Good morning, and welcome to the Library of Babel. Many of you will have heard of this great institution of learning only through the melancholy brochure produced for us by Jorge Luis Borges  in 1941. As, however, you will discover in the course of our tours, it is in the nature of this Library that the more closely one searches for the details of its foundation, the more complication one finds. The great logician Willard van Orman Quine  traces the founding concept to the psychologist Theodor Fechner  and notes contributions to the design by Kurt Lassiwitz, Theodor Wolff, George Gamow, and Willy Ley. There have even been attributions  to various ancient Hindu philosophers.
As Librarian, I am proud to announce the Library's new Mission Statement:
To make accessible to all the totality of human knowledge.
This is, to be sure, a challenging task, and an undenumberable series of further tours will be required to explore its different facets.
As Librarian, I am of course more intimately concerned with the information in the Library than with its history or physical structures. Historians and architects may thus protest that we are beginning our tours in the middle. Librarians will recognise that with this particular Library there is nowhere else to start.
At the time of Borges' brochure, our Mission Statement would have seemed impossibly ambitious. The Library had, it is true, an acquisition programme which, we believe uniquely, guaranteed absolutely that it would contain all human knowledge, and much else besides. But that brochure (probably the finest produced by any library) could not but reflect the misery of the then librarian, faced with the impossibility of the task of cataloguing.
We were immensely encouraged, however, by the emergence from the mid-1940s onwards of proposals for competing institutions. (I refer, of course, to the extraordinary prescience of Vannevar Bush and Ted Nelson.) And, from 1989, it seemed that the arrival of the World-Wide Web and then of its search engines  offered us salvation. We were no longer, it seemed, fated forever to footle in fallible catalogues. We could, it seemed, present our reference interviews directly to the Library:
(Alexandria NEAR hypocaust) AND NOT permanent
Try it.  You will find that our accession programme has some way to go before it fulfils its promise.
Today's tour will, from now on, concentrate on just one of the interesting issues which this this raises: and it will propose a solution which may startle or even shock many of you.
The Web search engines, in principle, offer us a new view of human knowledge and culture: and the more stupid the engine, the newer the view. Those which reproduce a more librarian-like experience involve human judgements: in the case of Yahoo, this seems to mean that the model user is concerned largely with financial information, and is perhaps a US citizen managing his own IRA (Individual Retirement Account) share portfolio. The stupid worms, on the other hand, simply index every word they can get hold of. A search for "East Timor" should, in principle, produce the official Indonesian government position  right next to an anti-occupation perspective  right next to all the newspaper articles.
But, whatever you seek in my Library, it is extremely unlikely that you will find any article from a newspaper. It is easy to verify that many newspapers are present in the Library: from the UK alone, the complete recent editions of the Daily Telegraph  and of the Times  are available; and they can be searched, but only by visiting their sites. If you do find a newspaper article with a Web search, it's probably a copyright violation on someone else's site - and this is a clue to the nature of the problem.
You won't find many professionally published books, either. Sample chapters, perhaps, and blurbs in their thousands are there.
This rather spoils the potential of the Library of Babel and, to be honest, it's putting the kybosh on our Mission Statement. Here we are, with the technology to put the whole of human knowledge at the disposal of the entire human race...
In fact, of course, we must not lose sight of the fact that the technology puts whatever knowledge we can muster only at the disposal of those members of the human race who possess:
These points will have to be covered in detail elsewhere. 
The obstacle which concerns us now is the fact that the more resources have been directly invested in producing a document, the less likely it is to be accessible through a Web search. You will find only what people are able and willing to give away for free, and what you will find therefore exhibits a distressing tendency to be worth every penny. It can be divided into three broad categories, in order of increasing value:
The point is that authoritative, well-researched, fact-checked, accessible, edited and attractively-presented material requires a very significant amount of human effort; and at present the only means we have of exchanging that effort at a distance are the tokens of stored labour which we call "money". The precise protocols we use in these exchanges have significant effects on where labour is applied; and the conventions which are all too rapidly spreading through the Web lead naturally to the locking up of significant chunks of our great Library of Babel in a myriad of closed collections.
This is a betrayal of the potential which the technology offers us and may yet turn out to be the biggest obstacle to the fulfilment of our Mission. The reason for this sad state of affairs is (I would say at my most charitable) a failure of imagination on the part of the Men in Suits who run newspapers and other publishing businesses.
And the way to remove this obstacle to the unhindered dissemination of human knowledge is to charge money for information.
The one reason that newspapers are invisible to the Web indexers is that publishers are hedging their bets on the elusive Internet Economic Model - so they are hiding. They are preserving the option of running their Web businesses in the two ways which most resemble what they already do: by charging subscriptions to access their product, and by selling advertising.
The subscription model leads publishers into elaborate coding to ensure that their site can be accessed only through a single gateway. At present, all but a few are offering their Web product as a loss-leader and an experiment. One exception is the Wall Street Journal, charging US$49 a year for access to its online edition  and reputed to be the only internet publishing operation making money. All those requirements to register and to remember dozens of irritating passwords are immediately useful for internal market research and may well represent a future revenue stream which no traditional, mashed-tree newspaper could dream of - selling the readers' names to direct marketing operations, sorted by what they read. But they are also the visible tip of an iceberg of a subscription mechanism.
