John Battelle's Searchblog

This drives me nuts, too, John... but more as in it pisses me off, since the publishers are so frigging clueless and/or greedy.

I dare the 'good guys' to completely purge their search indexes and include ONLY those sites that specifically opt-in. From a copyright perspective, this would at least make the naysayers less hypocritical. Either it's mandatory opt-in for including snippets of *ANY* copyright material, or it's not, end of story. Search engine snippets aren't any more or less fair use than book snippets.

And no, robots.txt is NOT opt-in.

Adam: you're being overly legalistic. The GPrint imbroglio is about a more fundamental issue of respect and fair-dealing from a company whose primary asset, as John has so frequently stated - - is trust.

As I posted elsewhere (I couldn't access Searchblog earlier today - - was that just me?):

If I am a writer, then Google and I are both, ultimately, in the IP business. I don't get to use Google's IP unless I agree to their Terms of Service. As an author, my TOS is the copyright line. Why is it ok for Google to ignore mine while insisting that I respect theirs?

To tie this all back into the Oodle/Craigslist debate for one neat package: compare and contrast Google's approach to GPrint with Craig Newmark's approach to the rollout of paid ads on Craigslist way-back-when. Craig's attitude was the embodiment of community-focus and fair-dealing. He basically said look, we need to generate some revenue here, this is what we're thinking about doing... but we recognize that our primary value comes from you, the community, so we'd like to get your feedback first on how to do this in the best possible way for the community.

It is precisely this attitude that has engendered such fierce loyalty from folks like myself that I will stay up until 4 in the morning writing a rant on some random weblog (this one, actually) to defend Craig's motives, even while having absolutely no real vested interest in the fight one way or the other. Google on the other hand said basically - we know what's best for you and the world, we're taking your stuff and repurposing it for ourselves. Don't like our approach? Too f-ing bad. See you in court. Yeah, yeah - I know they offered an opt out procedure. See my earlier posts for why I feel that was a bogus offer.

Further communication from Google management implied that anyone who didn't get with the program was either an obstructionist interfering with Google's altruistic plan to save the world or a benighted luddite who "just didn't get it."

To the contrary, it's precisely because I do get it that I am making such a ruckus; the precedent being set here and now will quite possibly define the parameters for the next hundred years of learning and creative expression - - so let's take a deep breath and try to get it right - - for everybody. As hard as it may be to believe, there are some people in the world creating value that don't happen to live inside the Googleplex. Their work deserves respect and the possibility of reward too.

Based on an extremely cursory review I did yesterday, and with the added disclaimer that I am neither a professional writer nor a lawyer - - OCA seems like an eminently reasonable solution: a clearinghouse approach where copyright holders and creators can define what they're willing to give away and what they're not, without having to be constantly scouring the web for new services "repurposing" their property.

---

Coda: I would also like to note that I predicted this (other search engines jumping in) a little under 48 hours ago on this very site. Admittedly, I thought it was going to take a little longer to be proven right, but I guess the world moves faster now than even I appreciated... :)

Why should we presume a copyright statement equals a TOS? When most of the in-copyright books in the libraries were published, no one had any idea what a TOS was. It's not quite as misleading a metaphor as comparing Google scans to car theft, but it's close.

Here's the main reason I support Google on the core opt-in vs. opt-out issue:

- Most books published afer 1923 are in-copyright.
- Most of those books have been out-of-print for a long time.
- In many cases it's not clear who owns the digital copyright for those books - publishers or authors.
- In many cases the authors are dead, or the publishers are out of business. The transaction costs of finding the copyright holder climb through the roof.
- Result: if you insist on an opt-in approach for all in-copyright content, you simply won't get most 20th-century content into the library. Everybody loses on that one.

When most of the in-copyright books in the libraries were published, no one had any idea what a TOS was

ok - and when "fair use" (the legal fig leaf Google is using to justify its actions) was conceived, no one had any idea what the internet was, much less GPrint - what's your point?

The analogy seems solid to me: I just picked up the first book on my desk and read the copyright statement. It says in part: "This publication is protected by copyright, and permission must be obtained from the publisher prior to any prohibited reproduction, storage in a retrieval system, or transmission by any means, electronic, mechanical..." In what respect is that not essentially a TOS for a printed work?

Re: your core concern, I don't completely disagree. That's why I said in one of my first posts on the GPrint topic here that I think a next-best solution might be a unified opt-out system, like the national "Do Not Call" registry, and mentioned again above the potential merit of a unified clearinghouse. With this in place, I think an approach that defaults to opt-out after a reasonable effort has been made to contact the copyright holder would probably be acceptable to most people. Furthermore, the combined resources of Microsoft, Yahoo, Google and the OCA would make it extraordinarily easy to publicize the plan (like they do sometimes with abandoned bank accounts) giving copyright holders a fair chance to come forward and decide if and how they want to participate.

I think that a compromise solution along those general lines would likely be seen as extremely fair to everyone involved, and probably meet with widespread support. It would almost certainly have mine.

In a TOS, the "service" involves interaction with the provider's physical property (e.g. Internet servers, or phone switching system, etc.), and that interaction is active and ongoing. With copyright, the copyright holder does not own the physical book, and there is no active, ongoing relationship between copyright holder and book owner. So, I think analogizing a copyright notice to a TOS is pretty sketchy. IMO, a closer analogy to TOS for the content world would be the "digital rights management" systems that are legally protected by the DMCA (wrongly, IMO, but that's another issue).

Regarding the idea of a unified clearinghouse for opting out or opting in, I think it sounds like a really good idea. Of course, the devil is always in the details, and the system would have to be mandated by Congress. I'm not holding my breath, but maybe I'll be surprised.

If scanning books from a library for
showing snippets on a search engine is
considered fair-use, then can anybody
that operates a searchable web-site
just walk into a library and start
scanning books ?
How does one distinguish a google from
anybody else who claims fair-use?
Is it that we trust google to do the right thing?
But its hard to do business purely based on trust
without legal protections

This is big and very exciting for many of us in the library community. For a take on how this is a start for the future, see our recent blog on the Open Content Alliance launch at http://hangingtogether.org/