Sunday, August 19, 2007

A Rare Misstep from Google?

We've always been big Google lovers. We think the way they structured their business is brilliant: create a graduate-school like atmosphere. Serve gourmet meals. Embrace open source. Make everything simple and fun.

Paul Graham, a geek much smarter than either of us, pointed out how Google's clean, simple and mostly white space look was not so much an aesthetic choice as a recognition, years ahead of its time, that mobile would become an important part of how people access the Internet, and so it would be smart to create a look that would work well on tiny mobile screens.

It's rare for us to think ill of Google. But we believe the company has made a big mistake in how it's handling its Adsense program, and how it's dealing with clickfraud.

Much of Google's revenues (which topped $1 billion in a single quarter) come from its programs to help advertisers large and small place text ads on Web sites large and small. It works like this: advertisers buy space for text ads on Web sites that relate to their product or service. Owners of Web sites sign up to run these ads on their pages. Ever time a viewer clicks one of the ads, the advertiser is charged, and a small sum is paid to the Web site owner's account. (The difference between the two, needless to say, is how Google makes its own profits.)

Simple, and brilliant--except for the phenomenon of clickfraud, in which nefarious users manipulate the system for their own ends. For instance if you want to harm one of your competitors, you can click its ad over and over, thus running up its Adsense bill without the benefit of a possible customer actually looking at the ad.

A more common, and worrisome, form of clickfraud involves web site owners using either humans or bots to repeatedly click ads on their own sites so as to generate Adsense payments. This practice has led advertisers to sue Google (and Yahoo, which has a similar program), claiming they were paying for fraudulent clicks.

Google negotiated a favorable settlement in the lawsuit, but it also cracked down--hard--on clickfraud. In the process, it has alienated many of its smaller business partners who carry the Adsense ads, by summarily closing their accounts--and keeping any owed money not yet paid to them--without explaining why it was doing so. Site owners are informed that because of the confidentiality of Google's algorithms, they can't be told exactly what they did wrong, only that it appears they've engaged in fraudulent behavior. That's that. And don't even think about coming back.

Though the web site owners are given no information, a tiny bit more can be gleaned from Google's site directed at advertisers, which explains (with a helpful graphic) that "invalid" clicks may not necessarily be fraudulent clicks, that Google is casting a wide net for these invalid clicks and doesn't mind getting false positives. Although I couldn't find anything that said so on the site, a PC World story says that something as simple as having a user click the same ad twice (which can easily happen anytime someone's computer or Internet connection is slow to respond) is considered an invalid click from Google's point of view. Given rules like these, it's easy to see how almost every innocent site might be getting its Adsense account closed.

And that seems to be what's happening, to web sites medium sized and small, including the popular British gay site pinknews.co.uk, and our friend, Rocketeers author Michael Belfiore. Supposedly, there's an appeal process, but when even a publisher with a respectable quantity of hits every month can't get anything beyond e-mail repetition that we're closing your account but won't tell you why, and you won't be able to open another one, it seems unlikely that anyone with the authority to do anything actually is reviewing the appeals. The last threat, by the way, isn't true: pinknews.co.uk had no trouble creating a new account for itself after it was told it wouldn't be able to.

It's obvious why Google is taking these draconian measures: for fear of losing its advertisers (who, are, after all, its only paying customers), not to mention running afoul of another lawsuit. The question is, how badly does it need the other element in the equation, the web site owners? Many of them have decided that, all in all, they'd rather do business with Yahoo! Especially since Yahoo! promises to provide a live human being to talk to if you run into a problem. Google has not given any indication that it minds sending these thousands, if not hundreds of thousands of content providers away in such a way that they're virtually certain never to return. And, from a short-term point of view, that strategy may be right. There are millions of fish in the Web site owner sea. Who cares if some of them are peeved; they'll be easy enough to replace.

