OK Cupid launches Crazy Blind Date

This is crazy... So crazy I can see it actually catching on. I'm quite interested to see how many people start using this.<p>I also like how they say "[each] Crazy Blind Date account is tied to a phone" I'd argue that's a very false sense of security/safety, as it's trivial for anyone to obtain a different (or temporary) phone number. On that point though, I guess you're just as safe with a blind internet date as with one you "electronically know".<p>And heck, let's be honest with ourselves. Maybe it will work better if you, the technically-inclined HN reader, is <i></i>not<i></i> allowed to pick-and-choose who you're going to date next, but your next date is chosen based on your interests and some "magic". After all, isn't that how love is supposed to work?

They would need one HELL of an algorithm to pull this off.<p>I tried OKC. A few good dates, but also several women who were 20-30 pounds heavier than their profile pictures.<p>I've talked to women who use OKC. In the first week, any reasonably attractive woman gets 30-150 message, 95% of them creepy/lame/from old men.<p>You can still find good things on OKC, but you need to learn how to filter. Women set up harsh filters against the flood of spam, and men have to learn i. How to get past those filters ii. How to tell if a woman is really herself<p>Women face more risks than men. Most men aren't crazy. But all it takes is one crazy man to really mess up your life. And crazy men are more likely to use something that lets them bypass traditional filters.<p>Meanwhile, this lets women that won't get hit on in real life bypass the filter of physical appearance.<p>If this could be made to work, it would be incredibly popular. But it has a ton of hurdles. You'd have to evaluate safety and attractiveness algorithmically, or you'd end up with a ton of lame/scary dates.

From this Forbes piece: <a href="http://www.forbes.com/sites/jeffbercovici/2013/01/15/would-you-go-on-a-crazy-blind-date-ok-cupid-thinks-so/" rel="nofollow">http://www.forbes.com/sites/jeffbercovici/2013/01/15/would-y...</a><p>""" In truth, the most radical part of Crazy Blind Date (whose name and basic concept OK Cupid toyed with in an earlier form more than five years ago) may be the way users are asked to express their satisfaction or lack thereof. After a date’s conclusion, the participants are invited to rate each other by purchasing “kudos.” Spending a few dollars implies that a date was enjoyable...<p>Why pay good money to improve the dating prospects of someone you might never see again? Because there’s more than generosity at play here. The user who never buys kudos for his or her dates will find himself increasingly shut out by the algorithm in the same way that your own friends would eventually stop setting you up if you rejected enough of their suggestions. “It’s a soft payment for access,” says Yagan. “How much you donate determines your product experience along the way.”<p>""" Who will become successful at getting dates then? Those with the most kudos purchased? And, what does that do to the quality of dates one might expect to have?

I was really sad when they killed the original web-based Crazy Blind Date a few years ago -- I was single and in the Bay Area at the time, and it seemed just random enough to be awesome.<p>Would strongly recommend this to anyone in the market. Although hopefully it gets promoted in places other than just hacker news, or there will be obvious issues.<p>I know a surprisingly large number of tech people who have gotten long-term girlfriends or married via OkCupid. I only ever met actually crazy people (although interesting and fun for a few months), but somehow it seems to work for other people. I'm not sure how well it works outside the Bay Area, Seattle, Boston, and NYC, but in those markets, it's pretty good.

