When I buy eggs in the supermarket I always open the box to check the eggs are not cracked before I put them in my basket. When I buy a pair of shoes, I try them on first to see if they fit and look OK. Similarly when buying some items of clothing. It is generally accepted that before buying goods you can check the suitability of the goods.
Information is not like other goods though. If you open the box to read the information, then you have consumed the information already. This is a conundrum that makes buying and selling information resources a tricky business. Thinking about journal articles, how do you decide that you want to read an article? Perhaps it is because it has been cited, perhaps the article title tells you that it is interesting, perhaps the authors and the journal source help to confirm it. For many years that was all you had to go on, but over the last 40 years abstracts have become ever more readily available and they can sometimes give a stronger (but not infallible) clue as to whether or not an article is worth reading.
When online journals came along I thought that we would start to see publishers selling sections of articles – 20% of the full article price would get you the intro, 20% would give you the methods, 30% for the results and 30% for the discussion. You could sample a small part of the article for a charge less than the cost of the whole, in order to satisfy yourself that you really wanted all of it. But I was wrong. Publishers even now only sell complete articles, and actually they prefer to sell the complete journal or even large bundles of journals.
If you are in an institutional setting (e.g. University) you will probably have access to a wide range of subscribed journals and packages so this isn’t such a problem. If you are an independent practitioner, work in a small science-based company, or you just have an occasional need to access the scientific literature, then it is a problem. You have to buy a whole article (and these can cost as much as £30 online) on the basis of an interesting abstract.
A Californian company called DeepDyve are setting out to change this by offering article rentals for just 99 cents. Their press release sets out the rationale, suggesting that there are up to 50 million people in the USA who use the web for research and who might value the convenience and affordability of obtaining an article for 99 cents. In a subsequent blog post the company elaborates on this, saying that their audience is the non-institutional “knowledge worker” and suggesting that there could be 35-40 million of them in the USA, making a market of $2 – 4 billion.
The 99 cent price tag is a key feature, explicitly modelled on the iTunes pricing. DeepDyve believe that it is “a price which makes it morally convenient for users to purchase, not pirate“. Open Access articles can be viewed for free through the service. They also claim a database of 30 million articles, easily searchable. The one big snag is in the rental concept. You don’t get to save a pdf on to your computer, but you can only view an article through the DeepDyve software. Articles cannot be printed or shared.
DeepDyve’s announcement has set the library and publisher listservs and blogs alight. Phil Davis on the Scholarly Kitchen blog notes that some of the big publishers (Elsevier and Springer) are as yet not included in the DeepDyve service. But it is undoubtedly a most interesting entry to the market, and he suggests that “what the Big Deal did to the library’s relationship with publishers, DeepDyve may [do] to the reader’s relationship with libraries and publishers“.
David Crotty from CSHL Publishing, on Phil Davis’ comment thread, is sceptical and asks “can anyone explain why it would be in a publisher’s interest to split revenue with a third party?” Publishing guru Sally Morris, on the liblicense-l list says “I for one find this a very positive development, I just wonder why Google hasn’t done it first!”
Jill Emery, a librarian at UTexas, Austin, did some testing and found a few practical snags:
Many records marked free are just an abstract from PubMed
Many records are Preview Only meaning you can only buy them from the publisher.
Renting an article for $.99 gives 24-hour access in a flash player. You can never see more than ½ page at a time. In a fit of stubbornness I wanted to see if there was any way I could end up with the whole article saved. To do it I had to do screen shots of the article, 2 screen shots per page and the article I rented was a 31 page review . Needless to say I didn’t bother to do the whole article, but you could…
Searching is very clunky, does mysterious things with what seems to be a web search.
I think it is interesting, though perhaps too early to say if it will succeed. I think it is addressing a genuine need that has been largely overlooked until now. Whether the convenience of a cheap preview will outweigh the inconvenience of the clunky mechanism remains to be seen.
I guess there’s some rational behind the clunkiness1 – this service seems to be aimed at people who will only want to access a few articles, so won’t have to clunk through all those half pages very often. It’s also a strong deterrent to piracy (which Jill Emery seems to have discovered first hand).
Frank, can you briefly summarise how copyright issues will work out for the redistribution here? Surely DeepDyve have accounted for this somehow?
1 Firefox spellchecker suggests ‘funkiness’. I like it!
Mike – I think you’re right. Also, once someone’s had a taster of an article and decides that that they need it, they may want to purchase the pdf from the publisher.
As far as I understand, DeepDyve is only including in their rental service articles from publishers they have agreements with.
Generally contract law takes precedence over copyright law, so I think that the licence between the publisher and DeepDyve, and the licence between DeepDyve and the user will govern what users can do. The DeepDyve terms of service have lists of what you can and can’t do.
