Ad-Hoc Networks - Got Your Ears On, Good Buddy?

When I was getting my J.D. (1994-1997), laptops in the classroom were not unheard of, but not by any means commonplace. By the time I went back for my LL.M. at Cardozo, laptops were pretty much de rigeur, and my initial insistence on taking notes by hand (I eventually gave in to the dark side, for various reasons) flagged me clearly as old-school, emphasis on "old."

That plethora of laptops exists outside of school, as well. I doubt I need to belabor the point to anyone reading this in a Starbucks, or most libraries, or any number of public spaces in your standard issue metropolitan area; just look around. And it goes pretty much without saying that the overwhelming majority of these laptops are equipped with WiFi connectivity.

Now, if you want to see the future of things, stop calling them laptops. Call them nodes.

There's lots of coolness tucked away in the internets, but at the core, one of the truly big ideas of the internet was the shift from a one-to-one model to the many-to-many model. This is true both on the technical, file transfer level - a single file going from point A to point Z will pass, in whole or in part, through many systems and connections besides the ones leading directly out of A or into Z - and on the level of content - in that theoretical Starbucks you're sitting in, you (and hopefully some others) are reading a blog, and the tattooed hipster sipping his latte next to you is writing a blog, and the barrista is wondering if her YouTube video's gotten any views, and that little girl who just came in with her father is planning her next MySpace post, etc.

In simple terms, any computer (technical level) or site (content) on the net is (or could be) having a lot of simultaneous conversations with a lot of other computers/users/sites/whatever.

That's all well and good. But the many-to-many model is only accurate a few steps away from the edge of that famous internet cloud we all know and love. (Much cooler image.) At the very edge, it's still a very limited conversation: your computer talks to your router, and your router talks to your ISP.

The idea of tweaking that model to fix (read: interconnect) the endpoints - mesh computing - isn't, strictly speaking, a new one. (In fact, one of the best features of the $100 Laptop is it's out-of-the-box mesh capability.) (click the little computers to prick up their ears, drag 'em to move them around.)

But I'm blogging it now because I just stumbled upon WiPeer. (There's a hat-tip that belongs here, but I don't remember who gets it.) I dropped it on my laptop, did a quick install, and within a few minutes I had my ears on. This was the first application that I've encountered that moves the mesh idea forward with such a low transaction cost on entry.

That is, I'm fairly tech-savvy, but even I'm a little chary of messing with my networking and wireless settings. Moreover, if I'm going to be a part of this experiment, I don't really want to sacrifice anything for it; I don't want it to interfere with my computing experience, my regular networking, etc. And I'm not installing anything with adware, spyware, annoying-ware, intrusive-ware, or really anything-ware. And I kind of don't want to think too hard about the installation process or settings, either.

WiPeer - with whom I am not affiliated in any way - seems to fit the bill. And while it's not open source, I can't imagine that the protocol is so abstruse as to prevent clever third parties from developing clever manipulations of the system. One of the first I can imagine is sharing an internet connection among meshed computers. (Go back and look at the one laptop per child demo, but keep the laptop on the far right (near the globe) "deaf" until the end)

But why am I noting this here? What's the IP import of the evolution of a true mesh?

Basically this is just a peek inside the can of worms. Think about how my laptop becoming a server changes the way the law looks at me. Am I an ISP? A (protected) internet router? an end-user? Do I deserve DMCA safe-harbor? If you're in that Starbucks right now, and paid the $10 for the day pass to use the T-Mobile wireless there, and then you mesh with a few other laptops there and they piggyback on your bandwidth... what if the aforementioned tattooed, latte-sipping, blogging hipster is (a) downloading music illegally, (b) uploading porn, (c) uploading kiddie-porn, (d) posting a crypto algorithm to an internationally available message board, (e) hacking into military systems?

What about file sharing within the local mesh? Take "Jack My IPod" to the next level. Or what about mesh-casting (I made that word up), where I can have music come through your computer instead of the radio on my shelf? What counts as "publishing" something now?

So that's the can of worms. And it's starting to open. It's going to be a mess, but it looks to be pretty cool, too.

--Ben D. Manevitz

[PostScript: It turns out that "meshcasting" isn't completely my neologism, but I don't think the results of the linked Google search (on 3/30/2007) demonstrates usage with the meaning I've given it.]

Viacom YouTube Brouhaha.

Everyone else is talking about the whole Viacom-YouTube brouhaha, so I thought I'd pipe up as well. There's been a lot of good analysis done, pretty much, but I want to chime in with some small tweaks to the analysis that I think have been missed.

Also, any excuse to use the word brouhaha.

Note/HatTip: A very fine summary of the complaint is available from GoogleWatch.

I've heard and read people out there arguing about the question of YouTube as a service provider under the DMCA, but I don't think a lot of time should be spent on that question. Whatever you want to say about the mechanics of how it goes about its business, YouTube isn't any different from Compuserve or AOL or Joe ISP, and it's exactly the sort of actor that the DMCA was designed to protect. Viacom may try to make the argument, but it seems spurious.