But there's also a long-term view. Google thrives on its nice-guy image. All the way through its rise to search engine dominance, through a supposedly overpriced IPO that went nowhere but up, even through being included in the S&P 500 (!), it's retained its image of a bunch of nice Stanford kids sitting around making cool stuff. A major corporation that puts its software in the hands of the little people.

Shutting the little people out without explaining why doesn't help that nice-guy image. Keeping the money you promised them for ads that have already run on their sites doesn't seem like a nice thing to do at all. It seems like the arrogant behavior of a major corporation. The class-action lawsuit that seems like a no-brainer response to that move will probably make Google look even less like a nice guy than it already does.

So maybe there was something to lose here, after all. And maybe, for once, Google was too big and too full of its own power to recognize it.

Friday, August 10, 2007

Don't Blame Tech!

If you're a geek and you lived through the boom and bust of the dot-coms in the late 1990s and early 2000s, it must be nice to know...this time no one is blaming you.

The culprit this time is the most bricks-and-mortar of industries, real estate. Real estate spent most of the last decade in a delusional state that was quite reminiscent of the dot-com boom: things were going to keep going up and up forever. Real estate prices would rise and never fall. Remember when people though that about the Nasdaq?

You could "flip" houses. You could borrow at interest only and count on rising property value to pay off your loan. It was almost impossible to lose money in real estate.

It turns out it was possible, and also highly possible to get in over your head. Especially in an environment where all it takes, pretty much, is a pulse to qualify for a home loan. Where lenders practically follow you down the street.

Countrywide, the nation's largest lender, announced the subprime meltdown would threaten its financial position, and, according to CNN, that 20 percent of its borrowers are late on their payments. This is the kind of thing that tends to make markets go nuts, and they have been, yesterday and today, probably saved from imminent disaster when the Fed, and its counterpart the European Central Bank, poured over $200 billion combined into the markets to try and build investor confidence.

OK, it seems that real estate has come home to roost as other bubbles have before it. But keep a couple of things in mind. It's easy to forget this, but just three weeks ago on July 19, the Dow actually hit its highest point ever. And, despite a bad couple of days, it actually finished this week in the black.

By the way, just in case anyone thinks the oversupply of credit is a myth, or isn't still with us, here is the text of three of the five ads accompanying the Forbes article about the meltdown:

"Refinance at 5.35 Fixed
Get $300,000 loan for $875/month. Calculate Your New Payment. Act Now!"

"Refinance Rates at 3.0%
$150,000 loan for $391/month - refinance, home equity and purchase."

"Discover Card Application
0% Intro APR & Up to 5% Cashback. Apply Online: 100% Fraud Protection."

Wednesday, August 08, 2007

The Ad Campaign and the Grinch

Is it me?

Nokia's new viral ad campaign for its N95 centers around Internet warnings of jealous laptops viciously attacking their owners when the computer spotted the Nokia phone and responded with unbridled rage.

The site comes with warnings as to what to do in the event of a computer attack, offers of camouflaged ringtones (so the eavesdropping computer won't know it's a Nokia) and even the facade of a coffee cup you can use to disguise your phone so the computer won't see it.

I keep reading about this on blogs whose authors find the whole thing hilarious. Somehow the hilarity is lost on me.

This may offer a lesson about people-influencing and expectations. Remember the 2000 presidential debates? Al Gore was a famously effective debater, while George W. Bush was known to bumble on his feet. Expectations for Gore's dominance over Bush rose so high in the days before the debate that all Bush had to do was avoid tripping over his shoelaces (which he did) to be perceived as the winner (which he was).

I first heard of the thing via a blog post all about how incredibly hysterical it was. So I had high expectations when I followed the link to the jealous computer Web site. Maybe the big buildup made the whole thing seem anticlimactic. Maybe if I'd come upon it some other way, with no particular expectations, I would have been ROTFL like everyone else.

