Category: Uncategorized

On behalf of the who Global Nerdy team (that's me and George), I'd like to wish our American readers a safe and happy Thanksgiving long weekend! Being based in New York, I;m certain George is doing up the holiday in style. I'm not missing out on it either: I'm celebrating with my in-laws in Boston.

As you may have guessed, posting over the next couple of days is going to be a bit spotty, but come next week, you'll find an invigorated (if slightly heavier) Global Nerdy team giving you more tech news mashed up with our razor-keen insight. Have a good long weekend!

Techdirt took the post title I wanted to use ("If You Liked This Post, Perhaps You'd Like To Look At The History Of Failed Recommendation Systems," clever bastards), so would Slide's algorithm try to connect me to Mike Masnick?

Fortune takes a look at some of the newer businesses out there building themselves around recommendation engines (such as Pandora, the personalized internet radio station). The company with the broadest ambitions in this space is Slide (they've recently pulled down some major financing), founded by ex-PayPal bigshot Max Levchin.

Suppose, for example, there's a user named YankeeDave who sees a Treo 750 scroll by in his Slide Show. He gives it a thumbs-up and forwards it to his buddy" we'll call him Smooth-P. Slide learns from this that both YankeeDave and Smooth-P have an interest in a smartphone and begins delivering competing prices. If YankeeDave buys the item, Slide displays headlines on Treo tips or photos of a leather case. If Smooth-P gives a thumbs-down, Slide gains another valuable piece of data. (Maybe Smooth-P is a BlackBerry guy.) Slide has also established a relationship between YankeeDave and Smooth-P and can begin comparing their ratings, traffic patterns, clicks and networks.

Based on all that information, Slide gains an understanding of people who share a taste for Treos, TAG Heuer watches and BMWs. Next, those users might see a Dyson vacuum, a pair of Forzieri wingtips or a single woman with a six-figure income living within a ten-mile radius. In fact, that's where Levchin thinks the first real opportunity lies – hooking up users with like-minded people.

Fascinating how we're constantly repeating ourselves when it comes to this specific field. Let me tell you quickly about Opencola.

Opencola was a startup Joey and I worked at (actually, it was thanks to Joey I got the job with OC) built entirely around the idea that collaborative filtering could help you sort through all the crap on the internet. Basically, after watching users and figuring out what they did (and didn't) like of the content they saw on the internet, Opencola's software would find other like-minded users and start filtering the internet for you based on their preferences. Our pithiest pitch for our vision was that we wanted to "relevance-switch" the internet (I credit our co-founder Cory Doctorow for that little bit of genius wordsmithing). Just to add to the excitement, it was open source and built on a peer-to-peer networking architecture. You could imagine what it must have been like to take that pitch on the road during 2000 (and pre-crash 2001).

Step 1: Vision.

Step 2: PowerPoint

Step 3: Sit back and open up the checks.

So, Joey and I (and every other member of the Opencola diaspora) know recommendations. Here are a few things (I think) I learned:

1. The broader your topic space, the more difficult it is to determine relevance. It's a hard bias to shake, but we all instinctively believe that our similarity with someone in one domain indicates similarity in another. For example, a shared love of experimental jazz doesn't mean that someone else with an extensive Ornette Coleman collection shares your political views. Or your taste in films. Or your style in clothing. It's the basic problem of trying to build a recommendation engine for something as broad as the internet, or even something like eBay or Amazon.com. The only way around it is to have multiple, segmented, and deep individual profiles filled with preferences.
2. Deep individual profiles are computationally difficult to compare. Assuming you can narrow down the topic space to something where similarity does correlate to relevance, you still have to handle the problem of comparing very deep profiles. One of the reasons the Amazon.com recommendations seem so basic is that they go for breadth, rather than depth. In other words, your purchasing history isn't carefully being compared with the purchasing history of other individuals as much as Amazon.com's simply seeing what most other people bought when they also bought the item you're currently looking at. The profile is one data point deep, but hopes the wisdom of the crowds steers customers in the right direction.
3. Relevance and preference signals are difficult to collect. Ultimately the whole recommendation enterprise rests on the collection of signals from users. They have to indicate what they like, what they buy, etc. In some cases, the signals are explicit and obvious—if I buy something (from a stock to a CD) it's a pretty good bet that I'm sending a positive signal—but in most cases, there aren't many explicit signals to go on. So what can recommendation engines do to overcome this? Making users signal their preferences explicitly is one way to go: rating their music from zero to five stars, digging news stories, etc, are all examples of how some systems cope with the issue. Unfortunately, making people do stuff  they probably wouldn't otherwise do is often a dead end. It's effort, and people hate effort. That Pandora works for their current base of users probably has more to do with the fact that it attracts the kind of user who's predisposed to contribute to the overall effort by rating the songs they hear. To get beyond the most basic kind of recommendation algorithms, you need to pick up the subtle judgments of the individuals in the system, and that's hard to do without either being invasive (to the point of violating privacy) or forcing people to make their implicit judgments artificially explicit.