I had the exact same idea, right down to using ratings to improve the random match, so this is pretty awesome in that I get to see how this will play out for free.<p>The main idea here is that it's a behaviorial approach to optimizing mate-matching. The way it works now (e.g., on okcupid) is that you fill out a profile and answer some questions. You expose some signals in your profile which you think are correlated with what makes you a good mate (interests, income, etc.), and go about finding such signals in other people's profiles. The match % is also such a signal. However,<p>1. People lie, or at least exaggerate the truth/selectively filter<p>2. They might not lie but simply be wrong: they think they want x but actually want y<p>3. The match % is a completely made up algorithm. I mean, it kind of makes sense that people who agree on things they care the most about would make good couples, but unless you measure this in a scientific way (i.e., controlled experiments), you have no idea.<p>4. Even assuming that you can actually get a good signal from the profile, it is drowned out by the noise. Suppose that some girl and dude are perfect for each other. What are the chances that he will find her profile and that she will read his message?<p>The solution to 1-3 is to measure the thing you want to optimize, instead of optimizing based on intuition. The solution to 4 is to set people using the thing that you measured. So, after each blind date, you ask the participants, "would you go on a date with this person again?" Imagine being okcupid and being able to say, here is a person I think you'd be interested in dating, and I can say with 95% confidence that you will have a good time. If they can get to that stage, they'd be able to make good money.<p>There are some issues with this, of course. You could argue that the thing they should measure is whether the date leads to a happy long-term relationship, but that's pretty much infeasible for various reasons. Also, as a dating website, they'd be perversely incentivized to keep you single forever. Anyway, I am super interested to see how this does.

Personally, I think Grouper's model is more likely to be successful.<p>They take a group of three friends and pair them up on a blind group date with another group of three friends. The safety in numbers aspect is, I think, a smart addition.

I met my wife via a blind date set up by a friend. No dating algorithm would have ever put us together. However, I think people do get too wrapped up in superficial likes / dislikes as a filtering mechanism, when it's been my experience that those on the surface things really have little correlation with you who might end up with long term. So "forcing" people to get to know each other by actually meeting might have some value here.

Interesting, because in some ways it's the opposite of OK Cupid. As multiple (brilliant: <a href="http://blog.okcupid.com/" rel="nofollow">http://blog.okcupid.com/</a>) blog posts attest, people do judge each other heavily on looks and their stated interests.<p>I assume the app is still using OKC data, and just hiding it. I'm not sure that I would trust the OKC algorithm enough to rely on it, and the question is ever-present in my mind- if you're already on OK Cupid, why use Blind Date? Either you think it'll be a laugh (the best option, probably) or you haven't been at all successful at getting a date on the main site. Perhaps there is a reason for that.

A word of caution about OK cupid. The emails they send out for your QUIVER MATCHES actually have your login credentials embedded in the URL. Meaning, if you right click, copy link address, and send the url to your friend, THEY ARE NOW LOGGED INTO YOUR ACCOUNT.<p>I learned the hardway, I pasted one of these links into an 80+ chat room full of 4chan rejects on IRC.

I predict - men will think it a good idea, women won't - idea will tank.

After reading some of the comments. What do you think about this approach?<p>1) Woman sets up the date. Place and time.<p>2) The system sends an invitation to several men until it gets 3-5 positive responses.<p>3) The woman chooses one of the candidates.<p>The advantages for the woman is that she controls the terms of the date. She doesn't have to sort thru tons of messages.<p>The advantages for the man, is that you don't have to spend a lot of time sending invitations to many women. Invitations that are basically ignored

Looks like it's not quite live. Nothing happens upon clicking "Get Crazy".

Hmm, not too hopeful about this.<p>The idea of a "blind date" is scary enough when it's been set up by someone you trust.<p>I could certainly see creepy or slightly unhinged types making use of this and it getting a lot of bad press very fast amongst women.<p>I can also see it being popular amongst certain classes of gay and bisexual men who will try and use it to hook up with straight guys.

And those who use this will no doubt be the ones who can't get a date on the regular site (like, hm, me).<p>Which means your potential pool is limited to those who can't get a date in real life (which may be because they have just moved, are atheists in South Texas, etc) and who can't get a date on the site.<p>So I doubt this will be effective, but hey it is a nice try.

Probably takes only a couple of hours until someone publishes an app on Github to take a screenshot and descramble the user pictures.

The concept is interesting, although OkCupid actually launched this service in 2007 and it never really caught on. I'm guessing that they're betting mobile will change things significantly.<p>What I really don't understand is why the design is so bad. OkCupid has always looked reasonably impressive (at least in my opinion), and this new site feels flat and amateurish. The mobile apps don't look good either.<p>It almost feels like they've fallen for the new "me-too" trend of exceptionally flat design, inspired by Metro, but the execution here is just not good. The typography is particularly bad. Maybe it's just me, but I think it could really use some additional polish.

