Ask HN: Will there ever be an eHarmony for finding friends? | Hacker News Reader

Ask HN: Will there ever be an eHarmony for finding friends?

51 pointssendos96 comments

From what I know, eHarmony asks you a long list of questions, and then tries to match you up with people they think you would be a good romantic match with.

Not sure how well the above works, but I always thought a similar thing that applied to finding people whom you would click as friends would find many applications:

Moving to a new city, moving into a new apartment complex, starting a new job, starting college, going on a cruise with a ship full of people you don't know, etc.

I guess there are websites that help you expand your network of friends, but as far as I know, no website runs a test that is analogous to the one on eHarmony.

My questions are:

1) Do you think that a questionnaire + algorithm could ever be developed to find people that you would click with as friends, or are human nature and human interactions too complex and unpredictable?

2) Assuming such an algorithm is possible to develop, would it get traction, would people use it, would people pay for it?

View on HN

1) The former.

2) Not easily. The biggest problem is that the people who most need help finding friends are the ones you least want to find on a friend-finding site. Every dating site seems to be drowning in this problem, which means they have to resort to scam techniques and exploiting new, naive users as much as possible. OkCupid is the only exception that I've seen. Hard to say whether you can replicate what they're doing (that is, not sucking) in the Platonic realm.

OkCupid gives 3 ratings: a Match percentage, a Friends percentage, and an Enemy percentage. Isn't the op's idea already rolled into this?

It'd just mean using the matcher on people you wouldn't romantically pursue. I think there's too much of a stigma of okc being a dating site, so that people think using it as a friend finder would be weird.

People use okc to make friends all the time. This problem is pretty much solved for the 18-30 set. Now, how do you make friends in your 40's 50's or 60's is another problem.

They should re-brand it or do a spin-off if that's the case. OK Cupid, just sounds like a dating website, and I'd assume someone with a profile on that site is merely looking for a partner.

Why rebrand? Guys and gals you don't have sex with are friend zoned.

OkCupid sets all of those up with match questions: you answer a given question, and mark what acceptable answers your "ideal match" would choose. The match percentage is calculated by comparing your answers to the other person's "ideal match" acceptable answers, and vice versa. The friend percentage is only calculated by comparing your answers to the other person's answers. But the set of questions themselves is the same, which means your "friend percentage" with someone can go up if you both like to be tied down during sex, or want to have babies, when these issues are clearly more relevant for matches than for friends.

I disagree. There's a big difference between finding a date and finding a friend (in the sense that the OP uses, i.e. not in the friend with benefit sense). As you point out (unfortunately, but it's a fact of life) objectively beautiful/attractive people are rare in the population and they are even more rare on dating sites. However, social interests, may be rare locally but on a national scale can be quite easy to match up. For example I like to play Go but it's rather hard to find players in my immediate friend circle and I cannot go around asking people.

Book clubs, chess clubs, etc. have operated in this manner for a long time, the only difference is that they substitute a physical space for a web site.

The real problem here is that it would be hard to create a comprehensive questionnaire that covers, say, 90% of interests and it would be even harder to find users patient enough to fill such a monstrosity. The only way to go is for them to fill out a small questionnaire and try to rapidly learn their preferences by asking them questions, 20 questions style, I think.

I'm not sure focusing solely on interests is the way to go because, if I think of the people I consider my best friends, what amazes me is that we don't really share that many common interests or hobbies.

What bonds us seems to be that when we talk, we feel free to discuss and explore any topic, we have a curiosity about the world, etc. I don't really know why I click with those particular people versus others who might share similar interests.

In fact, if you look at my best friends from different friend groups, they don't really have that much in common with each other (one is heavily into sports, one is heavily into classical music, etc), so that's why I'm a bit unsure as to whether it is feasible to create an algorithm to find whom you will click with.