I like medical paper abstracts, where they put the results and methods right in there, so you know immediately whether the paper is worth getting. Scientific abstracts sometimes sound promising, but then once you find it you see that the promising bit was only one “maybe this could work in the future?” at the end of the discussion and not a result.
Interesting concept, and I admit it has a certain appeal as compared with the usual $25 or so that a journal publisher would normally charge to “purchase” an article.
But… the Flash player interface described by Jill Emery sounds irritating in the extreme. I for one would not use the service, just for that purpose. What if I wanted to read an article for a school assignment (say), and quote part of it? Would I have to do a screengrab, run it through OCR software, and then edit the resultant mess? Or open up a text editor window and shudder type out the text myself while reading from the DeepDyve Flash thingamajig?
On the other hand… what do you want for 99 cents, I guess.
Richard, I’d think you would have to go back to that old-fashioned activity, taking notes.
I think the exercise is an interesting one, but I won’t be buying stock anytime, soon. I wonder what the real revenues are from individual article purchase, anyhow? In all the many thousands (tens of thousands?) of articles I’ve perused over the years, I might have paid for one once. I’ve paid much more in photocopies in the distant past, and for photocopies of monographs from the distant past. Who is doing the kind of major individual-article purchasing for which DeepDyve is trying to take part of the market for itself?
Heather – in the DeepDyve blog post they go into more detail about how they have analysed the market and the need. In some ways it could be a race between this kind of service and Open Access. If OA delivers mass access anytime soon then DeepDyve is unnecessary. Until then, anyone outside of traditional knowledge-rich environments like Universities and Institutes may find DeepDyve helpful. Or they may not.
@Heather – quite my point. Write stuff down when I could copy and paste? I think we’re beyond that. Or we should be. Of course, if I read a “book” made of “paper”, I would have to take notes too, I suppose.
It’s a good idea.
I still have to manually type quotes sometimes, either from “real” books or old PDF’s that have been scanned into a repository – or from Google books.
This is much more the case in humanities than in science though, there’s a bigger market there I think.
A friend and I were talking recently about how it’s easier to search a Google version of a book than to look something up in the hard copy you’ve got. If only you could copy and paste bits if you could prove you owned a copy somehow.
Cobi – what you’re describing would be excellent…
I’m almost-certain that I’ve come across some old scanned PDFs where some kind of OCR has either been embedded in the file, or is being done on the fly somehow. Maybe I was hallucinating though.
Regarding Google Books, I find it a bit clunky (bit like Google Maps street view that way) – but very useful sometimes.
I’d think you would have to go back to that old-fashioned activity, taking notes
Quite. Whenever I’ve been in the research phase of writing a book and have had to scan a lot of journal articles fast, I’ve had to take notes – especially if the books concerned are bound journals in research libraries, which can’t be taken home. You get very good at eye-grabbing the essential points that way.
I like to take notes – rather than using a highlighter or lifting whole sentences – but I also am partial to bundling my notes with the article or conference abstract in question. Have recently enjoyed doing that on Mendeley. Endnote has a similar feature. It makes those notes searchable as well – and I lurve a competent search engine.
Hi, I’m the CEO of DeepDyve. Thank you for your comments, they are all extremely helpful and valuable. I thought I’d comment on a few of the many questions:
Re: Mike Fowler – we only make available content where we have a written agreements in place with the publisher. Our intent is to partner with both open and “closed” access publishers to make their information more discoverable (and monetizable for certain publishers) to this new, large audience of non-institutional users.
Re: Richard Wintle – we are using a Flash player to display the content. I also agree that our player has some shortcomings, most notably it doesn’t allow you to copy/paste. There is a reason for the madness. One key purpose for using the player is to protect the copyrights of our publisher partners by not allowing our users to rent, copy and paste the article for $0.99 and effectively bypass our publisher’s sites where they sell the article for considerably more.
Re: copying inconvenience – I agree that the flash player creates an inconvenience in note-taking. We are scratching our heads on how to support better ease of use without compromising copyright protection. That being said, while we fall short on this scenario, I believe we can add a lot of efficiency in other cases by allowing users to quickly and affordably search across many articles and subject areas, then read and sample the full-text of many items of interest immediately. Compare this to the current situation where there’s no convenient means of reading the full-text (other than purchasing the article first for $20+ or spending time searching for a free copy). Ory try doing an interlibrary loan where you have to search catalogs and wait. Our hope is that the low cost and high convenience will outweigh the ‘free’ choices available.
Our site is still beta so we are the first to admit that we have much work to do. Please keep us in mind and continue commenting on us so we can incorporate your feedback into our service. Thanks again, Bill.
Thanks, William. I totally get the point of the Flash interface (and not allowing copy/paste) – 99 cents, after all, is not what purchasing an article or book costs (as you note).