« November 2006 | Main | January 2007 »
December 30, 2006
AT&T Net Neutrality "Commitments" Unenforceable?
Before anyone gets too excited about AT&T's network neutrality commitments in connection with its acquisition of BellSouth, read this from the statement of Chairman Martin and Commissioner Tate, the two Republicans involved in the decision:
Importantly, however, while the Democrat Commissioners may have extracted
concessions from AT&T, they in no way bind future Commission action. Specifically, a
minority of Commissioners cannot alter Commission precedent or bind future
Commission decisions, policies, actions, or rules. Thus, to the extent that AT&T has, as
a business matter, determined to take certain actions, they are allowed to do so. There are
certain conditions, however, that are not self-effectuating or cannot be accomplished by
AT&T alone. To the extent Commission action is required to effectuate these conditions
as a policy going forward, we specifically do not support those aspects of the conditions
and will oppose such policies going forward.For example, today’s order does not mean that the Commision has adopted an additional
net neutrality principle. We continue to believe such a requirement is not necessary and
may impede infrastructure deployment. Thus, although AT&T may make a voluntary
business decision, it cannot dictate or bind government policy. Nor does this order.
In other words, from the perspective of at least two members of the Commission, AT&T is not under any network neutrality obligations, despite what yesterday's agreement says. (Add Republican Commissioner McDowell, who recused himself from the proceeding, and that would be a majority of the current FCC.) And of course, the agreement on its face applied only to AT&T, not other broadband providers such as Verizon and Comcast.
UPDATE: Further skeptical comments on the deal from Susan Crawford, Dave Burstein, and Tom Evslin.
Posted by Kevin Werbach at 2:51 PM | Comments (0)
December 29, 2006
AT&T Network Neutrality Commitment
After much wrangling, AT&T and BellSouth have agreed to network neutrality conditions in order to secure FCC approval of their merger. The key provision of the agreement, which lasts two years, is the following:
AT&T/BellSouth also commits that it will maintain a neutral network and neutral routing in its wireline broadband Internet access service. This commitment shall be satisfied by AT&T/BellSouth’s agreement not to provide or to sell to Internet content, application, or service providers, including those affiliated with AT&T/BellSouth, any service that privileges, degrades or prioritizes any packet transmitted over AT&T/BellSouth’s wireline broadband Internet access service based on its source, ownership or destination."
Tim Wu, a leading network neutrality advocate, praises the agreement, saying we may look back on it as a milestone in Internet history. It's certainly a tribute to the forces advocating network neutrality that AT&T agreed to something quite similar to the stronger network neutrality bills floated in Congress. The question, though, is what this really means.
I suspect one reason AT&T caved is that it realized that the "access tiering" network neutrality advocates are so incensed about wouldn't actually work. Or at least, AT&T won't have a service offering generating significant revenue from Internet content and application providers in the next two years.
When you look closely at the technical and business case for the kind of quality of service mechanisms that access tiering implies (as I do in a paper I'm working on), they become pretty shaky. What AT&T really wants is the ability to do is shape traffic on a per-application basis. And this agreement doesn't prevent that. Blocking or degrading all BitTorrent traffic, or all VOIP traffic, is permitted; just not degradation based on the specific company that offers the service. The practices that the network neutrality argument originally attacked -- blocking ports, limiting services like streaming video, and generally tilting the playing field for Internet applications -- are not prohibited under this agreement (which, it should be noted, applies only to AT&T). There is just a general commitment that AT&T will comport its business in line with the FCC's broadband policy statement. And the loophole in the agreement for IPTV, which isn't subject to the network neutrality provisions, could be significant if AT&T tries to stuff more services into that platform.
The agreement is certainly a victory for network neutrality. And overall, that's a good thing for the future of the Net. However, I have a feeling that, in the grand scheme of things, there is less here than meets the eye.
