Eric.

1) If we disaggregate the activities here into a “million little pieces,” many of those are bound to lose the fair use battle. It doesn’t seem likely that a court is going to sympathize with, say, someone’s desire to put together a collection of song samples or web articles on their computer. They’ll just say “get a license;” see, the CCC will only charge you $40 for a limited right to copy a web article.

2) If the Grokster litigation had been done in this way, perhaps I’d see a reason for looking at Google as proxy, too. But it wasn’t–as Glynn Lunney’s brief points out, the court never bothered to look at the alleged infringements on a “micro,” fine-grained level. To get to the secondary liability issues, it had to just assume that a lot of what was going on was illegal. And to avoid the issue of what constitutes “potential for substantial noninfringing use,” it had to whip up the inducement standard, focusing on the mens rea (?) of Grokster, rather than the type of empirical analysis done by the Sony district court.

3) In the end, I don’t think we should “disaggregate” here and look at individual acts. What Google is doing is of world-historical importance, as Mary Sue Coleman suggests here:

http://www.washingtonpost.com/wp-dyn/content/article/2005/10/21/AR2005102101451.html

It’s sad that we haven’t put aside gov’t resources to do this kind of cataloging and archiving, but we haven’t. I wish we had, and then such an archive would be in an even stronger position, akin to that of the NLM in its suit with Williams & Wilkins. BUt even if it lacks status as a public utility, the synergy here is clear and enormous…much more is going on than the sum of the parts.

Certainly, search engines manage their databases, and there’s no question that profit is a keen motivator. But s.e. co.s don’t try to manage consumers’ (searchers’) *desires,* do they? (If they do, I’d like to know how. Ads? But see my example, below.)

Here’s an example that I included in a draft of the post but deleted for simplicity’s sake. It illustrates my response to the point you raise: In some First Amendment contexts, news media are plausibly viewed as agents for “citizens,” and they (the media) get abundant First Amendment protection as a result. Yet media institutions actively manage their content, using agendas of their own design, and many of them are just as profit-oriented as any search engine (though few of them are as profitable as Google). (If you want to extend the “manage consumers’ desires” point, see the literature on agenda-setting, and of course consider advertising in traditional media.)

*If* you’re inclined to accept my First Amendment example, then what’s the difference between a news-organization-as-agent and Google-as-agent? Do the differences between IP jurisprudence and First Amendment jurisprudence justify the different treatments of the agency inquiry? Personally, I don’t see a difference grounded solely in IP v. FA. But if not that, then what? Maybe society should value citizen inquiry into the kinds of political issues that media cover (in theory), as a FA value, over the kinds of consumer inquiry that we might assume drives search engine use, as an IP value, so media are protected FA agents while search engines are unprotected IP agents. Different welfare metrics, as it were. See John Ely’s work. Again, I don’t buy it (cf. Neil Netanel), but maybe there’s something there to explore.

Note that I’m not making a positive claim that Google *is* an agent in any or all IP contexts; I’m merely raising the possibility and suggesting that the question deserves some careful exploration.

Mike