To: cni-copyright@cni.org

Subject: Re: Database Protection Legislation

Date: Wed, 12 Jun 1996 17:07:05 -0400

From: David Dailey <ddailey@hancock.cc.williams.edu>

By way of reaction to recent comments by Dennis Karjala, David Post,Jessica Litman, Dan Agin and others on this topic, the notion of what facts are and how they are used seems relevant.

One likes to think that facts bear some semblance to truth. Truth, as some sort of platonic ideal, begs of our sensibilities not to be levied out, weighed and sliced like fish. Hence, with Feist, one finds some sense of "justice" in the re-affirmation that facts are not meant to be copyrighted: doing so would actually impede rather than foster progress in the useful arts and sciences.

The sciences and the useful arts tend to rely rather heavily on "facts". In both, a fact is the outcome of some sort of methodological inquiry: The artist/scientist undertakes a quest (while adhering to the rules of the discipline of choice) and, in the end, discovers, via those rules, a fact. She then publishes it, not merely as the statement of a theorem, but with the evidence she used to amass it. The article = (argument + conclusion) is copyrighted, but the conclusion/fact is not. This seems to be a basic description of the portion of academic work that the constitution both recognizes and seeks to facilitate; mucking with this would be naughty, at best.

That others, D, should be able to cite the new fact that "author, A, has produced conclusion, B, via argument, C-as-summarized-by-D" is a basic tenet of the educational process. The new fact, as framed in the context provided by D, adds to "progress".

So, exactly what is a fact if I can't use it? Something gratifying perhaps on a personal level (like a revelation), but otherwise useless, as itself, in matters of commerce.

How might I use it if I cannot quote it? Well, I suppose I could allude to it -- oh yes,... "unquotable piece of information 1.0.3, which other licensees of the proper database can verify." As such, it is no longer a "fact", but has, for all practical purposes, become a trade secret.

Of what use is a database if all the individually useless things it contains are indeed secrets? (I actually know the answer to this, but am prevented, because of certain obligations from telling you.)

As with Section 120 of Title 17, if my creative energies have been so harnessed as to alter the public landscape (e.g., the building I designed has been constructed on a public street) then I cannot restrict others' rights to photograph it, as a part of the public landscape. So, it seems, it should be the case with nonmaterial constructions. The building, by virtue of its location, has become a fact -- a verifiable part of the context (be it material or conceptual) of public life.

It seems that by virtue of attempts to license facts, the would-be vendors may have so altered the concept of facts that what they are now seeking to sell are no longer facts, but data -- data so stripped of verifiable citation, that they are, for the purposes to which one ordinarily wishes to employ "facts," useless.

If the aggregation and arrangement of those facts into a database shows sufficient originality as to warrant a new copyright in the compilation, then why not just use existing statutes to protect one's effort? Why tamper with the public domain, the constitution, and the arts and sciences to try to own a bunch of stuff that the Supreme Court and common sense both concur do not belong to owners?

There has been some discussion of a requirement that others, in order to use the fact, must independently rediscover it, apart from its appearance in the database. It seems possible that a certain fact might have a unique methodology associated with its discovery.

Granted many (most?) facts do come with alternate paths of discovery: a phone company might compile a phone directory by examining its records of the numbers it assigned to its customers; or a third party might go door-to-door paying everyone $20 to reveal their phone number. Likewise, one might prove a theorem through a variety of substantially different approaches.

Yet there are likely to be, in the case of certain truths, canonical proofs, things for which parsimony has been maximized and for which all departures from the canonical proof (by way of embellishment) destroy the human's ability to comprehend. In such a case, the theorem as well as the proof are both facts, and both legitimately unprotectable (except on Neptune). Indeed, the harder one works to produce the "perfect utterance" the less copyrightable it becomes.

David Dailey (ddailey@williams.edu)


A collection of random postings to cni-copyright during the years 1994-2004. Humor and dread.