Before Opencola went through the first of many VC–imposed mini-implosions (resulting in the loss of our founders and many of the original mad scientist staff, and ultimately in the ignominious garage sale of the resulting technology to Open Text), we thought we had figured out one way to make our vision real. First, we reduced the topic space: rather than relevance switch the whole damn internet, why not simply help gamers find new games that would appeal to them? That way the topic space is bounded to games and what gamers think about them, and we weren't going to try to tell you what toothpaste to buy based on your gaming preferences.

By building on a peer-to-peer network architecture, we also went some way towards solving the problem of comparing deep profiles. We elected to distribute the work to the peers themselves by having each peer constantly evaluate their similarity (based on that deep individual profile) to others, rather than trying to perform all this magic in some ginormous data center.

Finally, game software is one of those rare things where you can actually collect a lot of subtle judgments through explicit signals. Names get searched on. Demos get downloaded and installed. Demo software gets played. Perhaps the user plays it once, but maybe they play the demo several times. They trash it, or they buy the full version. Do they play it longer than the average player for that game? Do they download mods? Make their own mods? In other words, you can collect tons of meaningful data without asking people to step out of their gaming routine.

It was a nifty idea and to this day, I wonder why nobody has put all of these pieces together yet (but I assume it'll happen someday).

The hoops Opencola had to jump through all point to the difficulties of building a recommendation-based business, and automating the subtle process of tastemaking. The ultimate prize of really relevant suggestions, however, means we'll see many more millions thrown at this problem for years to come.

Link

Tags: recommendations, collaborative filtering, social search, Opencola, Slide, Max Levchin, Pandora, Amazon.com, peer to peer

This story starts out with video of a tragically bad corporate reinterpretation of U2's "One" (commissioned to commemorate the union of Bank of America and MBNA's credit card businesses). It was so painful, it became something of an internet comedy senation. The New York Times, however, notes that the tale has turned tragically stupid:

A video of two Bank of America employees singing a version of U2’s “One” to commemorate their company’s acquisition of MBNA recently made the rounds of the blogs, prompting amusement and some ridicule from online viewers.

But the intended comic effect of their performance and the retooled lyrics (“One spirit, we get to share it/Leading us all to higher standards”) seemed lost on lawyers on the lookout for copyright violations.

On Tuesday, a lawyer for the Universal Music Publishing Group, a catalog owner and administrator, posted the text of a cease-and-desist letter in the comments section of Stereogum.com, a Web site carrying the video. It contended that Bank of America had violated Universal’s copyright of the U2 song.

All this really shows is the indiscriminate nature with which Big Content uses the legal weapons at their disposal, without a second's thought or reflection. I won't bother with the neo-hippie "it used to be about the music, man" lament that the music business is now dominated by corporate management, but doesn't this kind of nonsense make you wonder if the legal staffs at the major record labels have too much power?

The joke's on UMG, of course. The sheer awfulness of the song has spawned a cover featuring Johnny Marr (ex of the Smiths) and David Cross (ex of Mr Show and Arrested Development). Enjoy.

Note to UMG lawyers: please click on "Leave a Comment" to post your cease and desist notices.

Tags: U2, UMG, Bank of America, MBNA, One, Big Content, Johnny Marr, David Cross

This must have been an interesting weekend for Yahoo! employees. First, you get the direct fallout from the Wall Street Journal's story covering "the Peanut Butter Manifesto." Then you get the secondary analysis of just what ailes the world's most popular network of web properties. Finally, Yahoo! announces a deal with a clutch of newspaper chains. What's it all mean, and what's the relationship between these stories?