Posted by Kevin Werbach at 10:48 AM | Comments (0)
Creators vs. Consumers
Helen Cheng of Seriosity has an interesting post contrasting Second Life and World of Warcraft:
"Open ended virtual worlds like Second Life are best suited for the types of people who love to create content.... On the other hand, if you prefer to be a consumer of entertainment content, as I am, and as I suspect much of the world is, then games are your nectar of the gods."
There is so much excitement right now about user-generated content, that it's easy to make blanket assumptions about what people want to do. Most people are neither strictly creators nor strictly consumers; they are users. A user is a little of both.
Most of the time, a user wants a nice, prepackaged experience, but unlike a pure consumer, a user wants the ability to shape that experience when he or she feels like it. World of Warcraft does a magnificent job of creating the illusion of freedom within a fully-determined world, by giving users just enough room to manipulate things.
There's a need for the greater flexibility that something like Second Life offers, but most people don't want that flexibility most of the time. The minority who want it all the time just happen to be very vocal and visible, which is why we get stories about Digg being worth hundreds of millions of dollars and Wikisari challenging Google.
Second Life itself offers prepackaged experiences within the user-created world. With the growth of virtual worlds services companies like Electric Sheep and Millions of Us, I suspect we'll see more "consumer" experiences inside supposedly "user-created" worlds, even as next-generation massively multiplayer online games offer greater user freeom than World of Warcraft.
Posted by Kevin Werbach at 10:44 AM | Comments (0)
December 28, 2006
Vista, we hardly notice ye
You know, I was just realizing how small the buzz about Microsoft's new Vista operating system seems to be. There are the usual reviews and so forth, but no one seems to be particularly excited one way or the other. The non-story about Microsoft's PR company giving out review laptops is getting play in the blogosphere because there's not much else to talk about.
(Earth to bloggers: tech companies have been distributing review units to journalists, and analysts, and trade press, and user groups, and informal "influencers," and so on, forever. The mainstream media you love to diss are actually the only ones who tend to have ethics rules and transparency about this stuff. If you want the blogosphere to do better, spend some energy developing ethics standards and best practices, so that every blogger doesn't have to make up their own rules.)
Anyway, my point is that, whether it's a great OS or not, Vista just isn't that big a deal to most people in the tech world. And that's a story. I remember when Windows 95 can out. It dominated not just to formal news coverage but the informal discussions for months. After all, we're talking about a piece of software that, sooner or later, something like a billion people are going to use. Even when Microsoft just adds a feature Apple already had in OSX, it has a greater impact.
It's probably best for Microsoft to be relatively ignored (and I stress that this is relative). The company is going through a wrenching transition from software to Web-based services, even as it transitions away from Bill Gates' leadership and prepares for the day when Steve Ballmer hands over the CEO title. Let Google get the attention for a while, as Microsoft tries to execute its pivot.
Posted by Kevin Werbach at 12:19 PM | Comments (0)
SpamSoap update #1
As I mentioned, I'm trying out SpamSoap to cut down on the massive volume of spam I receive. Next week my free trial period ends, and I have to decide whether to start paying for the service at $25/month.
I'm planning to pay for it. After some tweaking and whitelisting of various newsletters and mailing lists, plus a bit of configuration on my hosting account to exclude spam routed around the SpamSoap filters, I'm getting excellent performance. Before anything gets downloaded to my machine, it first goes through SpamSoap as well as a separate gauntlet of filters maintained by my hosting provider.
Last week, 38 spams that made it through to my final layer of defense -- a Bayesian filter called SpamSieve that runs locally when I use my email program (Eudora). SpamSieve caught all but about 5 of those.
With approximately 5000 spams coming in per day, that gives me an effectiveness of 99.9% catching spam before it gets to my computer, and close to 99.99% overall. I'd call that pretty effective. (One nice feature of SpamSoap is that it gives me better reporting on what's hitting my mail server on the front end.)
The spammers keep coming up with new ways to evade filters. So, I expect performance to degrade over time. But at the moment, I'm happy. It's still amazing and frustrating that I could practically buy a new computer for what I'll spend this year just to filter spam. That's life in the digital age, I guess.
