linking back to brembs.net






My lab:
lab.png

A recent flurry of posts and discussions prompted me to summarize a connected set of issues that I haven't really seen covered simultaneously elsewhere. I don't doubt everyone's aware of the issues, I just haven't seen anyone make the case that all the following three issues are related and, ultimately, have the same cause and thus, the same potential solution.

The astonishing gall with which corporate publishers strangle research and teaching institutions worldwide, even in the face of drastic cuts to said institutions budget is not a white gauntlet but a fist in the face of every taxpayer, paired with a big, raised middle finger. And yet, little is happening to make sure not more tax funds are being siphoned into the pockets of corporate CEOs and their shareholders. As outrageous and infuriating as such corporate behavior is, it is but merely a symptom of a wider infrastructure crisis in the sciences we have brought upon ourselves. Two factors have critically contributed to this crisis:
  1. We outsourced the dissemination of our products to publishers
  2. In the course of #1, we have ceased to see our libraries as the obvious custodians of said products.
As a consequence, we now hardly know who is running our libraries, what services they provide or where they are physically located. One reason why publishers are allowed to strangle our libraries is because we don't see them (the libraries) as parts of our institutions any more. We don't seem to part of the same team any more. Any attack on libraries should be seen as an attack on us, meaning the entire institution.

What are the products libraries are supposed to be the custodians of?

To answer this question, one only needs to look at what we do. Experimental science (and the following will be, of course, different in different fields) generally proceeds today in three main steps:
  1. Experiments generate raw data
  2. Software helps evaluate the data
  3. The evaluated data is presented to the scientific community in a publication
Thus, today, scientists generate three types of 'products': data, software and publications. At a survey in 2007, there were more than 1300 databases containing raw data in the realm of biology alone. Many if not most of these databases face funding problems, as it is comparatively easy to obtain funding to start a new database project, but much harder to fund the maintenance of the database. Why is it not the most natural and obvious thing to store our data with our institutions? If we wouldn't donate billions every year to publishers, we would have more than enough funds to run all databases many times over.

To my knowledge, there is no central place where one can efficiently search for scientific software. We have deposited our software for a recent project on sourceforge. This site contains both the software for data acquisition and for evaluation. I imagine many other colleagues post their software on similar sites, if they make their software available at all. Why is it not the most natural and obvious thing to store our software with our institutions? If we wouldn't donate billions every year to publishers, we would have more than enough funds to run all software repositories many times over. On top of that, we'd be able to continuously modernize and update the resulting infrastructure.

To me, publishing is the most annoying component of my work. In fact, it's the deep frustration and anger associated with publishing our work which fuels my motivation to try and contribute at least a little to publishing reform. Why is it not the most natural and obvious thing to store our publications with our institutions? If we wouldn't donate billions every year to publishers, we would have more than enough funds to run all journals on this planet and have plenty of funds left for data and software repositories. On top of that, we'd be able to continuously modernize and update the resulting infrastructure and incorporate links between data, software and literature, such that one could, for instance click on a figure in a paper, specify different parameters and visualize the data differently than the authors did. This is technically trivial today and used, e.g. on the website of Reed Elsevier to visualize their financial developments. Why does Reed Elsevier use this technology on their website, but we can't use it in our publications?

On top of the infrastructure crisis we face with our dysfunctional publication system, we also face two more crises, those of data and software. Tragically, by outsourcing not just the service of publishing, but simultaneously also the rights to our products, we have paved the way for the two more modern crises. We don't see libraries as the obvious place for our products any more and this has proved to be a pernicious development. It is time we bring the fruits of our labor back into our control and create a thriving market for services around our products. All of our products have been funded by the taxpayer and must be freely accessible and re-usable for every taxpayer. It is of course more than just ethically justifiable to spend tax funds to increase the efficiency of this dissemination and re-use, specifically if the market is healthy and competitive. But the fruits of our labor themselves need to be fully under our own control and not under that of for-profit corporations, often with diametrically opposed interests from the general public that supported our work to begin with.

The general gist of this post was presented in my keynote address at this year's open access days in Vienna, Austria:

Apparently, Slideshare doesn't offer old embed codes any more...
Posted on Thursday 18 October 2012 - 21:28:48 comment: 0
{TAGS}


You must be logged in to make comments on this site - please log in, or if you are not registered click here to signup
Render time: 0.0856 sec, 0.0045 of that for queries.