The odd thing about all this is, the product being touted is the Nokia 95, a mobile phone designed, so far as I can tell, to engender jealousy in iPhones, more than in computers. If I were a laptop and feeling threatened, I'd be more likely to attack my owner if he or she came home with the Nokia 800 Internet Tablet, which, unlike any phone or other mobile device I know of, browses the real Internet, rather than the mobile 'net, and because it runs on open source, has infinite potential for new applications to be created for it as well.

Tuesday, August 07, 2007

Lenovo Thinkpad - Now With Linux

Lenovo, makers of the formerly IBM Thinkpad, have just announced at the 2007 LinuxWorld Conference and Expo that they'll now be offering Linux operating systems on new Thinkpad T60p laptops, one of their leading core business models. According to this Reuters article, the 3rd largest PC maker will be shipping Thinkpads pre-installed with Novell's Suse Linux Enterprise Desktop 10 (SLED 10), one of the leading Linux distros. Slated to go on sale later this year, they will be sold to Lenovo's business customers as well as to consumers.

This isn't the first PC company to offer Linux as an alternative to Microsoft Windows. Just this past May, Dell began offering laptops and desktops loaded with Ubuntu 7.04 Linux, one of the most popular distros. Evidently, even Michael Dell himself uses Ubuntu on his home laptop. Dell is not new to Linux - as far back as 1999, Dell has been offering Red Hat Linux on it's Enterprise systems, and thousands are now running corporate websites.

It's a long, slow process, but it seems that finally the general public is beginning to like the idea of an alternative to Microsoft Windows. I think that some of this new push for Linux is fueled by the release of MS's new Vista operating system, which has been panned by reviewers here, and here, and rejected by many users. There's even a BadVista campaign to get people to avoid buying it.

My suggested solution? Time to move to Linux! I use XP on my desktop, mostly because I need to know how it works so I can repair customers computers, but my two laptops now run versions of Linux. One is running MEPIS Linux, and the other is now loaded with PCLinuxOS, both excellent alternatives to Windows. There are several other Linux distros that would make the transition fairly easy for migrating Microsoft users. Xandros 4.0, LinuxMint, DreamLinux, and Linspire all offer an easy learning curve for former Windows users crossing over to Linux.

Take the plunge, and dump Windows! You'll feel better about yourself in the morning.

The Inscrutable Future

The new novel, Spook Country, is set in the recent past.

There would be nothing unusual about that, except that its author is William Gibson, who, for more than 20 years, has made his living and his reptuation writing books about the near future. Gibson is most famous for inventing the cyberpunk movement and the term "cyberspace."

Why the switch? In part, Gibson told Silicon.com, because the future is getting too hard to see. "The trouble is there are enough crazy factors and wild cards on the table now that I can't convince myself of where a future might be in 10 to 15 years," he said. He also expressed sympathy for those in what he terms "corporate futurism," by which I think he means suits charged with making decisions today based on their analysis of what might happen tomorrow.

Pity me: I'm in that position myself. One of my roles in life is as board member of the American Society of Journalists and Authors (ASJA), an association of freelance writers and book authors which next year celebrates its 60th anniversary. The board decided it was time for some long-range planning, so I foolishly volunteered to chair a task force devoted to peering into the future and recommending plans for our next five years.

If there's one thing we know about the next five years, it's that we writers don't know how they will affect how we do what we do. Twelve years ago, getting paid regularly for writing for Web sites was an near-unheard-of activity. Eight years ago, those of us writing for the Internet had more offers of work than we knew what to do with. Neither of those things is true any more.

Around then, I wrote a big article for now-defunct Office.com about search engine optimization. By which I meant, Yahoo, Lycos, Alta Vista. There was no such thing as Google. Now Google is how I find everything, not to mention how I get my e-mail, which might as easily by on my handheld or my mobile phone as on an actual computer.