The advertising model requires, to a less drastic extent, that readers be funnelled through the maximum number of the publishers' own pages before reaching the information they seek. Hence the legal action  threatened by by CNN, the Los Angeles Times, the Wall Street Journal and the Washington Post against TotalNews, which is in effect selling advertising space on a virtual news-stand which "surrounds" the publications and thus (the publishers can be expected to argue if the case goes to court) is a "ParaSite", detracting from the impact of their own advertising.
Consider, as an alternative, an economic model in which the online Wall Street Journal had no subscription fee, but instead charged readers by the article. The technology is all in place. It allows you, in principle, to have an "electronic purse" incorporated in your Web browser. You charge it up with actual money. You find yesterday's WSJ article on the death of Suharto in a a Web index; follow the link; and are asked whether you want to spend US$0.02 to read the article. You could well tell your purse not to bother you over amounts less than a nickel. No account numbers or passwords are needed; and, since you are dealing with electronic cash rather than an old-fashioned credit card payment, the transaction is anonymous and you won't be bothered by junk-mailers unwisely targetting you for holidays in Bali. It is possible that the WSJ would deliver the article in encrypted form, set to erase itself after a month; it is equally possible that the publisher would simply "sign" the article with the fact it had been licensed to you and rely on fear of the courts, rather than annoying technology, to prevent you re-selling it.
The only technical development missing is governments' agreement on the encryption needed for electronic cash and for "signing" documents. At present they're panicking, but there is a reasonable hope that they will calm down.
You thought you were visiting the Ultimate Library, not a nerds' convention on e-cash and encryption? You are.
If the WSJ paid the costs of good research, writing and editing this way, it would not be hiding its articles from the Catalogue of the Library of Babel®. It would be paying our Web worms to visit it. The Ultimate Library would become the Ultimate Newspaper (and TV channel and medium-yet-to-be-named). All Human Life would be Here.
Some sort of code of ethics might be required to prevent Web search engines selling top billing to the highest bidder. On the other hand, any search engine which did that would in short order find that some enterprising nerd had put up a ParaSite which submitted a search on a user's behalf and then re-sorted the results in descending date order. (Hmmm... not a bad idea, even under current circumstances... and, as of now, unpatentable.)
The pay-per view approach has other advantages. It removes the justification for publishers' increasingly desperate attempts to become outright owners of information - which aggrieved freelance journalists dub "superhighway robbery".  Before, freelances (and, on the mainland of Europe, staff writers and photographers too) owned their own work and could rent it to whom they pleased. Now they are surrounded by publishers saying "give us the freehold for the price of a month's rent, or never work again".
Pay-per-read makes it very, very simple for authors and publishers to share the risks and opportunities: you pay your $0.02 to to the publisher, and $0.01 of that goes to the author. This may seem of little concern to librarians - until you look at the possibility of Disney, Microsoft, Sony and Panasonic owning very significant chunks of the world's written culture outright. It can be argued  that pay-per-read will increase the diversity and quality of new content arriving in libraries.
But... what about the horrendous cost to actual libraries? The answer is very simple, though in urgent need of fleshing out with details: a Public Reading Right for electronic publications. This author is willing to accept a much smaller payment for readings done on the premises of a school or library than for those made in the homes of people who can afford to do so. The alternatives are for libraries everywhere to be saddled with paying electronic subscriptions without knowing whether users are going to use them - or for the content of our precious Library of Babel to be dominated by intrusive advertising.
Thank you for your attention. Please do not forget the collection box as you leave the Library.
 Borges, J. L. (1941) "The Library of Babel" in Ficciones. A text (of dubious legality, but let us assume fair use for critical purposes) is available at
http://jubal.westnet.com/hyperdis cordia/library_of_babel.html , together with a charming illustration of one of the Library's newer wings, designed in the late 1960s.
 Quine, W. V. O (1987), "Universal Library" in Quiddities, an intermittently philosophical dictionary, Harvard: Harvard. Probably: the attribution is taken from
http://jubal.westnet.com/hyperdi scordia/universal_library.html .
 Fechner, Gustav Theodor [Pseud.: Mises], 1801-1887. Pionier der Sozialwissenschaften; Ästhetiker - or so it says at
http://library.byu.edu/~rdh/prmss/e- g/fechner.html .
 Or so it says  at 
http://jubal.westnet.com/hyperdis cordia/akashic_records.html . This site turns up with distressing frequency in searches  on the subject. Honest, three separate searches led to it...
 It's lying. Probably. Like much of the ancient Hindu philosophy thrown around in the West, this bit was made up by Madam Blavatsky. Or so it is confidently asserted at
 The credibility of the site is corroded by its name, and somewhat further lessened by the fact that one of the "authorities" cited in reference  appears to be a pseudonym for a co-author of the Illuminatus books  . That's the Library of Babel for you...
 See, as one of the first examples to show up:
 The author declares an interest. Anyone replying, as owner of the words in which they do so, acquires an interest too. For details of the "superhighway robbery" see the Creators' Copyright Coalition pages at