For finding people you share interests with, today there is meetup.com. You may not click with them, but it is one place to start looking if you're in a new city. What I am asking about in the OP though is different (although I can see that any algorithm that tries to find people you click with will look at, among other things, your interests and hobbies)

You have a point. But look at the example you provide: meetup.com. It is hopelessly crude. The bad part is that it lets you connect with groups of people (clubs, etc.) rather than actual people. For example, when I enter Go, I get (among tons of unrelated content brought up by uninformed text matching) the Chicago Board Games Club. Cool, but I could have found this using Google (in fact searching for "Chicago go game" in Google turns up many better matches). This is definitely now what I need.

As for your main point: My guess is that you are bonding with these people because you are sharing a common outlook to life, a common way of thinking. This would be hard or impossible to discern automatically using algorithms.

the interesting thing I saw when I was looking a few years back was that OkCupid was full of people in open relationships- it seems those people would be easier to successfully serve, simply because coming back to the site after finding someone doesn't indicate failure in an open relationship. I guess the same is true for casual dating where neither party is looking for a long term commitment.

The algorithm is quite doable. A good-enough algorithm would be fine -- I have higher standards for those I exclusively date than for those I grab an occasional beer with. An opportunity to meet some new and interesting people would be great.

The viral loop probably lies in encouraging people to sign their existing friends up so that we can better classify the kind of people a user gets along with. I think people would use it, given that it had some mild game mechanics and some real social benefit. I'd push hard against making it in any way useful for dating and hookups at first, otherwise it would quickly get a stigma.

I had an interesting chat with some salesguys a couple of months ago, where we talked about an application that lets you say, "I'm in Denver on a layover for four hours. I'd love to grab a beer with somebody who sells into the enterprise software space and has some time to chat." That's one possible revenue source. Permission marketing is another one.

If I were to go the social-media-startup-route, this is the problem I find most interesting.

http://www.njl.us/social-networking-but-the-other-way-around

I've seen people on twitter with a lot of followers doing the "I'm in such and such for 4 hours lets meet" invitation, though of course that only works for people with a lot of followers. So I think there is an opportunity there.

That would be a great feature for Foursquare. They could add a profile for users, and let you opt in to sharing that with other users who have checked in at a certain place. It could be interesting to see who's nearby with a similar collection of badges and interests.

Meet Gatsby does this with foursquare (http://meetgatsby.com/)

Doesn't Loopt do this?

> "The viral loop probably lies in encouraging people to sign their existing friends up so that we can better classify the kind of people a user gets along with."

So something like: 'Hey friends, some of you should spend your time to take this online questionnaire, so that I can benefit by having this site find me some other people to hang out with besides you lot. Also, if you think it is a jerk move for me to ask you to help me replace you in my life, then the site might be able to help you find someone similar to me who can fill my slot in the social circle.'

More like, "Lets find interesting new people together."

One problem that springs to mind is that while a romantic relationship is usually one-to-one, a friendship is often part of a circle of friends - even if you'd get along well with somebody on their own, there's no guarantee you'd fit comfortably with all their friends, to say nothing of whatever group dynamics might be involved.

If there was an eHarmony-style site for friendships, it would probably involve mining social networks like Facebook for more information than just a simple personality quiz.

Maybe I'm old-fashined, but "a site for making friends" sounds a bit creepy to me. I would expect it to be populated by crazies mostly, so not very attractive.

A slightly different approach could be to have a site or sites for people with some niche interest to meet each other. I don't know, carpentry hobbyists or amateur blues players. It seems that Meetup already covers this? Maybe you could see if there are some niches that are not covered well enough by Meetup, and focus on these.

looking for other people to pursue shared interests would be one of the main purposes behind "a site for making friends" - at a glance Meetup looks very geared towards structured group activity and less towards "hey judging by your profile/blog we have very similar tastes; fancy a drink?" ad-hoc approaches.

The other completely "non-creepy" case where someone might want to use the site is when they've just arrived in a new city/country (though again there are plenty of expat-oriented forums and social networks) or people that were looking for transitory conversation with likeminded people whilst passing through (essentially couch-surfing without the couch).

yes, you're also going to get the "people who aren't very good at making friends" and the "people that like to exploit the vulnerable", but they haunt the dating sites too.