Six or seven years ago, there was practically no such thing as offshore outsourcing. Now, competition from India in particular is no longer just a concern for the poor unfortunate geeks, they're coming after our jobs as well. The Web daily Pasadena Now made news in May when it announced plans to have journalists in India watch Pasadena's government meetings streamed over the Internet and then write news reports about them. The publisher says he's hired two reporters to write 15 articles a week, each, at a total annual cost of $20,800. That's less than $14 per article, even if he gives his new reporters two weeks of paid vacation.

That's a very scary number for anyone who makes her living by writing.

Then there's InDesign, which some writers report they're expected to know how to use, so they can do their own production work, and Internet streaming video, which some Web sites are using to supplement their print articles, and which the writers are supposed to know about too. Not to mention blogging, which more and more writers I know are paid to do, another thing that was on no one's radar a few years ago...and...

Help! How the heck do we figure out what the future will look like for writers? And will it be a pretty picture if we do?

Here's what Gibson has to say: "Most societal change now is technologically driven, so there's no way to look at where the human universe is going without looking at the effect of emergent technology. There's not really anything else driving change in the world, I believe." Since World War II, he adds, technological change has been growing exponentially, to the current "vertigo," where now we can't have any idea what's going to happen next.

Yup. Things are changing too fast for anyone to foresee. And we can blame it all on the geeks.

Monday, August 06, 2007

The Mad Serb's Dream Comes True at Last

One of the most heartbreaking stories in The Geek Gap is that of Nikola Tesla, a man for whom the term "mad genius" could have been invented.

Tesla, a Serb born in Austria-Hungary, arrived in New York in 1884 with little money and few possessions, beyond a letter of introduction to Thomas Edison, whose new-fangled DC electricity was all the rage at that time. Tesla invented not only alternating current (AC) which we all use today, but also the radio, long before Marconi ever transmitted his first message--"S"--across the Atlantic.

Tesla himself benefited little from his own genius. Promised a huge bonus if he could solve some engineering problems for Edison, he succeeded, only to be told the promise had been a joke. Tesla quit in a huff and went to work repairing sidewalks.

Eventually, he hooked up with Edison rival George Westinghouse, who saw the potential of AC and began promoting it as a preferable alternative to DC. When the ensuing "current wars" between Westinghouse and Edison left Westinghouse financially on the ropes, Tesla magnanimously tore up his contract with Westinghouse, renouncing his right to benefit financially from his own invention. Westinghouse, of course, went on to build an empire that survives today, while Tesla died nearly penniless in 1943 in a cheap hotel room.

As soon as he died, Yugoslavian and U.S. government agents arrived at his hotel room to make off with his papers and notes. They recognized that, nutty though he was, and bad as he was at the business end of inventing, Tesla's was a rare mind that could clearly see what technology could do long before it became reality.

One of the things he saw, but was never able to get enough funding to build, was a tower designed for the transmission of electricity. That's right: in Tesla's vision, charging your battery or powering your electrical device would no longer require wires connected by plugs to electrical outlets, but could simply be done by pulling power waves out of the air, the same way we do with music when we switch on a radio. Tesla always swore it could be done. "It is a simple feat of electrical engineering, only expensive," he wrote.

Cut to summer, 2007. Tesla's name has been popping up again because some researchers at the Massachusetts Institute of Technology (MIT) seem to have built their own smaller-scale version of what Tesla had in mind, transmitting enough electricity from one coil to another across a room (and, harmlessly, through a crowd of researchers sitting between them) to power a 60-watt light bulb. For the moment, the most promising application seems to be getting rid of the tangle of wires and wall warts that plague the area around any computer setup or entertainnent.

But, as the technique is perfected, larger-scale uses are coming, and "now is a good time to start thinking about commercializing it," according to Marin Soljacic, the professor who led the MIT team that made the breaktrhough. (Soljacic is Croatian, which makes you wonder if there's some kind of electrical-engineering-genius-juice mixed in with the water in the former Yugoslavia.)

