Marcoullier.com

Had breakfast with Joe Greenstein from Flixster and we talked about the future of web destinations. I agree with Fred Wilson– it’s the data, stupid.

That data gets far more interesting when attached to people. Why settle for the results of Yahoo Term Extractor, when we can attach highly structured data from sources like Flixster, iLike and Shelfari?

Once we turn people into highly structured lists of what they produce and consume, the sky’s the limit (privacy issues aside). I imagine that’s where the semantic web guys are headed. Maybe the lifestream folks will get there first.

(Length: 97 words)

Update: Just reread this post and I want to clarify that I am in no way calling Joe an idiot.   Good job, Eric.  Way to make the friends

In June 2007, comScore announced they were tracking widgets. Because only flash units were counted, MyBlogLog (with ~65M unique viewers that month, theoretically 3rd on the list) was excluded.

Today, I was shocked to see the Alexa and Compete graphs for MyBlogLog’s post-acquisition traffic. We were constantly growing during my tenure and Todd laid out some pretty explicit (and impressive) numbers on the official blog recently.

It seems that traditional metrics services still haven’t wrapped their minds around a data-portable world. Information wants to be free, but I wouldn’t mind for it to be audited properly along the way

(Length: 100 words)

Chris Charla turned me onto a cool little service at GDC called Kwiry that helps you remember all the cool little things you come across in your daily life.

1. Learn about something interesting, like a book, band, place, person, whatever
2. Send an SMS to KWIRY with the thing you want to remember
3. Receive a reminder email with a link to a google search for said thing

Dead simple. Dead useful. I’m in love.

(Length: 76 words)

I recently learned that an acquaintance was the victim of a Gmail phishing scam, leading to the deletion of her account and four years of mail. I lost all my 2006 MyBlogLog and Minerva mail, and understand this sense of loss at a visceral level.

Not long after, I read a moving blog post from someone who had their massive Flickr presence deleted under similar circumstances.

Dave Winer has questioned “who will maintain my online presence after I die?” It seems even worse to have to live through it yourself.

(Length: 90 words)

I’m currently engaged in a conversation with Dopplr about why they should offer opt-in public profiles. I’ve suggested several reasons, but really, there’s only one that matters. In the graph below, that massive spike represents the exact moment that LinkedIn a) enabled public profiles and b) stood a chance of becoming a publicly traded company.

(Length: 55 words)

Thingfo, Mike Grishaver’s long-gestating new micro-community service, has finally been released into the wild. Congrats for setting a launch date and then beating it. Check Mashable for the full deets.

(Length: 30 words)

The promise of Dopplr.com is fantastic – effortlessly share your travel schedule with others in order to serendipitously connect with them in person. Unfortunately, the current experience blows. Someone has to search the site for me in order to see my schedule (i.e. the Facebook model).

I would happily promote my Dopplr schedule on my various online profiles (this blog, LinkedIn and Facebook), but there does not seem to be a mechanism for doing so.

As a result, Dopplr seems more like a data harvesting tool than a way for me to inform people of my future whereabouts.

(Length: 98 words)

The news feed is at the core of Facebook app virality. I throw a sheep and all my friends learn that they too can throw sheep. Simple enough. There is an emerging group of services that aggregate one’s online presence into a global news feed – what you do at Flickr, Twitter, Delicious, etc.

Should SocNet-based apps should be included? It would obviously benefit the app providers, but what about the users? FB users have an expectation of privacy, so the answer is likely no. But what about sites like MySpace, with primarily public profiles? Should they be included?

(Length: 99 words)

Former Slide PM Yen Lee just blogged about The Four Viral App Objectives (via Dave McClure). It’s a great metrics-based analysis of the factors that contribute to virality. Whether you’re building FB apps or marketing physical goods, this is worthwhile reading.

“Infection duration” just rubs me the wrong way, though. Yes, it’s critical to computing virality, but it implies a focus on the acquisition of users solely for the purpose of acquisition of users. As long as someone is using your app they are infectious. The focus should be on prolonging engagement, not “infection.” Semantic debate? Probably.

(Length: 94 words)

I’ve been trying to deconstruct how Techmeme works. It appears that Gabe Rivera’s clustering algo basically works in the following manner:

1. Find an interesting story (only on a Techmeme Leaderboard site, I believe)
2. Ping the permalink against Technorati on a regular basis, looking for referrers
3. If the referrer meets a minimum authority rating, add it to the cluster

Given the limited number of sites that meet the authority criteria, it would appear that the leaderboard is highly resistant insular, as is Techmeme in general. No long tail here. Anyone got a better understanding?

(Length: 96 words)

Read more »