Will mobility revolutionize Web 2.0?

On Monday, I moderated a panel at the Smartphone Summit, held in San Francisco as part of this week's Wireless IT & Entertainment conference. The main conference is organized by the CTIA (the main US cell phone trade association), while Monday's "Summit" was privately organized and heavily sponsored by leading mobile phone companies (notably Symbian).

The session I moderated was called "Smartphone Interactivity (Social Networking & Personal Communications)," one of two parallel sessions that closed out the summit. The claim of the program was that:

In today's world, MySpace, YouTube, Facebook, and many other social networking experiences have made their way into everyday life around the globe. The next natural extension of the social interactive experience is to the wireless smartphone device. With the capability to run audio and video along with GPS locating, there is no limit to the type of social interactive applications that could be deployed over smart mobile devices. Join the pioneers taking social networking wireless for an in-depth look at the implications of this technology, including how it may be monetized for ROI of those enabling these experiences, as well as security of the experience.

WIth only a few minutes to present, I focused on two goals. One was providing a preview of the study by my graduate students (Eduardo Sanchez and German Benitez) who are doing their theses on social media business models for the mobile phone industry. The other was to provide an overview of what's going on in the industry, based on what I know from working with Eduardo and German, reading the trade journals/websites and going to industry events (particularly last month's Mobile Monday event).

In my slides, I joked that the buzzwords "Social networking" or "social media" or "Web 2.0" seem to be used interchangeably, so it was a relief to hear that a morning panel had been unable to agree on a definition of "Web 2.0." (Tim O'Reilly claims to have invented the term "Web 2.0," but people seem to use his buzzword more than his definition.

I saw people taking pictures of my slides with a cameraphone (you can download them free from my website). I didn't think the slides were particularly insightful because I was rushed between two conferences and a backlog of grading (which I've been working to clear today).

I had a chance to attend a few earlier sessions, and what I heard confirmed most of what I'd prepared. "Social networking" (or "social media" or "Web 2.0") business models seem to be driven by two major trends. One is user-generated content (like this blog), and the other is taking advantage of the value created by direct network effects, i.e. the N x (N-1) possible linkages of a population of N users. (This is called Metcalfe's Law, but researchers last year showed that it grossly overstates the value of a network due to the long-known distribution of value via Zipf's law. In fact, this power law describes many of the interactions on the Internet.)

Obviously the strongest possible business models are those that combine both. While both Facebook and MySpace are about allowing friends to stay in touch, MySpace also plays a major role in spreading word-of-mouth for members' favorite music. (Jason Ling of MySpace was up from the fires in Los Angeles to speak on the panel, but his prepared slides were left behind when Delta lost his luggage.) However, in visiting a booth at the trade show Tuesday, someone showing Nokia's download site there is a difference between organizing around social networks like Facebook (I want to see people who I know) and around content as with Flickr or YouTube (I want to see a video of the latest politician's gaffe).

One of the unresolved questions regarding mobile social media is whether there will be mobile-only and PC-only networks, or whether the successful sites will support both. (I almost said "platform agnostic," but with 500 different phones out there, developers of mobile applications have to do a lot of work to be platform agnostic just within the mobile space.) Of course, it takes considerable work to turn a PC website into a good cellphone website, which is why everyone was excited to see how well Google Maps works on the iPhone.

I think such dual PC/mobile content drives a related trend that I picked up on at the Summit, which is the rising use of web-based applications. One reason is that developers need to span not only the PC and phones, but also the various platforms within each. The other reason for web-based apps is that it's finally practical: the iPhone has one of the first decent phone-based browsers, but Nokia (and others) are going to make sure that it's not the last.

This led to the only heated argument on our panel, in which Faraz Hoodbhoy (CEO of PixSense) said I was wrong. Perhaps I didn't make myself clear, or he wasn't listening clearly, as I never said web apps are the be-all or end-all. My point about mobile web apps is that they're a least common denominator which are getting more practical and will be good enough for many of the Google-, Yahoo- and MySpace-type applications that are already designed for them.

There are plenty of apps that require offline access, or low latency, or persistence, for which a native app will be a much better solution. There's many things you can do on the native Google Earth application that you can't do on Google Maps. I don't know what the relative mix of the two approaches will be -- and even web apps have to be tuned for the different form factors -- but I think mobile web apps are here to stay.

What's with "taxing" music downloads, eh?

Just back from my visit to Calgary for a conference.

In a decision Thursday that made headlines here in Canada but was largely ignored elsewhere in North America, a government body decided to retroactively impose a mandatory 3.1% royalty on legal music downloads to compensate songwriters and composers.

Among the very limited news coverage south of the 49th parallel, Slashdot proclaimed "Canada May Tax Legal Music Downloads," based on a secondhand report in Electronista (called "Canada to tax legal digital music downloads"). Any money collected under the coercive power of the state is a form of taxation. On the other hand, the money is not being paid to the government, but to SOCAN (Canada's answer to ASCAP).

In the only decent story on the subject, the Global and Mail notes that the new 3.1% royalty adds onto the 7.9% fee being made to music publishers. On a 99¢ (CDN) download, the label gets 60¢, these royalties run 11¢, the bandwidth is about 1¢ and the credit card companies (for small transactions) get 10-20¢. While the iTunes Store has set a 99¢/song ceiling on prices, one executive predicted these royalties could push pricing past one loonie (CDN 1.00). Canada also has a tax (or mandatory royalty) on blank CDs and music players to collect royalties for artists.

