There are a lot of things to love about JSTOR for ejournals. It’s easy to search and has such a wealth of content that I find it easy to understand why for a lot of professors a while back it was synonymous with library ejournals. “My professor told me to search JSTOR,” students would tell me. And for research in many fields, it’s still not a bad place to start. There’s the small irritation of having to click to agree on their terms every time I want an article PDF, but at least when I do it works. And there’s the time I was trying to do an exhaustive literature search and JSTOR thought I was a bot of some kind and shut down the session, but I was able to bypass that in a couple of minutes. Overall, though, a great experience.
Then we get to JSTOR ebooks, and things change. In my Library Journal column on the mess of ebooks, I complained about JSTOR ebooks among others, because after a certain amount of friction trying to download an ebook chapter I simply gave up. It just wasn’t worth it. After that column, a rep called me and we talked about JSTOR ebooks and their many advantages, and they do have some advantages. However, when it comes to downloading, they make the 18 steps it takes to download an Ebrary ebook for the first time look almost appealing.
I decided to give it another try, though. The first time I tried was just an experiment. I didn’t want the book, I just wanted to test the service. Yesterday, I found a book I actually wanted to read, but the print copy was checked out and every copy in the Borrow Direct system was also checked out. The JSTOR ebook came up, because the book was published by the Princeton University Press and we buy the ebooks from PUP. I don’t want to read chapters in the Flash reader online, but If you don’t like the Flash reader, you can download the PDF, supposedly. According to the JSTOR rep I spoke with, that’s available on only 60% of the titles, but it was available on this one. It’s a long book, so I figured I’d download a chapter at a time and read through it when I got a chance.
I went to the page for the ebook. The first thing I noticed was the warning. “This book has viewing and download limits.” That’s for sure. “There is no printing or copying allowed,” because it seems like a good idea to take a potentially useful technology and make it impossible to do simple, basic tasks that everyone would expect it to do. Deliberate hobbled technology makes it unlikely I’ll invest in it, but the ebook was already paid for.
I could download a PDF of a book chapter, that is, if I logged in to my MyJSTOR account.
I didn’t have a MyJSTOR account, because I don’t want one. What I wanted to do is download some of the book the library paid for that says downloads are available. That doesn’t seem like much of a demand. Since completing that simple task was made impossible for me, I spent a few minutes creating an account I don’t want and shouldn’t need, filling in all the blanks with meaningless or wrong information. Then I logged in to the account I don’t want and shouldn’t need. Supposedly, now I can download the book.
Oh, but not yet. Replicating the outstanding JSTOR article platform would be far too harmful for the publishers, I assume, so I get some more friction. I need something called the FileOpen program, because JSTOR ebooks can’t just give me a PDF once the book is purchased and I’ve created this pointless account.
I was already pretty irritated, but what the hell. By then I was suffering from the “sunk costs fallacy,” where I’d invested enough time that I would feel bad giving up, even as my benefit-to-time ratio rapidly shrank. So I tried to load the plugin that’s only purpose seems to be to allow me to open a PDF that I should be able to open anyway if it hadn’t been screwed up by DRM or whatever they did to it. I couldn’t open it without the plugin, that’s for sure.
And finally, success! No, wait. Not success. Here’s what I got next:
If you can’t make that out, it reads, “Note to Safari users: Due to Apples’s updates and fixes it has become no longer possible to view PDF files in your Safari web browser. We apologize for this, and we hope to be able to restore this functionality in the future.”
That was really weird. First of all, I wasn’t using Safari, but Google Chrome. Second of all, here’s a screen shot of me viewing a PDF file in Safari a few minutes after I got that message.
Thus, the statement that it’s no longer possible to view PDFs in Safari was a lie. What it seems to mean is that they’ve added so much DRM to the PDF that it’s not viewable by standard web browsers like ordinary PDFs are. Let’s get clear who’s keeping me from viewing PDFs. It wasn’t because of changes that Apple made, or even Google. It’s because of changes to the PDF that JSTOR made. This turned into one of those “don’t pee on my head and tell me it’s raining” moments, only less messy.
I couldn’t view it in Chrome, either, so I went back to the note to Safari users and pretended it applied to me. “For now it seems that the best that can be done is to use Firefox together with stand-alone Adobe Reader or Acrobat.”