Posted by Kevin Werbach at 11:34 AM | Comments (0)
LA Times piece
The LA Times included me (along with smart folks like Steve Ballmer, Chris Anderson, and Rafat Ali) in a package of predictions about what might be next year's breakout online hit, ala MySpace and YouTube. Check it out for my answer....
Posted by Kevin Werbach at 11:31 AM | Comments (0)
Beyond Internet Video
Mark Cuban: "HDTV is the Internet video killer."
As usual, Mark makes some good points. There's more to the video market than just eyeballs. Advertising is what makes the media world go around. And while there are good advertising opportunities for Internet video services like YouTube, plenty of dollars that will go elsewhere.
Posted by Kevin Werbach at 10:55 AM
December 27, 2006
Check this prediction in November 2008
I think John Edwards will be the next President of the United States.
I'm not saying he's my personal favorite of the likely candidates; he's not, although he has many positives. I just think that, when all is said and done, he's in the best position to win the race. A lot can happen in two years, but that's the way it looks right now.
Posted by Kevin Werbach at 9:26 AM | Comments (0)
Everyone loves a good storyline
So, Wikia, the company started by Wikipedia co-founder Jimmy Wales, is launching a search engine, called Wikisari. I like Wikia -- its CEO, Gil Penchina, spoke at Supernova earlier this year. Leveraging the collective knowledge of users for what my friend and colleague Yochai Benkler calls "peer production" is a powerful technique, as Wikipedia, Digg, and other sites have shown. And I'm completely on board with the claim that there is plenty of room left for innovation and commercial success in Internet search. (One of the other panels at this year's Supernova was called, "The Future of Search.")
But why, oh why, do so many of the press reports on the Wikia story have to lead with something like, "Wikipedia founder challenges Google?"
Give me a break. Microsoft is challenging Google, and Yahoo! is challenging Google, because those are big companies that recognize the strategic importance of search to their businesses. Wikia isn't going to challenge Google. It's going to offer a service that competes for traffic against Google, and probably 50 other venture-backed search companies. Whether or not it succeeds will depend on quite a few things, few of which have anything to do with Google.
Search isn't an all or nothing game, and the sites with the best results aren't necessarily the most successful. Casting this as a battle over which search technique will "win" is just silly. If Jimmy Wales had opened a retail store, no one would be writing, "Wikipedia founder challenges Wal-Mart."
These mano-a-mano storylines are fun, but they aren't all that helpful for understanding the market. Back in the dotcom boom, there was a tremendous overemphasis on "business models." A great business model would get you funding, and maybe even an IPO. Well, guess what? An innovative model helps, but it still needs good execution and the right timing, among other things. That's the reality of business. The coverage of Wikisari suggests that lesson is being forgotten once again.
Posted by Kevin Werbach at 7:55 AM | Comments (0)
December 22, 2006
Name game
One of my Wharton colleagues just asked me what to call the scholarly field I'm in. It's a good question. My work -- and that of fellow academics like Susan Crawford, Phil Weiser, Christopher Yoo, Yochai Benkler, Barbara van Schewick, and Tim Wu -- sits somewhere between telecommunications law (focused on regulated networks) and cyberlaw (focused on intellectual property and other rules governing online activities).
One reason for the difficulty is that the underlying markets are in flux. Telecom and media industries are converging, and the regulated communications world is coming together with the historically unregulated Internet. That's the reason issues like network neutrality are arising in the first place.
I've taken to calling what I do "network infrastructure policy" -- the legal and associated issues governing digital communications and information platforms.
It's not elegant, but it's the best I've come up with. It gets at the notion that there is something distinctive about networks (which ties us to the many scholars in other areas studying networks and other complex systems), and something distinctive about infrastructure (as people such as Brett Frischman and Doc Searls have explained). But I'm open to other suggestions. Anyone?