In some ways, that's double jeopardy because [if it's at all like the US] the songwriters were already getting compensated by the record labels out of that 60¢; however, I suspect (as on the US) the royalties were based on a physical good rather than an information good. When I did my “digital music" case 5 years ago, I estimated that 28% of the wholesale price of a CD was for manufacturing and distribution, which record labels used as an excuse to limit songwriter royalties. With digital downloads, everything but the royalties is gross profit: if the download volumes ever match physical ones (they won't), record company profits should be higher than they are today.

Despite this potential inequity, I think it's a bad idea to set royalty rates via government fiat. Exhibit A is the March 2007 decision by the equivalent US panel to increase record label payments by Internet radio stations (beyond those of other countries) to more than 100% of revenues. Over the centuries, the only accurate way of setting prices has been through negotiations between with multiple competing buyers and sellers. Whether or not the artists were getting a fair price beforehand, the actions of the US Copyright Royalty Board and now the Copyright Board of Canada have only reinforced that point.

Will the gPhone change the mobile industry's business model?

Blogging while attending a conference.

Rumors are that the long-rumored Google phone will be announced soon. A Fortune blog quotes the rumor that the announcement is coming next week (Oct. 24) while the blog of ZDNet's editor repeats the rumor that the announcement is Nov. 5.

So people are writing (if not reading) are rumors about rumors about rumors. The interest in the gPhone today is comparable to that on the iPhone 11 months ago, and exactly parallels the buildup to Apple's Jan. 9 unveiling. It's not clear whether this is because these firms are highly visible, because they're good at using secrecy, because the industry needs another entrant, or because bloggers just need something to write about.

The Fortune article quotes a securities analyst as saying 50,000 phones will ship this year, which seems very optimistic timing. Apple unveiled the iPhone in January (for June 29 availability) because details were going to leak in the FCC certification process. If the gPhone isn't being evaluated by the FCC today, it seems unlikely that it will pass by Dec. 15.

More interesting is the claim that HTC is (a) firm that's going to make gPhones. This seems a lot more plausible, because it knocks down (silly) speculation that Google would make gPhones. Also, HTC is quite plausible as a gPhone maker, since they are a generic Taiwanese ODM (trying to become a branded vendor) with US presence via Cingular (AT&T), and in fact is only known in the US because they have already used someone else's software (Windows Mobile) as a market entry strategy. But of course, if Google is a mobile software supplier, there will be many other vendors.

The main reason I raise this is not to speculate on release dates, features or partners, but because of one of the few new insights that I've seen on this topic over the past 6 months. This morning on Forbes.com, Brian Caulfield argues that the gPhone business model is far more important than its features -- because (like everything else Google does) it will presumably be ad-supported, and leverage Google's core competence in ad targeting.

Since Microsoft and Apple are targeting the high end, and Google (with open source) is targeting the lwo end, Caulfield's speculation is also that the gPhone will have the greatest impact on the developing world. That makes sense since the gPhone in the developed world, would be a substitute/replacement for existing cellphones, while in the developing world (India and China are oft-mentioned) is the only place left with any new adoption possible, not to mention the most cost-sensitive users who might be more willing to put up with ads.

This would be a quite a change for the mobile industry. A lot of people run around saying Asia is the leader of mobile innovation, with Europe second (except possibly for smartphones) and the US last. But "Asia" normally means Japan, as with NTT's i-mode or FOMA (UMTS), or the J-Phone creation of the cameraphone category. More recently it has also included Korea, with nearly universal adoption of 3G (via cdma2000 migration) and experiments in things like music downloads and mobile TV).

India and China were already the world leader in mobile phones as a substitute for wireline -- most countries adopted wireline then wireless but these countries (and others) are on a path to bypass wireline altogether. Could ad-supported mobile phones in India and China be the beginning of a global trend? In this case, it really would be an important step in Google's inexorable march towards Total World Domination.

On the other hand, it might be that Americans and Europeans and Japanese will be quite happy with their existing cellphone and business models and not interested in switching to an ad-based model. In that case, the gPhone would reflect a strategy for part of the world (and the most rapidly growing part) but an inevitable wave for the whole world. In that case, what influence will Google have on the mobile industry in the developed world?

Apple decides to ship UNIX(R)

Finished one course but still behind on grading.

Apple has finally decided to ship a UNIX operating system — its Leopard (OS X 10.5) shipping October is fully compliant with two Unix validation suites: the Open Group’s Single UNIX Specification (SUSv3) and the IEEE’s POSIX 1003.1.

Since March 2001, has been shipping OS X based on two Unix-derived operating systems — FreeBSD and Mach (which was mainly associated with NeXT). However, its rivals have been sniping at its claim of shipping a UNIX operating system because it had not (previously) passed compliance tests.

A decade or two ago, people fought for the rights to use the UNIX® trademark. POSIX was invented so people could legitimate UNIX clones without having to pay to use AT&T's code or trademark. The 4.4BSD-Lite release in 1994 marked the first full release of a UNIX-like operating system without AT&T code.

While POSIX provided technical interoperability, by the time standardization was completed (and the “Unix Wars” were over), Windows had achieved total world domination. Jim Isaak — chair of the POSIX committee for 15 years and recently a candidate for IEEE Computer Society president — wrote about all the dilemmas of POSIX standardization in a paper for my HICSS minitrack on IT standards. He published two slightly different versions of the story in the HICSS (IEEE) proceedings and a special issue of JITSR.

Inexorable march to domination

The WSJ this morning published the latest installment of the Googometer, showing Google’s market cap. At $190 billion, Google has passed HP, Coca-Cola, Intel, IBM and now Wal-Mart. Next up: Cisco. After that, SBC (aka AT&T) and Microsoft. GE and Exxon Mobile are still a ways off.

At one point, there was a satire article about Microsoft buying a small country. When Google’s stock goes up another 50% or so, that baton will pass to Google.