Greenspun is assuming that a "perfect" dating solution would result in zero people who want to be married who are not married.

This is a pretty odd assumption. Let's imagine that my mythical friend Tom desperately wants to marry, but is kind of picky: he wants a supermodel who is also an expert Python coder and who can beat him at Tekken 2. Since Tom is poor, has rather poor personal hygiene, and weighs about 200kg, we can safely say that he will be involuntarily single his entire life. Does his existence mean that all dating methods and tools are doomed to failure? I'd say no. Much like there will always be some structural unemployment, there will always be some "structural" single people. At best we can just reduce the number slightly.

And this isn't the only flaw in Greenspun's argument. As others have pointed out, he is assuming that the only purpose of online dating sites is to get married - and yet a cursory examination of online dating sites and the people on them would indicate that this is not the case.

Meanwhile, Greenspun's reference to marriage statistics is deeply problematic:

1) First, and most obviously, correlation is not causation.

2) He phrases the argument in the context of "low marriage rates prove online dating is a failure", but he then turns to statistics that show falling marriage rates. He seems not to realize it, but what he's actually arguing is not that online dating is a "failure" that has failed to improve marriage rates, but that it is actually lowering marriage rates. This probably needs an explanation. How exactly is eHarmony destroying the institution of marriage?

3) When pressed in the comments, he talks about "involuntarily single people", but again, the statistics he references are relating to marriage rates. It's a bit like trying to show that fewer people are starving to death by pointing to rising obesity levels. An obese person is almost certainly not starving, but not all non-obese people are starving. And a married person is probably not involuntarily single, but not all unmarried people are involuntarily single.

4) Don't falling marriage rates actually show that what people really want is, incrasingly, not marriage? Revealed preferences, and all that? But if people don't want marriage, then we can hardly use low marriage rates to show that online dating is a failure. His statistics would seem to not just be irrelevant, but to actually undermine his own argument.

(Yes, I know, I'm overthinking this.)

There's a big assumption in this logic: that most people who date online are looking to get married.

A lot of people use online dating simply to hook up -- the virtual equivalent of approaching someone at the local bar, only with a lot of the potential awkwardness pre-vetted out. Though many of the biggest dating sites advertise themselves as places to get married, the use cases often tell a different story. That's not the fault of the sites, so much as the predominance of "off-label" use by the customers. (And the fairly recent plethora of hookup-specific online dating services are catering to an existing trend, not trying to create a new one.)

Frankly, none of this should surprise us. Many young people are primarily interested in casual sex. Still others are pursuing a try-before-you-buy strategy, and even among those people, the intent to "buy" is nebulous and flexible.

I have no doubt that a large contingent of online daters are genuinely using these services to find The One, and are earnestly focused on dating. But I'd be very curious to see what percentage this userbase represents. My gut tells me it may not even be a majority.

The parenthetical (from) tag on this post should be "Phil Greenspun" rather than "harvard.edu".

Phil G is a talented software guy, and has put the time/money in to becoming a very accomplished pilot. He writes very well on those subjects.

...

I got married early compared to my peers. I've been with my wife for about 12 years and married for 5. Over time, I've definitely noticed that in my group of friends that aren't in long term relationships are using online dating.

I feel like 12 years ago there was a negative stigma attached to it, similar to the chat lines. It was the sort of thing you used once you 'gave up'. Now, several couples I know are proud that they met online.

I wonder if I'm only noticing more acceptance of online dating because my peers are reaching the 'breed now or never' age?

Perhaps culture is changing, ancient religious customs are failing, and those of us who are inclined towards monogamy are eschewing marriage while still committing to our life partners? Also, perhaps homosexuality has hit the point culturally where those who otherwise would have lived an unhappy false life of heterosexuality are now free enough to be gay, and gay marriage is mostly illegal in this country.

I'm not even sure where to start. Failure to whom? The people making money by providing a service? The users of the service? The marriage industry?

One reason may be that the vast majority of young (very fertile) pretty women are not looking for dates online; simply because they have plenty of choices around them and therefore don't need to look in the virtual world. Males follow women in that category, wherever they are.

Is this really an argument? It's such a terrible strawman, I have a hard time believing this isn't logical satire.

If someone gives me a hammer, and I build a crappy house, it's not the hammer's fault.