Even without the MIT breakthrough, transmitted power has been tip-toeing into everyday use, on RFID tags, for instance, which usually don't carry batteries, but instead get all the power they need from the radio waves that come looking for them.

And British startup SplashPower is selling technology that allows users to recharge such items as mobile phones and digital cameras through magnetic induction, simply by setting them on a mat. These devices have to be fitted with a tiny SplashPower receiver to work on the SplashPower mat, so adoption may take some time.

But magnetic induction is already commonly in use in other items including a new kitchen stove that's extremely cool--both metaphorically and literally--the heating surface doesn't get hot! (I want one!) And electric toothbrushes. I remember being quite fascinated with my Oral-B, whose plastic handle sits on plastic charger with no metal contact for the transmission of power, but Bill understood right away how it worked.

Someday soon, we will be free of wires, and there will be room-wide, then later on, city-wide power transmission. Tesla always knew it could be done.

Saturday, August 04, 2007

Too Many Patents?

It's unusual for Apple, Microsoft, IBM and Google to all agree, but in the past few years, they've been unanimous in their opinion of the U.S. patent system: It's broken, they say. We spend too many billions defending nonsensical infringement claims. Time for reform!

Well, reform is coming in the form of the Patent Reform Act of 2007 which would dramatically (or not dramatically enough, depending who you ask) alter the way patents are managed in this country.

The Patent Office, everyone agrees, is completely overloaded. Patent applications take three to five years to even reach the review stage, examiners don't have enough time to check them out thoroughly and too many "junk patents" get through. Large companies argue that inventors pick up as many patents as they possibly can, not so as to use them (most patents are never used!) but in the hopes of winning the patent lottery by suing some larger company when it launches a product with technology or coding similar to one of their patents. Patent attorneys counter that any smart company needs to engage in patent stockpiling, since they're just as vulnerable to being sued as the big guys and "When they come knocking at your door, you want to come back with something other than a check," as Michelle Lee, head of patents at Google put it.

It's hard to figure out who the black hats and white hats are.

The biggest change in the new law would be to grant patents on the basis of first-to-file, as opposed to the current first-to-invent system. Inventors groups' rail against this change, saying it favors large companies with armies of patent lawyers who can get an application in quicker than the little guy can. The first-to-invent principle famously granted some very valuable patents--after a long legal battle--to a hapless graduate student who came up with laser technology but mistakenly thought he'd have to build a working laser before he could file for a patent on it.

On the other hand, the United States is the only nation on earth where first-to-invent predominates; everyone else uses first-to-file and there's no question that the change in law would dramatically simplify the patent process which currently involves a lot of investigation, because it's obviously much harder to determine who first created something than who was the first to get the paperwork through the door.

Meanwhile open source proponents like Eric Raymond think the reform doesn't go far enough to turn back the wave of junk patents that threaten the open-source community.

But of course, that's the philosophy of open-source: no patents, no copyrights, no ownership of anything. And that's the way they see the future.

As someone who makes her living creating copyrighted material, not to mention a class-action plaintiff in a lawsuit by writers against databases that put our material on the Internet, I find the concept of a world without ownership a little scary. But maybe the time has come for that. Maybe we need a whole new model for how we manage the ownership of ideas.

Friday, August 03, 2007

Better than Cloning?

From the You-Can't-Make-This-Stuff-Up files:

In 2004, South Korean Hwang Woo-Suk announced that he was the first in the world to extract live stem cells from a cloned human embryo. Stem cells, of course, are the darlings of medical researchers everywhere, as they can pretty much turn into whatever you want them to. They make the prospect of growing someone a new kidney, or repairing brain cells destroyed by Alzheimer's seem tantalizingly close (which is why the Reagan family has broken ranks with the Republican Party over stem cell research).

Being able to produce the things by cloning opens a wide range of scientific possibilities, and Hwang, who had paid for his schooling by working on a farm and claimed to work seven days a week was ballyhooed. He became a national hero. Stamps were issued with his picture on them.