I actually used CBD in it's first iteration briefly several years ago. I only went on one date but it actually ended up being one of the best internet dating experiences I've ever had. Didn't work out in the long run with the girl, but we did date for a little while. The idea of actually creating a "profile" for dating I think ends up resulting in a good amount of disparity between the curated persona you offer to the digital world and who you really are, no matter what you do.<p>Wish these guys the best of luck with it. And if you are curious about the service, recommend giving it a shot!

As someone who is extremely indecisive and laid back on little decisions like "where do you want to eat" or "what do you want to do today", I love this idea.<p>Just pick for me, and I'll go do it. It's fun!

Based on my experience with OKC I am doubtful this will be successful. I went on a lot of first dates w/girls from OKC and most of the time ones that were supposedly 90% matches were the worst dates.<p>I met my girlfriend on there too, and we were only supposedly 70% matches according to OKC but she's my perfect partner.<p>In other words, the algorithm is kinda off. At least in my case.

Or an alternative could be a blind date game show if an in person meeting with a stranger is a bit much. We just launched a dating game show that solves many of the shortfalls of online dating. HN thread <a href="http://news.ycombinator.com/item?id=5060477" rel="nofollow">http://news.ycombinator.com/item?id=5060477</a> (disclaimer, co-founder)

There is a value in social services "smoothing out the details" for a user. Naturally, this step requires the user to feel comfortable about entrusting those decisions to an app and it remains to be seen if people want to do this with dates.<p>The horror stories (which are always smirkingly relayed - nothing tragic) I hear from girls about OKC messages don't exactly give me confidence in this part of the company's hypothesis. I would be happy to be proven wrong, however. In the end, even ChatRouelette attracted a wide variety of people (for a very short time). Maybe there is an opportunity to use a Chat Roulette/Airtime interaction as the precursor to a (not-so-blind) date.

Prediction:<p>Good evening. In tragic news today, a woman was killed by her "Crazy Blind Date."

Tested CrazyBlindDate last night. Recognized a friend from okcupid labs (despite the surrealist facial shuffling) and clicked to set up a date. It gave me two options to meet him at Philz in SF - 3:14am or 3:44am. So clearly there are still some bugs to work out. That said, the mobile app for CBD feels much more modern than anything I've seen okc release recently.

My prediction: burner phones or Google Voice leads to some very bad scenarios. Stick with blind dates from friends.

I always enjoy the blog posts they do analyzing big chunks of data. Hopefully they will do one regarding this.

There must be something off about the Android version. I've never seen that high a proportion of 1-star ratings.<p><a href="https://play.google.com/store/apps/details?id=com.okcupid.cbd" rel="nofollow">https://play.google.com/store/apps/details?id=com.okcupid.cb...</a>

This is great if you have a smartphone and only want to date people that have smartphones.

I'd actually trust the OK Cupid guys to do an interesting take on this.<p>Their blog has been a fantastic source of interesting articles of the things they have discovered with their unique dataset.<p><a href="http://blog.okcupid.com/" rel="nofollow">http://blog.okcupid.com/</a>

Technically, this is a relaunch: it used to exist a few years ago. I'd assume the relaunch is an answer to Grouper?

And...it doesn't accept my Australian mobile phone number, nor does it accept my Spanish mobile phone number. :(

I think they mean re-launch, because this is something they had done a long time ago. With the same name even.

I threw a wish in the well, Don't ask me, I'll never tell I looked to you as it fell, And now you're in my way<p>I'd trade my soul for a wish, Pennies and dimes for a kiss I wasn't looking for this, But now you're in my way<p>Your stare was holdin', Ripped jeans, skin was showin' Hot night, wind was blowin' Where you think you're going, baby?<p>Hey, I just met you, And this is crazy, But here's my number, So call me, maybe?

A friend of mine actually started something like this, it's called meetcute.org

I'm surprised no one mentioned coffeemeetsbagel.com yet.

Launches? I used it in Boston ~4-5 years ago.