September 2009 – Page 2

Month: September 2009 (page 2 of 3)

Netflix prize awarded

Netflix has officially awarded the $1 million prize for improving its ratings engine by 10%. I’ve been fascinated by the prize since it was announced, and I’m thrilled that a team managed to win.

Netflix has also started a new contest. Teams will compete to see who can offer the best recommendations to users who have rated very few (or even zero) titles.

Being a blogger

September 20, 2009 / Rafe / 4 Comments

In a post about how economists often separate their personal beliefs and professional work, Tyler Cowen describes bloggers like this:

In many ways the core of blogging is a willingness to apply what you know to every problem you encounter, and see how good a job you can do of it in a more or less integrated fashion.

That’s pretty much a perfect explanation of what my goal is for this blog. I break down the problems I encounter in terms of the things I already know and believe. I make mistakes and try to learn from them. There are a lot of blogs that don’t fit this model, but it certainly describes many of the blogs I enjoy most.

The literacy revolution

September 20, 2009 / Rafe / 0 Comments

Literacy is going up, not down, people are writing more than they ever have, and it’s all thanks to kids spending more time online.

Why did Mint.com sell to Intuit?

September 19, 2009 / Rafe / 9 Comments

Jason Fried’s rant about Mint.com selling to Intuit has really, really stuck in my craw for some reason. I have no special knowledge of Mint.com’s situation, and from my reading of Fried’s post, I don’t think he does either.

Fried constructs a fantasy world in which the noble founders of Mint didn’t really want $170 million and instead the mean old investors (who enabled Mint to exist in the first place) forced them into selling out and immediately becoming rich.

Anyone who reads Signal vs. Noise knows that despite their rejection of the label “lifestyle business,” 37 Signals is a company that brags a lot about the fact that its employees do not work a lot of hours. And I think that’s great, but something tells me that the 35 employees at Mint.com were not in the same situation.

The strength of Mint was that it was integrated with everything, banks, credit card sites, and so forth. What was obvious to me as a user was that they achieved this integration without the cooperation of those sites. Mint is at its core one of the world’s great screen scraping applications, and if you’ve ever worked on an application that does screen scraping, you know that it’s a ton of work to maintain. When Bank of America or Wells Fargo changed things up and broke Mint’s integration, people were at work until the integration was fixed. And Mint is integrated with hundreds of banks, all of whom can make changes to their Web interface at any time. In fact, I couldn’t use Mint because they never got their system to work with my bank, and yet they still tried. All that effort goes into the hidden plumbing that just makes the site work. Mint also has a shiny Web interface, a nice iPhone application, and the same uptime expectations as people have for their bank.

There were two very strong incentives to go through with the Intuit deal. The first was the opportunity to rake in a ton of cash as a reward for putting in the massive amount of effort that the company must have demanded of them. And the second was to change the relationship between the company and the banks with which it is tightly integrated.

Fried looks at Mint and sees Intuit as their main rival, but Mint faces a much greater threat from banks. Look at their revenue model — they use what they know about your accounts to present you with better offers for similar services. Mint offers checking accounts with less fees, credit cards with lower interest rates, and refi deals on mortgages that will save the customer money. In other words, Mint makes its money by cutting into bank profits. At some point Mint was going to get big enough for banks to go from ignoring them to actively trying to disrupt them, and they needed a strategy to counter that.

Sure it’s gratifying to look at an acquisition and say “greedy investors killed the dream,” but there are plenty of alternate explanations that make more sense and don’t require us to think in childish terms about heroes and villains.

Update: Mint.com uses Yodlee to integrate with banks, so they’re not doing the actual integration themselves. However, it does look like Yodlee uses screen scraping to get data from banks. My main point on the screen scraping was to point out that Mint probably demands its employees work long hours. After writing that, I noticed this on a Mint job listing:

We’re flexible when needs dictate you work from home (although you’ll miss out on free dinner!)

Best career advice you’ll read today

September 19, 2009 / Rafe / 0 Comments

Chris Dixon writes about climbing the wrong hill.

I am a food geek

September 19, 2009 / Rafe / 0 Comments

The New York Times has an interesting blog post questioning whether cropping a photo of Dick Cheney cutting meat distorts the truth. My thought: if Dick Cheney rested the meat longer before slicing it up, it wouldn’t bleed so much.

Grading the Baucus bill

September 16, 2009 / Rafe / 4 Comments

Everybody who’s been paying attention to health care reform was eager to see what sort of monstrosity emerged from the Senate Finance Committee. Today, they presented their bill. The New York Times has provided a forum for health care experts to grade the bill. Everyone seems to hate it.

It seems to me that the fundamental problem with the bill is that the health insurance that it mandates that people purchase (and that employers provide in order to get credit for providing health insurance) is very low quality. I don’t know how cheap it will be, but even if you have it, it will still be very easy to go broke because you get sick. That’s a big problem.

Ezra Klein also notices that the bill punishes employers for hiring poor people. This is what conservatives talk about when they warn of unintended consequences.