Posted by Kevin Werbach at 9:34 AM | Comments (1)
Spam -- it's even worse than I thought
Daily spam volumes in November 2006 were roughly triple the level of just a year ago, according to Ironport, which makes email appliances. Yikes.
Posted by Kevin Werbach at 9:32 AM | Comments (0)
December 21, 2006
Implications of Traffic Shaping
Chart of ISPs blocking or, in most cases, restricting bandwidth to BitTorrent P2P file transfers. (Via the Wired News Monkey Bites column.)
So, are all these ISPs violating network neutrality? Or are they engaged in a legitimate and non-discriminatory effort to manage capacity on their networks?
Posted by Kevin Werbach at 3:29 PM | Comments (2)
Spam update
Over the past several months, there has been a major increase in the volume of email spam. Apparently, the spammers are using larger and more sophisticated "botnets," comprised of thousands of virus-compromised machines. They must also be coming up with better ways to avoid the Bayesian pattern-matching that most spam filters employ today, because I've found more and more spams getting through.
I've had the same primary email address (kevin@werbach.com) since 1995, with numerous aliases, and my address is widely distributed on the web. That's a recipe for spam. My best guess is that I now receive 5,000-10,000 spams per day. That means roughly 99% of my incoming mail is junk. Without good spam filtering, email would simply be useless for me.
For several years, I've employed a multi-layered anti-spam strategy. I have filters on the web hosting server where my account is located, and more filters running on my own machine. Despite this, the recent uptick in spam volume was starting to overwhelm my defenses. So I looked around for better alternatives, that wouldn't require too much of my time for configuration and installation. I'm unwilling to go to a challenge/response service, which blocks all mail unless authorized by me or the sender, because I'm worried about losing messages I might want to receive (from mailing lists, connected with my Supernova conference, etc.).
I'm now trying out a service called SpamSoap, a server-side spam filtering solution that intercepts mail as it comes into my ISP. It's one of the few providers of this type that offer a consumer service -- most others focus on businesses. With SpamSoap, I now have three layers of defense: before the mail hits my hosting server, at the hosting provider, and on my local machine. All three use a combination of techniques, including blacklists, greylisting, blackhole lists, content filters, and Bayesian classifiers.
So far, the results are good. Everything requires some tweaking, and the filtering will never be perfect. I'm pretty confident, though, that the false-positive rate is miniscule (especially for personal emails), and hardly any spams (no more than 1-2 per day) are able to run the gauntlet and make it to my inbox.
At $25/month, SpamSoap ain't cheap. It's odd to think that I'm paying more just to filter spam than I used to pay for my entire dial-up Internet account. On the other hand, email is now indispensable to my life, and to my business activities. $300/year doesn't seem unreasonable to make it work.
I'll see if the current good performance is transitory. I hope not!
Posted by Kevin Werbach at 9:16 AM | Comments (1)
December 20, 2006
The new Mac
Readers of this blog know that I had a suprisingly bad experience with the Mac Powerbook G4 I purchased last year. The crashes, incompatibilities, and odd behavior were just tolerable enough for me to keep the machine, because there are many things about the MacOS I still found preferable to Windows. I suspect the machine had a bad motherboard, although the experience suggested that Mac OSX, while good, isn't as bulletproof as it's made out to be. I was also mildly disappointed at the machine's performance, given that it was the fastest laptop Apple made, with all the options maxxed out.
So, last month, I purchased a new laptop. Glutton for punishment that I am, it's a MacBook Pro.
The results so far? Significantly better. I still have a few issues with peripherals and switching between my home and office setups. On the whole, though, these are orders of magnitude better than what I experienced with the last Powerbook. And the Intel Core2Duo is definitely a performance improvement over the G4.
One lesson I've learned is to approach the Mac the same way I approached my Windows machines. Assume that anything you do -- installing an application, adding a peripheral, updating your system software, changing a setting -- might totally screw up your system. Take appropriate precautions, and weigh the value of something new against the risks.
Posted by Kevin Werbach at 9:09 AM | Comments (0)