Posts from August 2006

YouTube is starting to answer the question of how they intend to monetize the service. We all knew that banner and text ads weren’t the solution for a video service. According to this piece from MediaPost, they’ve settled on "participatory video ads" plus "brand channels".

The participatory video ad is on the upper right in this screen shot and is a video ad for the new Rockstar Games offering Bully (which we’ll probably be seeing a lot of this fall in our house).

When you click on that video, you’ll be taken to a "brand channel", in this case the Rockstar Games channel where you will see more video ads for Rockstar products. Here’s a screenshot of what the Rockstar Games channel looks like in YouTube.

This plan makes sense to me. It’s a "native advertising system" that works well within the social viral community that YouTube has become.

Here’s the video for Bully in case you are curious about the game.

The New York Times had an editorial piece yesterday on the stored search query debate.

To quote from the Times’ opinion piece,

Whatever the government does, the companies themselves should be acting
more responsibly. People who use the Internet have a right to expect
privacy. If companies do not have their users’ affirmative consent to
keep data, they should delete it, and make money from the many other,
very profitable parts of the search engine business.

I agree with that except the part about "affirmative consent".

I think we need to think long and hard before moving to an opt-in approach to solving all Internet privacy issues. Sure, we have the right to privacy on the Internet, but if we force users to opt into stored queries, stored behavior, stored logins, etc, it will make the Internet a lot less useful.

Today Amazon is storing our clickstreams on Amazon and using that to create a custom home page where we get really useful purchase recommendations. If that was an opt-in feature, about a tenth of the people who find it useful would actually be using it.

Today, Google uses the search and click histories of its users to deliver more relevant search results. If users had to opt-in, most people would find Google less useful.

Today, Rhapsody tracks all the music I listen to and make recommendations for new music that I haven’t listened to that I might like. If I had to opt-in to that feature, I’d probably never have used it.

I believe that what’s needed is user friendly opt-out, not opt-in.

By ‘user friendly opt-out” I mean the following:

1 – Boldly and brightly advertise the fact that you are storing user behavior in a way that everyone who uses the service will understand that fact. And clearly explain the user benefits of storing user behavior. This should be done in plain english not legal boilerplate.

2 – Make it simple and easy to opt out and opt back in. The opt-out should be available directly from the main interface of the service (home page, advertisement, login page, whatever)

3 – Allow users to see their stored behavior anytime they want to do that.

4 – Allow users to delete their stored behavior anytime they want to do that.

I believe that “user friendly opt-out” is vastly superior to opt-in and will give the vast majority of the people who use the Internet the comfort to stay opted in and obtain the benefit that all this stored behavior provides.

For more on this topic, I suggest everyone take a look at the principles behind AttentionTrust.

My friend Fred gave me some mixed CDs (old school but cool anyway) and I was listening to them on the train out to Amagansett yesterday afternoon. This version of North Country Girl (from Nashville Skyline?) came on and it just gave me chills. I had to share it with all of you. Something about Johnny Cash’s voice just does that to me.

North Country Girl – Bob Dylan and Johnny Cash

And while I am on the subject of this amazing song, here’s the cover that E did at the Eels Town Hall Show that the Gotham Gal and I went to last summer.

North Country Girl – The Eels Live At Town Hall

And no blog post on this song would be complete without Pete Townshend’s version from the Chinese Eyes record.

North Country Girl – Pete Townshend

I bought Yahoo! stock about a month ago so I’ve been watching Yahoo! closely since then.

This morning Comscore Media Metrix reported in the July qsearch report that Yahoo! has gained search market share for the second month in a row. The gains are not huge but it’s a feat in itself to be gaining share in a market that has been going Google’s way for a long time.

Speaking of Google, they lost one point of search share in July. It would be interesting to know where Google is losing share and where Yahoo! is gaining share (demographic, geolocation, search terms, etc).

There is a great session at the Web 2.0 conference called Launchpad where 13 new web companies will launch their services. I attended last year and it was a very high energy event.

I am a member of the Launchpad Advisory Board which includes some great people on it. We will be starting the process of whittling down the list of over 100 submissions down to the final 13 in the next week.

If you want to be considered for Launchpad in November at the Web 2.0 conference, this is the final week to get in your submission. John Battelle has the details on how to do that on his blog.

One of The Gotham Gal’s favorite records of 2005 was Picaresque by The Decemberists. We must have listened to that record 100 times!

So we were excited to get a pre-release listen to their new record The Crane Wife which will be out on Oct 3rd, just in time for The Gotham Gal’s birthday.

We’ve only listened a couple times, but we’ve already fallen in love with this song, called Sons and Daughters, which is the closing track on the record.

Sons and Daughters – The Decemberists

Tom Evslin makes an interesting observation in his post about The Long Tail.

Tom says:

The power law also explains why it is bad idea to be the twentieth social networking site or VoIP phone company even if #1 was just purchased for a gazillion dollars.  Let’s assume that a gazillion means a $24 million just to make the math easy.  Power law says that #2 is worth  $12,000,000 (still not bad), #3 is worth $8,000,000 (you’d get by even after your VCs’ cut).  But #20 is worth only $1.2 million.  When you check the liquidation preference (see Brad Feld on this) in your financing, you’ll find that you don’t get any of that – it all goes to the investors and they won’t even be happy.

The lengthening tail affects copy-cat entrepreneurs as well as authors.  The ever lower cost of starting and running an Internet business means that #20 will always have to contend with #21 through #50 if it looks like any money is going to made in the category.  Tough to get the investors’ money back.  Actually, since so much of the value is at the head of the curve when talking about network businesses, it is impossible for any but a handful of network business to succeed within a category.

I’ve always felt this way about venture deals, long before Chris Anderson coined the term Long Tail. I dislike copycat deals because for one, they are not likely to produce big gains for anyone, but also because, as Clay Shirkey explained in 2003:

“We also know that as the number of options rise, the curve becomes more extreme. This is a counter-intuitive finding – most of us would expect a rising number of choices to flatten the curve, but in fact, increasing the size of the system increases the gap between the #1 spot and the median spot."

So the more copycat deals there are out there, the harder it is for the second and third players in a market to make it. Competition is good for sure, but too much copycat competition is problematic.

Guitar Hero has taken over the family this vacation. Everyone is in on the action. We need to get a second guitar asap. If anyone has any advice for the absolutely best Guitar Hero controller (there are a bunch of them) we are all ears.

Until we get our second controller, we’ll just have to be envious of these two guys:

Lot’s of discussion in the blog world emanating from Bob Metcalfe’s response to the IEEE Spectrum piece suggesting that Metcalfe’s law overstates the value that an additional node creates on a network.

I think its great that Bob defended the law named after him and went on to discuss its role in social networks. But I was hoping he’d discuss Reed’s Law which states that the value growth of a social network is much greater, instead of n^2 its c2^n. Anything that grows at the power of ^n is a big deal.

To date, the best thinking/writing I’ve seen on this subject was a post Tom Evslin did last fall.

Given the value creation we’ve witnessed in the past year with
juggernauts like MySpace and YouTube, I wonder of Reed may be right. I
am not enough of a math wiz to figure that out. So I’d love to hear
what Bob Metcalfe thinks.