Really? That’s it? The best that can be done? Make me create an account, login to that account, install a plugin I shouldn’t need to read a PDF, fail to give me a PDF that I can read, and then tell me to go follow some special instructions and change browsers to view a file format I should be able to view with any standard browser. That’s the best that can be done?
No, that’s not the best that can be done. That’s a non-solution to a problem JSTOR created, no doubt at the behest of the publishers. The best that could have been done is having me click “Download this chapter” and then downloading the chapter. That’s the best that could have been done.Telling me my problem downloading a chapter was something other than their restricted file format isn’t tempting me to buy any JSTOR ebooks for the library. It did tempt me to write this blog post, though.
At JSTOR, we believe in offering broad access and delivering a great user experience. Over 19,000 of the books in the JSTOR program operate exactly like our journals do, not like you describe here. Every person at an academic institution has 24/7 online access, on campus and off, to these ebooks and we provide for unlimited concurrent use, unlimited DRM-free PDF downloads, and unlimited printing and copy/paste. There is no need for a plug-in, and no need to set up an account. JSTOR also supports interlibrary loan for ebook chapters from these books. This is the user experience that most people have of e-books on JSTOR and the one we expected most libraries and users to want.
The challenge is that publishers are not yet comfortable making all books available in this way, and so we have chosen to experiment with other models. Princeton University elected to try one of these other models – an ebook that can only be used by one person at a time and that carries DRM but is available at a lower price. We have considered this model an experiment from the start. Is has been a way that we could get publishers to make more books available and it is enabling us to understand whether libraries and users will tolerate some level of restricted access on an ebook for a lower price point. So we are learning together here, and appreciate your perspective.
Director, Books at JSTOR
In this case, JSTOR should include an icon or some words indicating that an ebook has DRM. And then, we can avoid them.
Lawrence, fortunately JSTOR has since removed DRM from their ebooks because of customer complaints. It’s platform is now as great as the ejournal platform.
Frank, thanks for responding. I agree we’re learning together here, and that’s one of the reasons I’ve started writing about these issues. For too long the movement for library ebooks has been guided by understandable (although I think unnecessary) hesitation by publishers and by less understandable but widespread complacency by librarians. The model JSTOR wants for all their ebooks is the same model I want, and when all or most scholarly ebooks are available that way I’ll finally make the switch to ebooks as a preferred method of buying books. That might take a very long time, or might not happen.
I was aware of the special PUP deal, but not aware that it involved experimental DRM. I’m not speaking for my institution here, but I think that’s a failed experiment. However, I decided to write about this specifically because of the notice implying that the reason I couldn’t see the PDF in Chrome was because Apple had made changes to Safari. That message was misleading, and it seemed to be blaming standard web browsers for not displaying content rather than the altered PDF that wouldn’t work. That’s not an Apple problem or a Google problem, but a problem with this particular type of DRM. If, while using a very popular web browser, the chapter had just downloaded after I went through all the steps I was supposed to, I wouldn’t have thought twice about it.
Part of the issue here–and it’s one I see often with e-books–is one of repeatedly creating and then foiling user expectations. I think most people would be all right with a multi step process if a) they’re aware at the outset what they need and b) have some sense that they’re making progress. Here you’re having obstacles constantly thrown in your path, like one of those fairy-tale quests where each step of the adventure yields yet another complicating factor. Not a good model.
I am in favor of “learning together” in a general sense in an educational environment but it’s not learning together when there’s a system that doesn’t work and one person is paying for it and one person is experimenting.
It is possible to try new things, but you have to explain the system that you are experimenting with, thank people for their patience as you work out the new system, and tell them to give you feedback on their user experiences with the new system as a way to actually determine what works and what doesn’t work.
I appreciate that this is new territory, but ebook vendors seem to have two simultaneous sets of customers who they are trying to satisfy and I know librarians would appreciate a lot more transparency about the processes that they are having to use and explain to users. At the very least, truth in error messages should be a baseline requirement.
We lend books to users at Open Library (one user at a time) using DRM that is significantly less onerous than this model. There is nothing inherent to DRM that makes this process this convoluted.
I appreciate and value the scholarly articles. Cheerful and keep on.