[open-science] "open science" definition?

Thu Oct 9 18:00:45 UTC 2014

Tom Roche Tue, 07 Oct 2014 14:09:29 -0400 [1]
>> is there a definition of open *science*, suitable for similar reference use [as the "Open Definition"]?

Jenny Molloy Thu, 9 Oct 2014 15:43:18 +0100 [2]
> There are probably as many definitions of open science as people you ask,

Perhaps, but

1. your response addresses only the empirical aspect of reference, not the normative.
2. multiplicity can "be part of the problem."

An example:

Presumably listizens all believe that open science is good, and seek to further its adoption. We can do this via positive or negative reinforcement. Positively, one might seek to reward those practicing open science. "Nobel prizes" being presumably beyond reach at present :-) one might pursue, e.g., badging. Badging without any sort of quality-control (e.g., some sort of certification process) is obviously suboptimal, but, again, is probably all that is feasible for all but the most-well-funded groups. Still, with some sort of definition (or, at least, checklist--definition by example), one could at least provide some sort of sanction on fraudulent self-certification. (E.g., public shaming: "You gave your project an open-science badge, but you lack this, that, and the other thing.") But a badge backed with neither certification or definition is not only meaningless, but seems abuse-prone: cf. the use of "natural" applied to food[3].

Similarly, I suspect that partial or convenient definitions of "open science" will be developed either purely for profit (see this photo[4] of "members of the Springer behavioral sciences team [who] wore advertisements for new journal policies"[5]), or to neuter it--to "embrace, extend, and extinguish" open science[6].

> the major complication is that the Open Definition addresses
> outputs, be they data, publications, code or other knowledge
> 'objects' to which licences can be applied.

> Open science as used by many people also encompasses open process -
> openness to participation and transparency at different steps of the
> discovery process and therefore also addresses issues such as inclusion,
> modes of collaboration, democratisation of science and a host of other
> factors which are really difficult to define.

... which is an excellent reason to exclude such factors, until such point as they are meaningfully definable.

Every day, scientists (and the better sorts of philosophers) exclude "factors which are really difficult to define," refuse to "allow the perfect to be the enemy of the good," and model what can be modeled. As relevant factors become definable, we include them; but, all too often, that which cannot be defined is found to be meaningless[7].

FWIW, Tom Roche <Tom_Roche at pobox.com>

[1] https://lists.okfn.org/pipermail/open-science/20141007/003535.html
[2] https://lists.okfn.org/pipermail/open-science/20141009/003541.html
[3] http://www.nytimes.com/2013/12/20/business/trade-group-seeks-natural-label-on-modified-food.html
[4] http://www.slate.com/content/dam/slate/articles/health_and_science/science/2014/07/140730_SCI_ReplicateScientists.jpg.CROP.original-original.jpg
[5] http://www.slate.com/articles/health_and_science/science/2014/07/replication_controversy_in_psychology_bullying_file_drawer_effect_blog_posts.single.html
[6] Note that there are already enemies at least of replication[5] and open access, practices that I suspect most listizens would consider components of open science.
[7] Cf. Wittgenstein: "Wovon mann nicht sprechen kann, darüber muss man schweigen."