TUCSON, Ariz. - In the spirit of openness that has traditionally defined both the scientific community and the open source software community, bioinformaticists working at the intersection of these two disciplines gathered here today to discuss the role of public funding agencies in supporting open source bioinformatics software.
A panel discussion today at the O'Reilly Bioinformatics Technology Conference served as a forum for a healthy debate on the range of interpretations of this issue within the field.
While a panel on open source software at an O'Reilly conference may seem a preaching-to-the-converted scenario, the discussion served to highlight emerging differences in opinion, sparked largely by a petitiondrawn up in September by a subset of the bioinformatics open source community that calls for all bioinformatics software developed using public funds to be released under an open source license.
The petition, supported by bioinformatics advocacy group openinformatics.org, launched a number of heated debates on slashdot.org, O'Reilly Network, and the Open Informatics mailing list. A key sticking point for many open source bioinformatics programmers was the wording of the petition, which explicitly demands that public funding agencies "require" publicly funded software to be released under an unspecified open source license.
"I am against the Open Informatics petition because the originators do not fully understand what open source means," wrote Andrew Dalke of Dalke Scientific Software in his O'Reilly Network post, "Why I'm Not Supporting the Open Informatics Petition." In this document, Dalke posited that requiring publicly funded researchers to publish any source code under an open source or free software license "will hinder the development of science."
In the face of the mounting discussion around this topic, the Open Bioinformatics Foundation - the non-profit umbrella group that supports BioPerl, BioPython, BioJava, BioDAS and other projects -- released a position statementat today's panel discussion. According to OBF board member Steven Brenner of the University of California, Berkeley, the statement was necessary to distinguish the OBF from organizations "with similar sounding names" and "because everybody was asking us where we stood on the topic."
The OBF statement takes a more moderate stance than the openinformatics.org petition. Noting that "scientific software developed with public support should be distributed under terms analogous to those applied to biological materials," it concludes, "We encourage institutions to delegate to their scientists the opportunity to select non-restrictive and open source licenses for their software."
Additionally, following the precedent Brenner set by renegotiating his employment contract with Berkeley to enable him to produce open source software, the OBF is investigating ways to support bioinformatics researchers in negotiating similar contracts.
Most of the developers present at the panel discussion were quick to note that this solidifying of philosophy among subsets of the bioinformatics open source community shouldn't be interpreted as a Free Software Foundation/Open Source Initiative-style rift.
Jason Stewart, one of the authors of the openinformatics.org petition, said he's taking many of the arguments against the petition into account and is planning to redraft the petition to clarify some of the more controversial points.