgerald (2002, Understanding Open Source Software. London: Addison-Wesley.) they list the following apects of the Open Source model and show how these are consistent with the goals of academia. Open Source:
Taylor and Riley take the Open Source model as an inspiration for how scholarly works should be produced and published. Although we agree with this line of thinking, we merely take this argument to confirm that peer-reviewed independent software development is a natural fit for academic neonatology. If Open Source is a good model for other aspects for academic productivity, surely it is a great model for academic software development.
It should be mentioned that university administrators often have a tendency to see computer programs developed by their faculty members as opportunities for income, not as opportunities to contribute to human knowledge. Finding ways to bring about a change in this attitude is a central aspect of the Prism vision. Dr. Walter gratefully acknowledges the willing and supportive cooperation of the University of Arkansas for this project (UAMS Invention Disclosure I02-24).
This page is licensed under a Creative Commons License.