MeetUp.com does a good job of this, albeit in a different format. Here in Raleigh, you can always find a pick-up ultimate frisbee game, a developers meet-up, or whatever else you're into.

I think the idea has legs, and the fact that people use MeetUp.com is a testament to that. Finding connection is definitely something people want. If the service existed, I'd absolutely try it out.

Any such site will immediately be used by people to attempt to find dating partners under the guise of a friendship, defeating the point if it being "friends only". I know someone (female) who recently tried to use the "platonic" section of craigslist to find some new friends, only to find later that all the people who contacted her (a few emails in) started down the path of "I'm bisexual". Some made continual references to how cute she was, but how they understand she isn't into that sort of thing, and would "respect her boundaries"...

This. I would add that meetup.com is a fairly decent avenue for making friends. At a minimum, folks you meet will have a common interest, and you'll be less likely to face the problem above (though let's face it, any social activity can manifest this problem.) There are also generalized meetup groups more oriented towards networking or simply making friends, which adequately fill this niche, IMHO.

I don't think going college or going on a cruise are that difficult places to meet people. In fact I think that holidays in general and college are easier than every day life since people often have few friends with them and people are more relaxed and open to making friends.

For moving to a new area, I would expect that real life networking events are best around a common interest or joining sports clubs.

I think eHarmony and similar sites are aimed at matching two people together, where as male friendships tend to based around groups of friends, so I'm not sure how you would modify it to fit. Might work for women though.

ya. being friends with someone is a social thing. If you can't befriend me in real life, I probably don't want to be your friend. (or rather it just would not sustain itself)

Also, at least guy to guy, you don't really go out looking for friends. That's just weird. You just happen to do stuff and gradually, subtly, you become friends =)

I don't think most people remember how exactly you become friends with your friends. That's indicative imo. A forum for "friendmaking" is just weird.. as in it probably wouldn't actually work due to the dynamics.

Music, movies & books are good indicators of taste and therefore you can use them as parameters in similarity search on user-user matrix. It is called nearest neighbors search or clustering in academic jargon and there are many possible solutions, mostly based on some sort of matrix decomposition. Look at this one for example, the winner of Netflix challenge: http://public.research.att.com/~volinsky/netflix/BellKorICDM...

This is interesting, since I've been thinking for a while that your Netflix queue might serve as a decent proxy in tests about whom you might click with (either romantically or just-friends).

If I think of the things my friends and I have in common, of all the aspects of their lives (interests, hobbies, politics, etc) maybe the one where we have most overlap is in the types of movies we like. Maybe this is not supported statistically across the population, but I would be curious to see a study on the correlation between the similarity of two people's Netflix queues and a measure of how much those two people click as friends.

The similarity of the queues is easily quantifiable, but one would have to come up with a quantifiable measure of how much two people click. If they are friends on Facebook, you could see how much they interact: the higher the interact, the higher they click (of course this may not always be the case).

Two thoughts.

First: "when you don't create things, you become defined by your tastes rather than ability. your tastes only narrow & exclude people. so create." --why

Second: I've had many fulfilling friendships with people who didn't enjoy the same music, movies, or books. I think that's a really crappy way to make friends, and a really good way to encourage people to isolate themselves even more into cultural echo chambers.

I genuinely believe that you're making a mistake over here; you're assuming that machine learning is a magic bullet that can solve anything. I am afraid that isn't the case. Even for a satisfactory level of functioning you need to factor in these "meta" variables;

- Emotional stability of the user: You need to be able to judge if the user can form serious long term relationships before recommending him or her.

- Level of Emotional maturity: Maturity is really different from stability. Stability means that you'll be roughly the same a week from today whereas maturity applies to your ability to handle problems effectively and rise above petty issues.

- Willingness for Engagement/Commitment: Your users are in a portal, but the levels till which they are willing to use your service vary, and it isn't something you can ask in a questionnaire.

- Acceptance: This is a part of emotional maturity, but it deserves a solitary mention. How do you judge if someone is willing to accept divergent points of view and learn from things?

