Question

Forum:Author affiliation and for-profit licensing of published software

1

Entering edit mode

9.5 years ago

mikhail.shugay 3.5k

Hello!

I was reading about software licensing recently and found the GATK model quite interesting. In that case the general (API) libraries are released as fully open-access (free - open), while the executable application is released under academic-only license (free - restricted), with an additional for-profit package that includes a licensing fee. Now I have a question regarding how this model could be put in practice, as I've been told by the people at my department that there could be author affiliation-related issues.

Lets assume that a group of authors have published a paper describing a new promising bioinformatics software, released under academic-only license. After finding that the software had sparked some interest in commercial (e.g. pharma) companies, they've decided to form a start-up/firm and sell this software under a commercial (for - profit) license. Would there be any possible ownership issues from the institutes they've put as their affiliations during the publication of the paper? Or it is necessary to be affiliated with some sort of commercial company during publication?

Thank you in advance!

commercial affiliation software • 2.8k views

ADD COMMENT • link updated 2.2 years ago by Ram 43k • written 9.5 years ago by mikhail.shugay 3.5k

Ram · Accepted Answer · 2014-10-04

I believe that the authors of a software may change the license of their software at any time they wish to do so and may change it to any type of terms they want to.

The new license cannot be applied retroactively only going forward - and as such will only cover those parts of the software that are actually different, since the prior code has already been licensed.

In the case that you list the issue at hand is the ownership of the code not as much the licensing. Depending on the organization, the laws of the country and terms of employment in effect, the code written during the employment within an organization may partially or entirely belong to that organization. Thus the people that wrote the code may not actually own the code hence the right to change the license.

This of course depends on many factors and the exact details need to be determined for each individual case. I would guess that when it comes to academia it is much easier to work out a deal with the employer.

Ram · Accepted Answer · 2014-10-04

I agree with Istvan. At least in the US, it's customary (required, in my experience) for students and employees of academic institutions to sign away their intellectual property rights when beginning employment. As random examples: "MIT owns inventions made by its employees while working under a grant or contract to MIT or making significant use of MIT resources" (MIT's policy, here) or "All potentially patentable inventions conceived or first reduced to practice in whole or in part by members of the faculty or staff (including student employees) ... shall be disclosed on a timely basis to the University. Title to such inventions shall be assigned to the University, regardless of the source of funding" (Stanford's policy, here).

So in summary, the core issue is the official employment situation of the authors, rather than the affiliation they put when publishing papers (though of course those are often related).

If they wanted to start their own company, the usual route as I understand is to do that and then have the new company license the technology back from the university. Those negotiations can be complicated, and in some cases cause controversy (for fun side reading, see "Researchers Cry Foul on Ion Torrent License and Inventors' Rights"). The Broad example is a bit different because it appears to be wholly internal, but hopefully these tidbits address your more general question.