Will libraries disappear?

A librarian proudly forwarded me this "10 reasons why" article about why conventional libraries are unlikely to disappear in the future.

While in general I would be inclined to agree, the article had me shaking my head in a few places. For instance:

Yes, we need the Internet, but in addition to all the scientific, medical, and historical information (when accurate), there is also a cesspool of waste. When young people aren’t getting their sex education off XXX-rated sites, they’re learning politics from the Freeman Web page, or race relations from Klan sites. There is no quality control on the Web, and there isn’t likely to be any.

First of all, putting the Freeman Center for Strategic Studies (a pro-Israel thinktank) on the same plane as the Ku Klux Klan and hardcore pornography would not reflect any political biases on the part of the writer, huh?

Secondly, it's clear that the writer never spent any amount of time in the more serious parts of the blogosphere, where relentless fact-checking is common currency.

Try reading an e-book reader for more than a half-hour. Headaches and eyestrain are the best results. Besides, if what you’re reading is more than two pages long, what do you do? Print it.

Well, buddy: I routinely read books on my Palm m515, which is actually gentler on the eyes than many printed books (particularly softcover editions with small print on yellowed paper). For straight text I prefer it over hardcopy, since I can read wherever and whenever I have spare time. Any material I print out for reading is liable to be scientific papers with oodles of tables and equations: and guess what, those I used to xerox from the journal so I could spread the pages out in front of me on my desk!

The great boon to libraries has been the digitization of journals. But full-text sites, while grand, aren’t always full. What you don’t know can hurt you: (a) articles on these sites are often missing, among other things, footnotes; (b) tables, graphs, and formulae do not often show up in a readable fashion (especially when printed); and (c) journal titles in a digitized package change regularly, often without warning

I guess I must be living in an alternate universe. All the scientific journals I read offer papers for download in PDF format, which is identical to the printed page. Many of them offer HTML versions of the paper as well, and there criticism (b) applies to some extent --- but thanks to two marvellous initiatives known as CrossRef and Digital Object Identifiers, the "references" thing is changing beyond recognition. In many online papers of recent vintage, you now not only have full footontes, but you can directly jump to the paper being referenced!

This is only one reason why scientific publishing is undergoing a metamorphosis right now. One publishing house (American Institute of Physics) is offering not just its usual journals, but also a number of "virtual journals" which aggregate, from their entire journal pool, articles that focus on a single (often interdisciplinary) subject.

And then there is the Open Access Initiative, which strives to offer freely accessible online editions of scholarly journals.

Indeed, when I was first introduced to the more advanced weblogging tools, I was struck by --- with some authentication and supervision to prevent crank-posting --- how adaptible this format could be to open peer-reviewing of scientific papers (preferably after they have undergone an anonyous reviewing cycle). There are enough issues with the reviewing process as it stands that this may actually be a change for the better.

The cost of having everything digitized is incredibly high, costing tens of millions of dollars just in copyright releases. And this buys only one virtual library at one university. Questia Media, the biggest such outfit, just spent $125 million digitizing 50,000 books released (but not to libraries!) in January [2001].

I don't know how much Amazon spent to digitize tens of thousands of books for which it is offering both full-text search and "look inside" features. But having some acquaintance with such matters, I would venture to say that the cost has dropped sharply since then --- especially if one is willing to outsource to "developing countries". And besides, it's a onetime cost.

Furthermore, online editions allow the presentation of material that cannot be presented in a book. A (bio)chemistry paper may, for instance, include manipulable 3D images of molecules or enzyme active sites, which the reader could rotate, zoom,... at will.

For reference works that obsolesce rapidly, a frequently updated online edition is probably a better investment than hardcopy.

I could go on for hours...

Comments

[dof]

> The cost of having everything digitized is incredibly high, costing tens of millions of dollars just in copyright releases.

The Distributed Playboy Digitizing Initiative, also known as usenet, proves that this need not always be the case.