- Perspective: What does the user seek qualitatively in relationships?

I think that one day this might be possible, but you're dealing with one of the greatest unsolved problems we have ever faced. Making a computer understand things.

Further, I believe that even though you might succeed someday it won't work for people all the time. Why? Because emotions are really hard to parse, and what I seek in relationships is far beyond common interests or anything like that. I seek kinship, love and understanding. How can you possibly judge and match that without having some degree of emotional understanding? This is yet again one of those great unsolved problems.

On the other hand, there must be a solution beyond simply matching interests from a long checklist. Maybe you should approach it like Aardvark. You could make a friend of a friend based contextual system. Imagine a giant grid of people remotely connected to each other through friends and you parse the daily routine conversations between users and create a personality profile using current psychometric testing (it's seriously inaccurate, but we need a shot in the dark). After that you offer to match them together anonymously at first (this is important) and then see how they hit off.

Yet again I doubt that it will work at all due to privacy issues and moreover it will be too computationally expensive to justify...

Why would you need to go through all of that for a FRIENDSHIP? I don't think you read his post.

Trust me I read his post.

Any relationship I engage in I take it very seriously. I want to be there for people and love them. That's what it means for me.

On the other hand, I was trying to list what happens in someones mind when they try to decide if someone should be close to them or not. The list must be pretty inaccurate, but it was more to show a point than to form a blueprint.

P.S. - I know that I tend to take stuff too seriously, but I just need to emotionally connect with people. So, I guess that YMMV.

I think most people do, just unconsciously. We don't keep a mental check-list going in our head. But if after interacting with a person for a while, we can say "I dunno, something about him just rubbed me the wrong way," then they failed an item on our implicit list we may not be conscious of.

Yeah, I think this is a little too complicated. Only people who are looking for friends would be on the site and I think if you supplied them with a list of people who have similar interests, are around the same age and describe themselves similarly, then people should be able to find friends. Of course, you would need to have a huge userbase.

"You can choose your friends and not your family".

1. I don't think an algorithm could be developed to find friends. On the other hand, I think people make friends on game sites where, they strike out on their own to make friends. Even meet up sites, where you go to play golf with a bunch of strangers or go play chess.

2. I don't think this kind of site will get traction and people will not pay for it.

eHarmony exists because it makes finding your potential mates easier, it is for people who have given up on bars and friend's recommendation. Yes, there are other ways to meet people but the frequency of finding a good mate will be very less, I think.

I think people just become friends who get along with each other and don't look for friendship actively, like we look for dates on a dating site.

The concept of a site finding a friend has always bothered me, like facebook is trying to find a friend for me or saying "share more info to find more friends"...

The best thing you can do is launch a portal for like minded people to mingle.

Everything cannot be solved by technology.

>I think people just become friends who get along with each other and don't look for friendship actively

I think this is right. So the question is can you sift out attributes that identify compatible partners? Or is it more biochemical?

Yes, there may be some common attributes and I don't think it is biochemical when it comes to male friends.

I feel that someone can look at the problem differently. To make a site based on a common interest and get people to pay for it. The users of the site will find a way to make friends.

What makes you qualified to dismiss the possibility of biochemical effects?

for example, tribal identification via scent, status chemicals

Personally, I lean toward thinking that sort of thing is too complex and unpredictable to determine algorithmically, but then again, dating sites have evidently done well in the last 10 years, and analog matchmaking sites did well before that, so maybe. My problem with solving the friendship/dating problem algorithmically is that it seems like a good answer to the wrong question. My best friends are the ones that I have a ton of shared experiences with (going to high school or college together, knowing each other for 10+ years, etc.), and the best "dates" I've been on have happened serendipitously. I don't know how you'd address those dynamics with something as necessarily superficial as a questionnaire. If anything, I could see a questionnaire being a good screening device for people you almost certainly would not click with, but I think it'd be almost impossible to predict the opposite. Just my two cents, though.

On the point about serendipity, there is a dating site called HowAboutWe.com that focuses on spontaneous dates rather than matching.

