Sign in or Join FriendFeed
FriendFeed is the easiest way to share online. Learn more »
Mackenzie Cowell
How much does an annual library subscription to the top 20 closed-access journals cost? Lets start a virtual library for non-scientists.
My guess would be on the order of $25-50,000 per year. The UC OSC maintains a curated list of price info for about 3000 of the "top scholarly journals" (http://osc.universityofcalifornia.edu/journal...). The average list price is about $1250/title (http://www.sennoma.net/main...), but I would expect the "top 20" to be more expensive than average. - Bill Hooker
Two potential problems: first, 20 journals doesn't get you very far -- there are more than 20,000 to choose from; and second, how do you determine the "top 20"? We have had many, many conversations around here on this topic... - Bill Hooker
I was niavely thinking we could pick based on eigenfactor ranking. Maybe we could afford many more than 20 - it all depends on how much the licenses could be negotiated for and the tension between number of subscribers and subscription cost. How much do alumni organization (that include access to the Uni library) subscriptions typically cost? - Mackenzie Cowell from iPhone
Why do scientists keep playing hostage to the copyright bandits that pistol-whipped their publicly-funded intellectual property from them to begin with? Let's consider the grey option, not too seriously, just as a fun literary idea: One could use existing hashes (pubmed ID, DOI, etc.) and paper databases (pubmed) to build a "swedish-hosted" bittorrent database of every article ever published and digitized as PDF. The nonextant hashtable entries (papers) could be given priority-ranks based on impact factor and age. Ten thousand greyhats behind firewalls (-most- university students) could populate the hashtable in the course of a few months/year and thwart any bot-centric countermeasures. PDFs are small, the core database could be kept alive forever once built. This story might happen by actors outside of america: a more accessible corpus is vital for "BRIC" and third world scientists, who get helped today through a patchwork of charitable proxies. - Anselm Levskaya
Anselm, good idea. We should focus from the beginning on empowering researchers from the BRIC. A successful international sci-pirate bay will be helpful in demonstrating the size and desire of this otherwise hidden market. Like how napster and bittorent forced the music industry to develop iTunes and Amazon MP3. - Mackenzie Cowell from iPhone
Anselm -- by "patchwork of charitable proxies" I assume you mean HINARI/OARE/AGORA, recently rebranded "research4life (http://www.research4life.org). They provide access to almost 9,000 journals to researchers in qualifying countries (http://www.research4life.org/Pages...). If someone were to try to put Mackenzie's idea into practice, that is a reasonable benchmark. - Bill Hooker
Bill -- HINARI/OARE/AGORA is charming, but the fact is that they provide access to third world scientists in countries so third world they don't have any science to speak of. Most of my experience with developing science is in Chile, a country that does -not- qualify for these free access points and in truth has very little general access to journals that is not provided by individuals extending proxies. Chile has a young and vibrant science community that could really use the help of ready access to science journals. The same goes for brazil, for argentina, for india, for any number of countries that aren't destitute but in great need of help in every way. (BRIC countries). I'm sorry to sound too critical, but this list and this program comes across as a PR stunt more than a real offer of international access to the scientific corpus. - Anselm Levskaya
I don't really disagree, Anselm; I was just using R4L as a benchmark for Mackenzie's idea. (Although I would add that even if a country doesn't have much research infrastructure it surely has some kind of health service, or at least some struggling doctors and nurses, for whom access to the primary health literature could mean a great deal.) - Bill Hooker
The bittorent idea is interesting, though blatantly illegal -- I could see a role for civil disobedience... Just some thinking out loud: you'd need more greyhats -- there are 19 million documents in PubMed alone. You would also need to withstand the full legal and technological might of a $5 billion/year industry, since what you propose would, if successful, destroy traditional toll-access publishing completely, so they'd come after you with every resource they have. You'd also have to worry about partial completion -- suppose you got 10 million documents into the database, what are publishers going to do? They'll take their stuff offline if they have to; you could end up making a large chunk of the existing literature virtually useless. Plus, you would pretty much destroy any commercial attempts to digitize back-catalogs... there are are lot of caveats to consider. - Bill Hooker
I like the pirate bay idea. A lot of people have a lot of PDFs stored for themselves already. I've been thinking along these lines for quite some time now, essentially since I started entering metadata into my Mendeley library. Now if one could get the PTP technology from PB and write an interface such that one could post your Mendeley library with metadata to the PTP network, one would have quite a sizable db to start with... - Björn Brembs
Bill does have a good point about shooting ourselves in the foot with back-catalog digitization efforts. Unless your army of disobediant first-world grad students is ready to spend some time with a scanner, this isn't a no-harm process. - Mr. Gunn
Can't we just rely on google for the back cataloging? I like the PDFbay Idea, although it's blatantly illegal, and I have no problem accessing every journal I could possibly want to read from my current university. There's also that pesky issue of PDF watermarking... - Brian Krueger - LabSpaces
Google can't index a book existing only in paper form. They do have a digitization program, but I don't know how much support is has. - Mr. Gunn
getarticles@googlegroups.com, run by http://contentliberation.com/, already demonstrates that at some scale, on-demand liberation of protected content works. We should think about how to build PDFbay on a foundation of on-demand liberation requests mediated by groups like getarticles. (I hear there is a similar friendfeed room?) - Mackenzie Cowell
Content Liberation seem to want to describe what they do as civil disobedience, but it isn't. Quite apart from the issues of unintended harm that MrG and I have pointed out, I am not down with simple lawbreaking. If you break a law because you think it's unjust, you should be willing to accept the unjust consequences, using the whole process as a means to challenge and change the law. *That's* civil disobedience. To apply that to access to research, you'd have to believe not only that toll-access publishing was morally wrong (and I think a case can be made on the basis of Peter Suber's scale argument), but that exposing that wrong would be sufficient to force change. Mr Swartz is hardly as sympathetic a figure as Rosa Parks -- nor, of course, does he have the same force of unambiguous justice behind his efforts. - Bill Hooker
The Refs Wanted room (http://friendfeed.com/referen...) operates deliberately and openly in the grey area between Fair Use and PR Nightmare for publishers who might want to prevent such activities. It's not clear that the room's activity is illegal, largely because it would be very difficult to demonstrate that it causes measurable harm to subscriptions. This grey status depends heavily on its peer-to-peer, as opposed to peer-to-many, distribution method -- sending one person a pdf is very different from making that pdf available free to all comers forever. It also depends heavily on inefficiency: it seems intuitively obvious that if all researchers everywhere could get any paper within a few hours just by posting a request to the FF room, that real damage to the subscription model would soon follow. - Bill Hooker
I gather that 2nd/3rd world distribution of article sets is already happening through the distribution of hard drives filled with indexed articles. 1.9E6*~5MB = 9.5TB. So it's not that big for modern storage media. A single drive would soon be able to hold it. I'd never recommend anyone in america or europe to post themselves up for legal annihilation by playing at revolutionary. But there are probably hundreds of thousands of tech-savvy students beyond the legal arm of these entities (in china and russia alone!). I pose this story more as a prediction of what we'll soon see unfold in the international scientific community. I'm guessing that corporate entities whose business models are predicated on coercive IP "rights" simply won't make it in the long run. - Anselm Levskaya