And then the news broke: Hwang had lied from the beginning. He never cloned those embryos. They were, in fact, donated by female researchers in his lab. Information in his Science article about his cloning achievement turned out to be mostly false.
"I was blinded by work and my drive for achievement," he told an interviewer at the time.

Disgraced, stripped of his many prestigious posts, Hwang was indicted for fraud and embezzlement. The mighty had fallen, and the world had another cautionary tale about an overachiever driven to deception when he couldn't produce the results he aimed for by honest means.

But wait, there's more. Yesterday, researchers from the newly launched journal Cell Stem Cell took a second look at Hwang's cells and announced that they did indeed represent a groundbreaking achievement. It was just a different one from the one Hwang claimed.

Parthenogenesis is scientists' term for the an egg that becomes an embryo on its own without the benefit of sperm. It's also aptly called "virgin birth," and researchers have been trying to do it for years. They've managed to create parthenogenesis in animal eggs, but have never succeeded in human eggs that lived long enough to extract stem cells. But apparently, Hwang did just that, accidentally stimulating the eggs into becoming embryos while he was trying to clone them. Human parthenogenesis raises the tantalizing possibility of creating cells and organs that are an exact genetic match to a female patient, among other things. Some researchers say Hwang's breakthrough was even more useful than cloning stem cells would have been. So he may be a liar and a criminal, but he's made a great contribution to science...by accident.

Sounds kinda like the plot of a TV movie, doesn't it?

Thursday, August 02, 2007

FCC Does the Right Thing...Sorta

Infrastructure.

There's a word most of us suits hate. We hate thinking about the little packets of information flying around that bring us our e-mail, and whether those packets are traveling along cables under the ocean or bouncing off satellites above the atmosphere. We just want to know for sure that the message will get there.

But every now and then we have to at least try and apply our brains to questions of infrastructure, and one of those times was three days ago, when the FCC issued its long-awaited rules for the auction of the C block of the 700 Mhz airwave spectrum (A and B are reserved for local use). The spectrum is being reclaimed by act of Congress from UHF television stations with channels from 52 to 69. Those stations are required to switch to digital by 2009.

It's a big, big deal because it's the biggest block of bandwidth to be sold for the foreseeable future and w0uld allow the purchaser to massively expand its footprint. On top of which, this so-called "beachfront" wavelength is great for going through walls, perfect for urban settings.

The auction itself won't take place till January, and is only open to well-funded bidders: the reserve price is a cool $4.6 billion. But at issue--before companies even decide if they'll bid or not--is what conditions, if any, the winning bidder will have to follow. Major telecom firms like AT&T were lobbying hard for no conditions at all. Google--and consumer groups--were lobbying for a set of four conditions that would have forced the bandwidth to openness, requiring the winner to accept all devices and all applications, forcing the winner to sell the bandwidth wholesale to other services, and requiring its network to accept third-party connections (from ISPs, for example) anywhere on the network.

The four conditions, experts say, would have set the stage for a "third pipe"--large-scale delivery of wireless high-speed Internet to rival the first two "pipes"--cable and DSL. (Note to those of you reading this in a cafe or park over a wireless connection--your Internet signal is almost certainly a wireless rebroadcast of a signal delivered by cable or DSL.)

The FCC didn't go for the third pipe, but it did take up the first two of the suggested requirements: the bandwidth must be open to all devices and all applications. Among other things, this means that in 2009, you should be able to switch your mobile phone company without having to switch to a different phone.

Observers are now speculating as to whether Google will bid on the bandwidth. There seems no logical reason why not--after all if it wins it can choose to implement all four conditions--but perhaps the real goal was to keep the price (relatively) down by making the bandwidth less attractive to the telecoms it would be bidding against.

That remains to be seen. Meantime it's nice to know--even if we're not getting that third pipe--that more open days are coming.