Yes, it's all about ME, me the user, me the person at the other end of the connection. I'm also not just any user, I am an advocate of libraries, a librarian, and I made the effort to go to HathiTrust -- a site that has not shown up for me in search engines.
This seems to be such a basic lesson that I do not understand why libraries can't learn it. User-friendliness.
Ooof! It just gets worse. I decided to see what login is about. To get to login, you have to search, select a book, and click on login. On the book page, you may see that a book is "Public Domain" or it may say "Public Domain, Google-digitized". When you log in, you log in either as someone from a member institution or a guest. The guest log in form states:
Does NOT provide access to full PDF downloads of public domain & open access items where not publicly availableHowever, it turns out that it DOES provide access to PD books (see comment by anonymous) if the book is not digitized by Google -- but that isn't what you've been told. "... not publicly available" isn't what you see on the book page, you see "Google-digitized." The page on policies has two different categories, "Open Access" and "Open Access, Google-digitized." Nothing in the definitions of those categories mentions member and guest downloading.
Basically, HathiTrust turns out to be a tiered system with member and non-member access. You don't encounter this until you try to download something that is PD but not "publicly available." Nothing on the home page mentions that this is a member-based service, therefore you don't know that as a non-member you will encounter walls.
OK, it is resolved, that from now on I will always go first to the Open Library, a site where Open means what I think it should.
Your friends at Google are the ones who digitized it. Hathi Trust requires a log in for Google-digitized volumes. Other volumes are available as a full PDF without logging in (here's a different Dewey digitized by Microsoft, apparently):
ReplyDeletehttp://babel.hathitrust.org/cgi/pt?id=yale.39002003927341
Google's getting you coming and going.
HathiTrust has told me this is because of their contract with Google -- they're not allowed to share those google-digitized PDFs except with HathiTrust members, by their Google contract.
ReplyDeleteOne could question whether umich should have signed such a contract, or whether they could have negotiated harder for a better one, or what.
But either way, umich/HT tech team gets credit for, once under these terms, not taking the easy way out and thinking "Well, since MOST of our copies are from Google..." and instead making sure their infrastructure DOES give you access to the PDFs that they are allowed to share, even without logging in.
I think HT used to not even show those PDFs at all if you weren't logged in, which was a problem for when HT affiliate patrons got there, no advertisement whatsoever that such a thing might be available if they were logged in.
Either way, yeah, access controls in general are one of the biggest barriers to good user interface for e-content. (certainly not the only one, and others are our fault)
Jonathan, just to make it clear, I am not questioning the value of HathiTrust. I am having trouble with the site as a user coming in with no underlying knowledge of the rules and from outside the set of member libraries. A home page with search boxes and no mention of access restrictions doesn't appropriately inform the visitor. In fact, my visitor experience was not a good one. I compared this to JSTOR's home page and they do a bit better. They have a login link on the home page (a hint that logging in my be relevant to the access) and a notice that says:
ReplyDelete"You are not currently logged in through a participating institution or individual account. See the login page for more information."
That's all I ask. It's that simple. And that is so much better than a user hitting an unexpected wall at the point that they've found something they really, really want to download.
Why is it important? Because that user will then turn to Google (which is actually what the HathiTrust documentation suggests) and will never return. OK, that's not a great loss, perhaps, but I think this is yet another "libraries as a negative experience" that we just do not need.
I agree they could do better, it's been very slowly and steadily improving, but there's still room for lots of improvement.
ReplyDeleteIronically (actually, this has nothing to do with irony, it's something else) " that user will then turn to Google" is of course exactly the intention of the Google contract that prohbits HathiTrust from sharing these PDFs with the unauthenticated public. That's exactly the intention, right? But yeah, HT still has some things they could do to amelioerate.