US only (demands Zip code). Shame on them.

I feel like meetup is a "good enough" algorithm and found many friends when I moved to a new city.

There was a site called Simler (http://simler.com) that attempted to do this. Its goal was to help you find new people who you might be friends with. It worked similar to Facebook or Twitter, via status updates.

People tagged themselves with dozens or hundreds of arbitrary tags, then their engine would recommend people who share same interests.

The challenge they faced was getting anybody to come to their site. You are not going to get people who feel they are already "friend full" there, which means your percentage of "abnormal" people (hey, I'm one - people who don't make friends easily).

I think the algorithms will be easier to overcome than the problems with the self-selected population.

Personally I think the whole eHarmony model is dead (or dying).

I'm interested in people's social interactions and how that is changing with new people growing up with no digital prejudices against technology.

I believe social networking replaces traditional online dating. Especially the younger you get the more true this is. The eHarmony model tends to prosper over inflation of the details you put down.

Okay so the trend that I see happening in younger people is that traditionally Myspace actually took this space for finding new friends online with people under 18 (non-tech). Facebook has a social contract that you typically have to meet the person or heavily been in a person's social circle before becoming friends on facebook.

The site currently ruling making new friends online is Dailybooth and YouTube which is immerensly popular with under 21s. Not only do participants make "online friends" they usually translate into offline friends. I have personally observed a high percentage rate of meetups and "gatherings" with both these communities.

So what makes Dailybooth popular with making new friends. 1) Use of actual photos of the person on a dailyish basis: This allows for people to follow people that they would be generally interested in hanging out with. The use of images from a webcam also means that you can verify the person, that they are real, and again images are a very powerful tool in considering if you'll like the person or not.

2) The details on dailybooth has the standard social networking info and also the location. Dailybooth users when they gather or meet up with one another they tend to take a booth/photo of the meetup each tagging the other people. This drives other people to meetup in groups.

------------

In consideration of an algorithm I've been meaning to pitch Dailybooth on an idea to populate a user's photo stream with people that you can calculate based on the network graph of who the user has followed/followed back. And based on location, to try and increase the percentage of people making offline friends using the service. Which decreases churn rate and brand value.

So yeah, I'm heavily researching this area right now so if you interested in discussing it further my details are on http://benreyes.com

I could talk about this topic for days.

Interesting stuff, but you mention under-21s. I think this age group never has that much trouble meeting friends, since they are still in school, see a lot of people, maybe are less picky than older people in who they interact with, etc. So, historically, there isn't much need for a website for under-21s to find friends, but maybe this is not the case with today's youth.

I may be wrong here, but I think any site that would want to attack this problem and find enough people who would use it would have to do it for people who are older than 21, maybe people above 25 or 30? These are the people who move to new cities because of a new job, or move into a new neighborhood because they bought a house, etc

Sexual love is a human need as fundamental as food and shelter.

In America's individualist society, friendship is a "nice to have."

Dating sites get over the hump of "oh my god this is creepy" because their users are desperate. It also doesn't matter how well they work: desperate users will pay money for hope, even if their hope is unfounded.

Without that driving desperation, I don't know that a friend-making site would ever get traction.

I've tried match, harmony, yahoo, friendfinder, and tickle. okcupid actually has compatibility testing & matching that seems to work and that does not suck. No one else I've seen has been able to do that. The problem with okcupid, is that it seems that women would rather interact on the site instead of meeting.

"The problem with okcupid, is that it seems that women would rather interact on the site instead of meeting."

Use crazyblinddate instead of OKCupid. It's down now but hopefully it'll be back online in the next year or so.

"The problem with okcupid, is that it seems that women would rather interact on the site instead of meeting"

That's a problem with you, not okcupid.

When I go out to events and parties, I get their digits and a date. Perhaps it's my profile.

The vibe in your profile or your messages, yes.

"The problem with okcupid, is that it seems that women would rather interact on the site instead of meeting."