Let's try to connect some dots.

This year, Yahoo has suffered from slumping shares, slowing revenue growth, staff defections and a delay in a crucial project aimed at boosting online ad sales. As the memo shows, even some current executives have been fretting that the Internet company's top management isn't prepared to take the strong medicine they feel is needed to right the ship.

Some worry that Yahoo — whose activities range from online dating to fantasy sports — has stretched itself thin and lost track of priorities. Recently, the company has been outmatched in key areas such as search advertising and social networking. (See related article.)

Last month Brad Garlinghouse, a Yahoo senior vice president, wrote the memo, titled "The Peanut Butter Manifesto," for top executives. His contention: "Change is needed and it is needed soon."

Mr. Garlinghouse, who once shaved a "Y" in the back of his head, argued in his manifesto that Yahoo is spreading its resources like peanut butter on bread, thinly and evenly across all its activities. "Thus we focus on nothing in particular," he wrote, saying the Sunnyvale, Calif., company needs to pick specific areas to focus on and make bigger bets on them while dropping nonessential activities.

So, is Yahoo! broken? So heavily matrixed, sclerotic, and unfocused that it needs radical surgery? Or is Yahoo! simply a successful upgrade or a deal away from returning to their former glory atop the internet advertising pile, thanks to their industry-leading reach? It would seem that Wall Street can't figure it out either. Accoding to the New York Times this morning:

So is Yahoo stock, which now trades at $26.91, down 31.3 percent this year, a bargain suitable for value investors? Or is the once highflying company — which in its heyday was regarded not unlike today’s Google — destined to bring further disappointment?

Yahoo management, led by its chief executive, Terry S. Semel, is counting heavily on a technological upgrade of its search engine to enhance profits. The delayed upgrade, called Project Panama, is now scheduled for introduction in 2007, and analysts expect it to narrow Yahoo’s search gap with Google. Merrill Lynch figures that the project will raise revenue per search to around 5 cents in 2008, or as much as 7 cents if Panama proves a rousing success.

“Monetization is a hard thing; not too many people do it very well,” said Mr. Befumo at Legg Mason. But, he contended, Yahoo has some appealing alternatives.

“If you have a traffic problem, then you have a fundamental business problem because you have nothing to convert into revenue dollars,” he said. “But if you have a monetization problem, which is what Yahoo effectively has, you always have options.”

So the (investment) world is splitting into two perspectives on Yahoo!: those who believe the company is caught between aggressive competition from Microsoft and AOL as well as niche players like MySpace and YouTube, and those who see strength in their massive reach—a huge audience simply awaiting the right tools to properly address their needs.

What's the state of Yahoo!'s traffic? Fred "A VC" Wilson has a very useful post on the matter. It's clear from his numbers (which come from comScore) that Yahoo!'s still the biggest property manager on the 'net. Unfortunately, Yahoo! isn't growing their audience at the average rate of the internet as a whole. Google itself is barely keeping pace (if you don't factor in YouTube's audience growth). Here, I drew a quick graph (using Fred's comScore numbers) to show you what I'm talking about:

As Fred points out, the internet may not be mature (there's still triple-digit growth available for new services), but the warhorses of internet eyeballs—the portals—are looking pretty mature themselves, relatively speaking. Yahoo! has to keep pace with the net as a whole, otherwise every point of growth in the size of the internet audience comes at its expense. The chart above, however, makes me think that a portal-based strategy, essentially creating an all-encompassing destination for an audience, isn't where the internet's growth is coming from. Audiences are being attracted to sites that showcase within some kind of niche, either by virtue of the medium (YouTube for video), or the subject (Facebook for college kids), or the purpose (Wikipedia as a reference source). The portal's dilemma is that it's simply untenable to be as good as a specialist might be when that specialist is simply a click away.

So, should Yahoo! acquire audience growth? One of the things Yahoo! has taken some heat for has been their seeming inability to close deals for Web 2.0 properties like MySpace and YouTube. When you see the audience growth rates for Fox Interactive (ie, MySpace) and YouTube, you can see why: if Yahoo!'s core portal offerings aren't even keeping pace with the growth of the medium, then they can only increase their share by quickly incorporating the up and coming destinations.