And so, Tweak the first:

Let's assume away an argument that YouTube isn't a service provider as imagined by the DMCA. Instead, Viacom's going to have to try to get Google/YouTube out of the safe harbor by other means. Most likely under 512(c)(1)(B), under which Viacom's going to claim that Google makes money from the infringing material and has the ability to control the activity.

I don't know if Google's ever taken an official position here, but I've heard it argued that Google keeps ads off the pages where the videos are actually displayed specifically to try to stay on the right side of this clause. And in this, it seems it's Google's turn to be silly; It's hard to countenance an argument that Google's not making money from content on a certain page just because there aren't any ads on that specific page when there's clear monetization of the property as a whole. Google's business is ads and eyeballs, and the videos - infringing or not - drive the eyeballs.

Instead, the question is going to turn on Google's ability to control, which will probably devolve onto an analysis of the state of technology. If Google has that ability, which is to say, if it's technically feasible for Google to control the uploading of infringing works, then it's possible that the statute means that Google has to do so. The judicial interpretation of "ability" should certainly include a feasibility/practicality aspect, but who knows how that's going to go. Google has started using a filtering technology, but - presumably - Viacom's not happy with that.

It's possible that Viacom's also trying to work on 512(c)(1)(A)(ii), claiming that Google is "aware of facts or circumstances from which infringing activity is apparent," but I think that the analysis comes down to the same inquiry; Taking Viacom's allegations as true for a minute - Google is aware that there are a lot of infringing works on YouTube - but even so, identifying infringing clips among the bajillions of clips on YouTube is a technical question.

There's another argument of Viacom's that I haven't seen getting a lot of attention. It's a bit technical, but might turn out to be the stealth claim that landsharks Google. Since I haven't seen any analysis on the claim... Tweak the second:

As nicely summarized in the aforementioned GoogleWatch post, Viacom claims that "YouTube is also deliberately interfering with copyright owners' ability to find copyrighted works" by limiting search returns to 1,000 hits. 512(i)(1)(B) limits the safe harbor to service providers that "accommodate[] and do[] not interfere with standard technical measures." Said technical measures are defined in 512(i)(2). It's not untenable for Viacom to argue that by limiting search results in the manner claimed, Google is interfering with such technical measures.

Google would counterargue that the limitation is necessary to avoid the imposition of "substantial costs on [Google] or substantial burdens on their systems or networks," as allowed in 512(i)(2)(C), but I don't know... seems a bit tetchy to me.

Finally, I just want to poke at the DMCA for a minute. A lot of the people with whom I find myself sharing coffee don't like the DMCA. But I actually think that it's not so bad; particularly in that it has two qualities admirable in a statute: (1) it actually comes close to doing what it set out to do and (2) it's at least pointed in the right direction, which is to say that what it sets out to do is what I think should be done with regard to an otherwise thorny problem.

{Edit} In response to the comment below, I clarify that I'm addressing here very specifically the mechanism for dealing with online copyright infringement. Other aspects of the DMCA, including the anticircumvention provisions, are subject to a separate analysis. {/Edit}

That tepid defense of the statute aside, there are some important weaknesses in it, and I think the Viacom/YouTube situation makes some of those weaknesses clear. And so, Tweak the third, which is not so much a tweak as a whinge.

Without going into the whole story, I should disclose that I have taken advantage of the DMCA notice and takedown provisions, in order to successfully (and quickly) stop a splogger from using (whole cloth) a post from my other (pseudonymous) blog. So I have first-hand experience of the DMCA working the way it's supposed to work.

On the other hand, the first phase of the dust-up between Viacom and YouTube demonstrates what I think is a significant weakness in the scheme as it stands now. Of the 100,000 take-down notices lodged with Google by Viacom, many (if not most) were probably overbroad. Say it ain't so, Viacom, but it seems that that they just did keyword searches and sent the notices. (Note: the link points to Just an Online Minute; the third response down is ostensibly from a Viacom rep denying the claim.)

Under 512(c)(3)(A)(v), the notification of claimed infringement must include a "statement that the complaining party has a good faith belief that use of the material in the manner complained of is not authorized by the copyright owner, its agent, or the law." Under 512(c)(3)(A)(vi), that statement is made under penalty of perjury.

But the "penalty of perjury" isn't really where the statute has any teeth. The real insurance against abuse (on this side of the equation) lives in 512(f)(1), which imposes liability for damages including attorney's fees on anyone who knowingly misrepresents that the material is infringing.

But those remedies don't extend to the knowing misrepresentation of other elements of the infringement notice - and I would argue that the remedies should so extend. Moreover, I think the DMCA should make clear that the "knowing" standard in question is an objective standard, which is to say that it includes "should-have-known."

It's a small change, but one that would, I think make the DMCA even better at getting to where it seems to be pointing.

--Ben D. Manevitz