Steve Harnad on the difference between open access to code, to text, and to data: an update

I had the recent opportunity to be involved in an email exchange with very interesting people involved in the open access and free software/open source movements, initiated by Richard Poynder.

Steven Harnad, a well-known open access and open content activist and practitioner, wrote a wrap-up on his position as to why, despite the commonalities, it is preferable to keep in mind the separate goals of the 2 movements, i.e. not to forget their differences.

We featured Steve Harnad in an earlier contribution on this theme, see here, and click also on the comments for an update by Steve.

Here is his very cogent explanation of why it is better to view it that way:

OA, OS, FS & CC share commonalities but also have some differences: there are genuine convergences and genuine divergences. Forcing them all into a single procrustean mold does them all a disservice. I agree that they should not do one another harm, and they do not and will not. But it would also do a kind of harm to conflate things that are logically and practically distinct. And it doesn’t work, in concrete terms, no matter how much abstract discourse it may generate:

Here is the heart of the matter. Let me try to convey it purely by analogy first: I am personally in favour of open-code pharmacology (“OP”): The formula for potential cures should not be kept secret, or prevented from being used to sell or even give away the medicine.

It does *not* follow from this, however, that if a commercial pharmaceutical company develops a non-OP cure for AIDS today that I will refuse to use it or promote it! Nor will I try to suppress or refuse to cooperate with OP research or OP researchers, while there are still diseases and patients, needing to be cured now.

To take this allegory further: OA is like Medecins Sans Frontieres (MSF): All those doctors have already dedicated themselves to providing their

  • services* for free, but it does not follow from that that they should

only use, or only collaborate with OP drugs and OP suppliers. They should use whatever they can get, and favor OP whenever they can do so, *but not at the expense of the immediate needs of their patients*.

That’s how it is with OA and OS/FS/CC: I am personally 100% in favour of OA, FS, and CC for my own output (though I think I shall write little software in this lifetime!) and for the output of others of a like mind on OA/FS/CC. But I would be against imposing OS/FS/CC (or OP) by fiat, for fear of losing potential creativity and productivity. (This is a point on which otherwise like-minded people can and do disagree.)

And the immediate situation for OA is this: 100% OA is within reach now. It needs OAI-compliant OA Institutional Repository (IR) software. I would personally prefer if all OA IR software (indeed all software) were GPL, but I would not mandate it, nor would I refuse the immediate help of non-GPL software in reaching 100% OA, any more than an MSF doctor would refuse the immediate help of a non-OP drug for treating AIDS, now.

(Maximizing research impact is not as pressing a problem as saving people from AIDS; so this analogy, like all analogies, has its limitations. But it’s not a bad idea to ask ourselves quite explicitly whether we take research so lightly that we agree with the above reasoning for AIDS, but not for research.)”

Leave A Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.