There’s been a slew of posts and other news items which convince me that we are at a discontinuity in the way science is communicated and valued. I’m too busy hacking Chem4Word (150 units tests work, 300 to go) to do it justice. But I’ll mention two:
-
A very thorough and compelling analysis By Michael Nielsen of why conventional science publishers will crash http://michaelnielsen.org/blog/?p=629. Not “whether” but “why”. “When” is more difficult but I hope we start seeing it in the next two years – the publishing industry in its current form is increasingly seen to be harmful to science and the technology is there that will allow people to vote by action, not words. Take time to read it. If you are a publisher, change what you are doing today or the world will change it tomorrow
-
Beyond Institutional Repositories
Laurent Romary
INRIAChris Armbruster
Research Network 1989; Max Planck Society-
in http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1425692
This argues that Institutional repositories have essentially failed, certainly for science and that we must adopt an Open model (most IRs are obsessively closed with copyright Mordor drooling over the portals). We must also adopt national and supernational models. To the average scientist this makes sense – why look for science in a thousand empty IRs when PubMed and arXiv aggregate the disciplines in Open fashion.
I will argue this when I have more time, but here are some predictions for scientists. Let’s look at this in 5 years.
-
Institutional resources (libraries and repositories) will be irrelevant to scientists. A new business model (possibly through national libraries) will provide an infrastructure.
-
Commercial scientific abstracting services (Chemical Abstracts) will be seen to be obsolete compared with the WebOfOpenData.
-
Scientific information will be managed in an Open manner by Learned Societies which are seen to be acting for the benefit of their community (and not for their income). That’s a hard business model to predict but it will happen. In the UK it’s Wellcome Trust, EBI and the British Library
-
A few publishers will be involved in managing science quality and metrics for a reasonable income. They may not be the current publishers.
-
Science will be authored with communal tools (hence Chem4Word :-)) and researchers will use memex-like infrastructure (e.g. Google Wave or whatever overtakes it).
-
Current Open Access models (“Green and gold”) will have played their part in history because authors will publish their papers Openly.
There is a great opportunity for universities to reclaim their commons.
-
They will unfortunately fail to grasp it, some some new business model will take a critical role in managing academic scientific information.
Now back to C#. Glyn Moody and Richard Stallman say I shouldn’t be doing it. One benefit is that it forces a constant refactoring of the abstract design – we have made a lot of progress in CML as a result of the project. I feel an urge to refactor JUMBO yet again…
-
wow, mentioned in the same breath as RMS – I’m clearly moving up in the world…
And yes, as far as C# is concerned: *stop* it, now, or I’ll send Richard around to sort you out.
@Glyn I shan’t I’m afraid. It’s a Faustian bargain but since I don’t believe in hell I can live with myself here for a bit.
NOOOOOOOOOO!
Thought about IronPython? Can access all those .NET dlls too… And a shameless plug here for the fact that you can access OpenBabel from any .NET language (subject to GPL of course).
@baoilleach yes we have though I don’t think we’ve had huge amounts of experience. I’m still wary of cross-language distribs – InChI and OpenBabel cause more installation problems than everything else together. Personally I think we have to work towards a BO solution where the algorithms converge and we have offerings for all main platforms.
Thanks for the link, Peter.
Your point about how passive Universities are is interesting. Many seem to simply follow the money, without much independent guiding vision or values. Since funding in most countries is quite centralized, the natural result is homogeneity and lack of innovation; the Universities mostly reflect the vision of the central funding agencies. I guess the flipside is that the big grant agencies are in a position to drive innovation if they wish. In this vein, it’ll be interesting to see what the long-run consequences of the NIH and similar open access policies is.
@Michael – yes. I’m also concerned about the absurd gearing between research funders and publishers – the latter provide the metrics which drive the former and the metrics are under their control.
Peter, Much as I admire the work of colleagues at the BL, what does it have to do with OA? The BL, as part of the establishment, is too timid for ‘open’, although I think Lynne Brindley is slowly changing that.
On publishers moving to control metrics, yes it’s happening http://info.spotlight.scival.com/
Like you, I look forward to the time when OA is not an issue and just happens, but I’m not sure we’ll get there in five years with this prescription.
Steve
@Steve Hitchcock thanks. The point about national is that institutions haven’t and I suspect are not able to manage repositories ion scientific disciplines. Whether the BL can address it I don’t know but we have to have organizations who can pull partners together.
The control of metrics by publishers is very frightening.
@pmr: “InChI and OpenBabel cause more installation problems than everything else together.”. I can’t speak for InChI, but I’m not aware of any installation problems with OpenBabel. If you find the installation instructions confusing or that they don’t work, please let us know…
See: “Beyond Romary & Armbruster On Institutional Repositories”
http://openaccess.eprints.org/index.php?/archives/606-guid.html
Voir: “Beyond Romary & Armbruster On Institutional Repositories”
http://openaccess.eprints.org/index.php?/archives/606-giud.html
10: The problem isn’t with OpenBabel or with InChI specifically, any platform specific code that requires installation would cause the same problems in distribution and installation.