The obsession with Google and China was quickly replaced last week by an obsession with Google's earnings, as the numbers came in a bit lower than analysts expected and kicked the stock down about 10 percent.

Much of the concern seemed undue. The miss was due primarily to heavier-than-expected taxes, which was a result of complicated international allocations, and a gift to the Google Foundation; without these, Google came in about as expected.

Of course, because Google regularly exceeds expectations the real expectations were that it would do better than expected, which meant that merely meeting expectations was actually a failure to meet expectations. (Clear?)

The matter has been subject to high volume (in both senses - quantity and noise level) comment, conveniently linked on the Internet Stock Blog.

But there has been little reference to what may be the biggest imponderable concerning Google's long-term kingdom, power and glory, which is the degree to which its margins may depend on free riding on investment by others, and the various ways in which these others will fight back to share in the loot. (And the "may" should be emphasized in that sentence, because the answers are murky. This post is more query than conclusion.)

Three crucial inputs are involved: software, content, and connectivity.

Software

Google is a proud member of the Free and Open Source Software (FOSS) community, and the operating system that runs its thousands of computers is a variant of Linux. But not garden-variety Linux, because Google has tweaked it in various ways to incorporate its own complex search algorithms, which are the crown jewels of the company.

This tweaking can be kept secret because of a quirk in the General Public License, version 2, which governs Linux. GPLv2 requires anyone who modifies a program and then distributes the modified program to make the source code available to the public. Google, and other companies that modify Linux and then provide services over the Internet (referred to as Applications Service Providers, or ASPs) are not distributing the code, so they need not make it public.

Some members of the FOSS community regard this as a loophole; the ASPs are taking from the community without giving back. The GPL is now entering a year-long revision process, and one issue that will assuredly receive considerable attention is whether this loophole will be closed and the ASPs forced to make their modified code public.

It seems doubtful that Google would want to comply with any such requirement, so it could be on the edge of having to revamp its fundamental operating system, a task that cannot be routine even for as fine a hive of tech brains as Google. It might turn to Sun's Solaris, or it might even have to start paying for software, unless it wants to write an operating system of its own.

Content

Google's goal "is to build products that organize the world's information and make it accessible to our users." And, it should be added, to make money by selling ads designed to appeal to those interested in particular bits of that information.

It is a worthy goal, and a great business model, but the problem is that the world's information was produced by other people, who are not always happy at the thought of the Internet in general and Google in particular making their stuff available for free while the search engine reaps rewards. Even information producers who possess their own strong reasons for posting material, and who would willingly do so without recompense, may bridle at the thought of search engines profiting when they do not.

The grumpiness is particularly marked in the old mass media. Their immediate problem is not only that the Internet is using their content but that, paced by Google's sophisticated matching of searchers and advertisers, it is destroying their economic base of advertising. This looks irreversible, because the Internet has such massive cost advantages. A department store or home seller need not pay the costs of printing a million copies of a page of the Washington Post to reach the 10,000, or perhaps 10, people who might be interested in its ad. An auto maker need not pay $2.5 million for 30 seconds of Superbowl time to reach million of viewers who couldn't care less. It can reach people when they are actively searching for information about cars.

So the old media must find alternative revenue streams, and the only possibilities are to capture for themselves the advertising value of their content, or to put it behind walls of digital rights management and charge either the readers or the aggregators, such as search engines.

When the Internet exploded, a big backlog of content existed, ready to be put on the Net for trivial cost, and people shoveled it on. Google is reaping the benefits of this. But producing good content costs money, and as the backlog of worthy no-cost material declines, the problem of paying to produce future content will come to the fore.

The bottom line is that whoever has the revenue stream will ultimately be forced to pay. And that can put a dent in profits.

Connectivity

The third area in which Google might be vulnerable is in basic connectivity. The telecommunications providers claim that the Internet-dependent companies are not paying enough for their bandwidth use, and are working hard on mechanisms to charge more.

As with the legacy content, there is a legacy telecom dimension to this issue. The telecom boom and bust wiped out billions of dollars worth of investment in fiber, but not the fiber itself. Much of this remains dark. Prices are still low, and competition keeps them there. But not a lot of new trunk-line fiber is being laid, and there is a lily-pad issue. If, as some claim, we are only using 5 percent of present capacity, but that capacity is doubling every year, then it will be less than five years before the current glut turns into a shortage.

And of course both cable and telephone companies are building fiber to the premises as fast as they dare, an expensive enterprise in search of a good revenue model.

It is difficult to tell how this topic will develop, since the fight over what is called "net neutrality" is turning intense, but again it seems like a safe bet to conclude that there cannot be an indefinite separation of revenue streams from production. In one form or another, the companies that make money off the Net will wind up paying for the infrastructure.

* * * * *

So what does all this bode for the future of Google's earnings? Who knows? Much of it depends on Google, which needs to develop effective methods of sharing, as indeed it is doing in contexts such as Google Print for publishers or AdSense, in which Google gives big chunks of revenue to partners.

But were I a Google groupie, whether as investor or analyst, I would broaden my focus to include the business models of the content and connectivity providers, and developments in the GPLv3 debate. The one assumption that is insupportable is that the other players in this complex game will not react, and react strongly, to what they perceive as free riding.

posted by James DeLong @ 4:47 PM | Markets: Business, Investment & Innovation

Link to this Entry | Printer-Friendly | Email a Comment | Post a Comment(0)