I met with a few women that I met on OkCupid, which was usually not preceded by long times of messaging. So far, a high match percentage usually meant we would get along with quite well. I didn't notice a "difference" between 80% and >90% matches in real life though, but it's a good indication to find people you might like.

Disclaimer, I'm located in Germany, and most women don't have the "replies very selectively" label here.

Edit: spelling

"Sexual love is a human need as fundamental as food and shelter."

That's preposterous. Which one could you live without?

If you are going to nitpick about this: as long as you procreate, you could die the next day due to lack of food and shelter and humanity would continue to evolve in this manner as it has been for thousands of years.

That wouldn't particularly affect me, would it?

It does, unless you somehow escaped the influence of thousands of years of human evolution...

Lots of people live long, happy lives without ever once making babies. That's not true of food, water, or shelter.

In fact, you could probably live a long happy life without even fucking. But food, water, and shelter are still necessary.

Tibetan monks can survive in 40 degree environments just by meditating.

http://www.news.harvard.edu/gazette/2002/04.18/09-tummo.html

Does this mean that shelter and clothes are unnecessary? No. Just because a handful of people are able to cope with extreme conditions without X doesn't mean X isn't a fundamental need.

Fundamental != you will die without it.

You're turning this into a semantic argument. The original discussion began with the question "Which one could you live without?". Whether or not sex fulfills some other evolutionary function is not the point.

Uh no. What you quoted was the reply. This is the original discussion.

Sexual love is a human need as fundamental as food and shelter.

Yes--sex is less fundamental a need than food and shelter. You can go without sex for years at a time, but it's hard to go more than a couple weeks without food or shelter.

Maybe it is __less__ fundamental, but the point of this discussion (which you are confused about) is that it is __still__ fundamental.

Maybe you won't die, but humanity wouldn't exist without it. And don't pretend that most people don't have a primal urge to have offspring, or at the very least, a sexual relationship.

No, the point of this discussion is the claim that "sexual love is a human need as fundamental as food and shelter". That's an exact quote. "As fundamental as" is a comparative phrase which means sex is on the same level of fundamentality as food and shelter in the hierarchy of human needs. Since food and shelter are necessities for survival on a daily basis and sex is not a necessity for survival at all, it follows that food and shelter are more fundamental needs, hence the statement under contention is false.

I was arguing using your incorrect assumption that the topic of the discussion was "which one could you live without", and tried to explain that the necessity to live was not the sole criteria for measuring the degree of fundamentality.

I myself believe it is just as fundamental as food and shelter, as some people are willing to give up both food and shelter to attain it.

Dating sites get over the hump of "oh my god this is creepy" because their users are desperate.

I think to imply that the general users of dating sites are "desperate" is not only judgmental, but completely inaccurate. Of couse there are SOME desperate users, but they are far from the majority.

I tried out Match 4 years ago because I was out of college and just moved to a new city, both factors which make it much harder to meet men I'm compatible with. I thought that by looking at profiles online, I could efficiently choose people to date who had the exact qualities I was looking for. My boyfriend (who I met on Match) used the site for the exact same reasons. I know at least a dozen other couples who are also normal, social people who met the same way. They certainly weren't desperate, nor were most of the people they went on dates with. Because so many non-desperate couples have met through online dating sites, the number of non-desperate users must be a pretty large pool.

This is precisely how my friends use okcupid.

Are they men or women?

Both.

Would this be possible as a Facebook app? That would help get past the initial chicken and egg problem any project like this would have.

I'm about an hour away from the nearest decent sized city so something like this would be useful for finding new circles of friends in my area. At the moment I meet new friends through existing friends. While I've met many great people this way I do find it's difficult to meet people who share particular interests that my current social circle doesn't have.

We're working on it at SimplyAdaptive We call it Inner Space Technologies. Will release a preview in a couple of month and shooting for a working website in 6 month. We're doing out of passion with no external funding at the moment for this project. If interested in collaboration/funding/other details please contact us. http://www.simplyadaptive.com/ist.html

This would be an incredibly powerful program if you could make it work.

But the "making it work" part is non-trivial.