On the other hand, if you're a Peanut Butterite, you'd have to believe that Yahoo! couldn't have digested those properties, leaving them as-is alongside existing (and competing) internally-developed services like Yahoo! 360 or Yahoo! Video. Even if you think Yahoo!'s traffic is fine but the monetization needs a tweak, then you'd have to imagine that the incremental traffic delivered by either acquisition would have come in handy later rather than sooner. In either case, it would have been a short-term waste of shareholders' money.

Back to Garlinghouse's assertion, then, that Yahoo! needs to focus on high-impact businesses where it can dominate. That sounds partly right, but it's a half-measure unless Yahoo! also addresses how they earn money with those visitors. Yahoo! has to get their ad network (search, contextual, rich media, and otherwise) right.

Google, of course, has neatly sidestepped these issues by not being a destination and concentrating on brokering individual audence members' needs with the information in their index. By focusing on building an ad network rather than a content destination, they allow the specialists to chip away at Yahoo!'s share of the audience (in fact, they finance the specialists through ad revenue).

And while it's critical for Yahoo to bring their ad network up to parity with Google's, they'll have to cut deals and spend money to ensure Google doesn't erode Yahoo!'s third party reach any further. That's the motivation behind today's announcement that Yahoo! will work with 176 local papers from seven different newspaper chains on a long-term project to share ad revenue and local content. Whatever the details of the deal, the main point is that it takes some powerful local ad distribution out of Google's reach. Yahoo! will be able to offer advertisers regionally targeted ads distributed by well-known local brands (such as the Atlanta Journal-Constitution).

Yahoo! has grown up to be the biggest media property on the web but their slice of the web's audience is declining all the time. Google, with their focus on search and advertising technology, has positioned itself to not only put ads in front of more people than Yahoo! can, but to earn more money when they do it. Yahoo! has rightly started to take a hard look at itself: what good is all the traffic it generates if it can't convert it into profits the way Google can? With competition on all sides, should the company consider reducing the scope of its services to focus on ways it can compete with Google, or is its breadth of offerings a strength, hidden behind second-rate ad technology?

Yahoo! CEO Terry Semel will come under inreasing pressure to find decisive answers to these questions, and soon.

Tags: Yahoo!, Google, internet advertising, content, Peanut Butter Manifesto

Happy 21st birthday, Microsoft Windows!

Although you've been of legal drinking age in France and Italy for five years, the United Kingdom for three and most of Canada for two, you're finally old enough to procure Bud Light in your home country!

In honor of this milestone, the site Connected Internet has published a list of 21 things you never knew about Microsoft, Windows and Bill Gates. Among the items in this list:

• The price: Microsoft Windows 1.0's retail price was US$100. The Connected Internet article states that 100 1985 dollars is worth about $177 today, which is about a dozen dollars short of what it costs to buy the full version Windows XP Home edition with Service Pack 2 at Amazon.com.
• Requirements to run Windows 1.0: 256KB of RAM (remember, that's kilobytes, not megs!), DOS 2.0 and two floppy drives.
• Plus ca change, plus c'est la meme chose: Windows 1.0 was out a full 2 weeks before it had to be patched to fix some bugs.
• Why Bill doesn't work in branding: Bill didn't want to call it “Windows”; his preferred name was “Interface Manager”. You've got to remember that this was the early days of personal computing. You have to remember that only four years prior to Windows 1.0's release, IBM's PCs were produced by an independent business unit whose name was “Entry-Level Systems”, which implied that PCs were something you bought as a gateway to buying a real computer.

If you're curious as to what installing Windows 1.0 was like, DigiBarn has a photo essay that goes through the whole procedure.

We'll close by leaving you with this cult favorite: a video of Steve Ballmer doing what I presume is an internal promotional video for Windows 1.0:

If you're wondering what Nintendo's up to amidst the noise surrounding the PlayStation 3 release, here it is:

Everything I hate about SOAP — the so-called “Simple Object Access Protocol” — has been summarized quite nicely in the dialogue The S Stands for Simple in Pete Lacey's Weblog.

Link