Links for September 14

September 15, 2009 / Rafe / 4 Comments

Marginal Revolution: Unacceptable thoughts about Quentin Tarantino. Really interesting (and disturbing) thoughts on Inglorious Basterds.
New York Times: Times Web Ads Show Security Breach. The malware ads were injected by an advertiser who pretended to be Vonage. Corporate identity theft.
Varmint Bites: My Life as a Line Cook. First person account of a (very) skilled home chef who worked in a restaurant kitchen for a night. It gave me a new appreciation of the difficulty of restaurant work.

Learning to love an upwardly mobile coach

September 14, 2009 / Rafe / 3 Comments

So the big news for me this weekend was that the University of Houston football team beat Oklahama State (then the number 5 team in the country) 45-35 on the road. It was their first win on the road against a top 5 team since 1984. For one weekend, my team was the toast of college football and the subject of widespread discussion. It’s not every day that UH leads the highlight real on the college football scoreboard show. Unfortunately, in the wake of the win, some people are saying unhelpful things that need to be addressed.

Houston Chronicle sports columnist Richard Justice wrote a column on Saturday congratulating the team on its big win, and in the process gave the school and its alumni some really, really bad advice. The context here is that UH has a second year coach, Kevin Sumlin, who’s doing a great job. The team went 8-5 last year and won its first bowl game in 27 years, and just won its biggest game in 25 years this weekend. Things are going well. So his advice is to do whatever is necessary to keep Sumlin around:

UH has found one of those special coaches in Kevin Sumlin, and now it’s a matter of holding onto him. UH should be aggressive, not reaction. Sign him. Now. Offer him 10 years or 15 years or whatever he wants. UH president Renu Khator wants a Tier One university. She wants money for research and facilities and all the rest. She knows a winning football team can do wonders for a school in terms of enrollment, donations and exposure. At the moment, Kevin Sumlin is about the best ambassador UH could have.

Let me say that I would love for Sumlin to stay at UH forever. When UH was looking for a coach, I came up with a set of criteria for the kind of coach I’d like to see them hire, and Sumlin fit every one of those criteria. But UH cannot put all of its eggs in the Sumlin basket. I think Sumlin has a real chance to take the UH program as far as it can go, but Sumlin is an ambitious young coach, and I’m sure that once he’s taken UH as far as he can, he’ll want to pursue opportunities to take his own career further.

What I worry about with Justice’s column is that it risks convincing UH fans that things won’t be OK if Sumlin does move on. It’s the UH program, not the Kevin Sumlin Program. It wasn’t the Art Briles program either. Briles, the coach before Sumlin, came to UH, took the team to four bowl games, and was hired by Baylor at twice his old salary. Of course he took the job. And that was OK, Briles did a great job for UH for four years, and put the program in position to hire an even better coach. Thanks, Art! Things are working out well for him and for UH.

For a program like UH, this is an approach that will work. Every program in college football with a coach under the age of 65 is subject to having their coach hired away, either by a bigger, richer program or by the NFL. So counting on building around a particular coach for the next 15 or 20 years is not a realistic strategy. What UH needs to focus on is hiring well and getting the most out of the coaches that they do hire in the time that they have them.

This is the thing that worries me most about UH’s having lost athletic director Dave Maggard earlier in the year. He was great at hiring football coaches. Art Briles was a high school football coach who had spent a couple of years as running back coach at a school that never runs the football. Kevin Sumlin was co-offensive coordinator at Oklahoma, a great program with a great head coach, but he wasn’t at the top of anyone’s list of head coaching prospects. And yet Maggard saw something in both of them, and UH has gone from the dregs of Division I football to being the country’s media darling, at least for a week.

That’s the strategy UH has to pursue. Remaining a place where coaches can make their bones and advance their own careers is the most a school in a non-BCS conference can hope for. At one time, Nick Saban was head coach at Toledo. Rich Rodriguez was head coach at Glenville State. Urban Meyer was head coach at Bowling Green. Those guys are all making many millions of dollars now but they started out somewhere. UH needs to focus on being the very best stepping stone it can be.

And UH supporters have to learn to be the kind of fans who don’t get their feelings hurt when coaches take a better offer. Many UH alumni were so depressed by losing Art Briles that they wanted to hire former coach Jack Pardee, who is 73 years old and has been out of coaching for 13 years, because they knew he’d be loyal. Loyalty is an elusive commodity in college football — Pardee himself left in 1990 after only three years for a job in the NFL.

UH fans, enjoy the team’s success, and worry about the coaching situation if and when Coach Sumlin gets that offer he can’t refuse. And when he’s coaching UCLA in the Rose Bowl or OU in the Fiesta Bowl, you’ll be able to say, “I remember when …”

Links for September 13

September 13, 2009 / Rafe / 0 Comments

Jay Robinson: Some Notes On iTunes LP. Unsurprisingly, the album info is basically a packaged Web site made of HTML, CSS, and media files. I like that Apple is trying a new way to get people to return to the experience of listening to an album.
MySQL Performance Blog: 3 ways MySQL uses indexes. Short and useful.
The American Prospect: The Life and Death of Online Communities. There are plenty of articles on this subject, but they never get old for me. This one’s about GeoCities.
Ask MetaFilter: New York Times malware ads. Looks like the New York Times is running ads that attempt to install malware. I noticed the same thing on Haaretz a couple of weeks ago. Sounds like something’s wrong with the ad brokers. The New York Times owns up here.

rc3.org

Strong opinions, weakly held