Sounds like you are missing a hook -- something that pulls more people in

I keep bringing this up (I keep my drum on me 24/7), but I run http://EveryMentor.com/ and we're trying to be something like that but for folks that are looking for a more business aspect. Basically, its a place where you get a workout buddy for work in general. Kinda aimed at startups or single entrepreneurs that don't have lots and lots of contact as is.

First of all, lol at the drum visual.

It's amazing how much I can learn about my own craft by critiquing others. I feel like you have wayyyy too much text on your site

That being said, I think your section headings captured the information I needed anyway, so I signed up. Your distance filtering doesn't work. It's sorted alphabetically, not numerically.

Hi there. I'm working on something similar. We're not starting out with questions, simply with friends-of-friends whom you haven't yet met, but there are a lot of interesting next steps we are contemplating, that being one of them. Consider signing up for the beta at http://Kliq.in; we're launching next month.

I'm the cofounder of SubMate ( http://Www.submate.com ) and we want to introduce you cool new people in your areas based on your commute. For the moment , the website is pretty new but we plan to improve that algorithm quickly to allow you to make new friends !

Maybe its just that I'm getting grumpy and old, but I'd lose the "cool" in your pitch. For some reason, with "cool", it sounds almost desperate; without "cool", it sounds interesting.

as long as it doesn't look desperate (rather than going on eharmony for friends destination site, you'd give a service on your favorite socialnetwork/communities which proposes you //twitter"who to follow") ? For the "algorithm", why not partnering with Hunch a bit? Actually, it may be a good idea if it is an information site(rather interaction site not a match.com w/chat etc): people would just check which profiles around "could" become friends (based on near values/tastes/etc). It gives you an idea of who is around in your new town for instance. So that if you meet them IRL, you can say hello.

I don't know about you, but I wouldn't want to hang out with too many guys/gals who are 'just like me'. Their algorithm would need to account for that or adjust based on user input before the query.

You are very correct and most people feel the way you do even if they wont admit it. I have worked at sever large hookup and dating sites and all the problems are the same. The main one called the 90/10 rule is the root of all their problems.

I am building a dating site right now that eliminates this issue and introduces a totally different model.

Sounds really interesting! You seem to have some domain expertise as well.

Can I help you out? I've always wanted to do something in the online dating space.

I like the idea of organizing and expanding the idea of "Anyone want to grab dinner before the concert?". At least you'd have stuff in common, and you'd avoid the feeling of desperation.

Well there is certainly a market for it, there are a lot of lonely people, I doubt the ability of a questionnaire and algorithm to match friends together. There are too many variables.

There already is (or...was):

Facebook.

Back in 04 that's how people found friends before they went off to college; they would look people up in their class who had a common interest in music, movies, activities, etc.

I have been using EH for 2 months now. Every week I ask myself, "why couldn't this just be for meeting friends?"

I know a guy who's working on this: http://preacquaint.com

He should change the name.

But you definitely shouldn't, ever.

you don't like http://shirtlessfriends.com ?

This does seem to be the type of site I asked about in the OP. I'm curious to see how well it does when it launches.

meetup.com

I just joined after meaning to for a while, and it seems awesome. There's a hash run in Hong Kong tomorrow I'm thinking of going to and an entrepreneurs group I made a post to asking who wants to have a coffee in Central/Sheung Wan. I do a lot of my work from a cafe in Central/Sheung Wan anyways, so having someone stop by for an hour to chat would be a cool way to break up the day. Looks like a promising for a way to meet people into cool things, athletics, adventuring, business, etc.

eharmony sucks. If there were a site to make friends that uses an algorithm similar to eharmony's, it would be a complete failure.

Finding friends on the internet is going to become socially acceptable very soon. NicoNaco.com - list your values and get match-ups (e.g, startups, drinking, bowling, politics, dubstep...) I have found that my only true friendships are based around shared values.

What? Facebook isn't working for you?

All you can find are "friends" there.

Facebook is more for existing friendships, not finding new ones. From speaking with people, it seems most people reject friend requests from people they don't know.

Reply on news.ycombinator.com