Clay Shirky

Notes from the opening address at CALI. The comments in italics are mine.

Clay Shirky: Data is the new book. The change is a slow process. Things get broken for a long time and stay broken for a long time. When the printing press was invented the old systems broke for a long time before we figured out how to use it.

The library as canary - users are opting out and libraries have to deal with that.

First reaction: STOP! There is a threat of de-coherence. Often we try first to bring users back to the old model. Change is driven by a market model - 10% change can be enough to change the way we do business. The classic mode of dealing with this is driven by the impacted institution. "Eat More Lard" - A message from the lard council.

Common theme is often that convenience drives changes. The printed bible is inferior to a hand printed bible in every way except convenience. This is what the RIAA is facing. Users tolerated studio distribution when it was a service that we couldn't do without. Now there are more convenient ways of getting access to media.

There is tension between wanting what you do to be highly prized and simultaneously to be scarce.

The second reaction is careful transition. Librarians are very good at creating taxonomies.

Question to self: what will be the way for classification of data

Consider chess, bicycling and basketball - two are games, two are sports, etc. Ontologies are poor at at classification of ideas and are much better at classifying things. For a long time ideas have been presented as things - the physical book. Our concern with ontologies is a 500-year patch to a problem - where to put books. Full text search means indexes go away. URLs make the need for classification schemes go away. Yahoo! started as a classification system. They did what people were comfortable with. When the data is is data what is the need for a classification system?

This is the problem I have with paper filing systems - too many different classifications for the information

Ontologies have consistently declined in use. Hidden Problems exist in several arenas

  • FCC: Spectrum is scarce
  • RIAA: Music is hard to distribute - somebody has to know what people will like
  • Colleges and Universities: We have more books than you do. You have to come to where we are to talk in small groups

All of these principles are coming under pressure.

In the future we will be judged more by how we take advantage of the flexibility we have not the way we watch the current system break.

Audience questions

Is closed source the library of 2004?

Other notes: Users are much more accepting of access to raw data. Lexis-Nexis and West are the "new church" of the printing press. Lawyers are rapidly becoming scarce more for their credentials than their information. Taking an idea like Page-Rank and allow users to specify their own nominations of contextual experts. Google is not the be-all and end-all of the access to information, it is just a departure from the previous way of classification of information.

Users are simply opting out.

What is the implication for Lexis/West of the way that RIAA users have opted out of the system they were given when another option became available. Will there be open source models for the sharing of legal information. When this collides with the idea that the scarcity that a lawyer provides is information and moves towards the scarcity of what lawyers providing being the bar admission they have, not the information they have.