I think/hope my start up, Krewe (https://www.gokrewe.com), has found the right formula. I believe the key elements are access and comfort. Krewe places people into a group with five of their peers, making it comfortable to hang out at first. You don't have to single anyone out, it works kind of how things worked in your youth. And everyone lives within a half mile radius, so it's really convenient to meet up often.

As you say, having lots of stuff in common isn't usually the defining factor in what makes people "click." I think most people can get along with pretty much anyone, so long as they spend enough time together. When you're in school, that's built in since you have to show up to class everyday. Krewe makes it possible to hang out with your group everyday since everyone's within walking distance and can get together for a quick beer or whatever.

I thought this was a cool idea, and so I just signed up, but I was so alarmed by the process that I made an HN account to address this here.

Please correct me if I'm wrong, but you drop people into randomly assigned groups of people who live within a certain distance without giving the user a chance to screen that group for people they already may know and need to avoid?

This is such an obvious potential problem that it didn't occur to me that your service wouldn't account for it. In my case it's at least not a dangerous situation, although I'm betting the other person in my new group is really not going to be pleased to get this particular notification email. But I imagine overlap will be a problem, particularly when your userbase is still small.

I have to ask--are there any women working at your company? Anyone over 25?

This may seem like a nuisance question to you, but for many women I know (and for myself), the inability to screen potential groups is an immediate dealbreaker. It's a safety issue at it's most extreme, but it's also just common sense. For many women this is an absolute dealbreaker in a way men might not understand.

Thank you for very much for your feedback. I really appreciate it.

Yes, Krewe does drop you into a group as soon as you sign up and you don't get to see who's in it. Keep in mind that we never give ANYONE your location or email address. You also have the choice to leave the group right away and join a new group and you have the ability to vote people out of a group.

But I strongly believe the platform simply won't work if people are able to pick and choose who they want to be in a group with. It's not hard to imagine that men will only want to be in a group with attractive women. If people can choose, then they'll only choose "the best."

I believe that the issues you raised apply to essentially anything in life. Every time you go to the grocery story, you run the risk of bumping into someone you don't like. If you go on Tinder, you can come across your ex. If you join a recreational kickball league, you could be put on a team with your arch nemesis. Krewe doesn't make you interact with anyone, it just give you the opportunity to do so. You should always meet up with new people in a safe, public space.

With that said, I will be making some small changes that I hope will help address your concerns.

"Picking and choosing" is not remotely the same thing as "screening," and that you conflate the two indicates that you either didn't read my comment in good faith or you missed the point by an almost insurmountable margin.

I'd urge you to listen to the needs of your users. You managed to dismiss this out of hand in a way that implies you haven't thought much about women at all. Obviously you don't have to, and you wouldn't be alone there. But this is quite literally a case of a woman giving you feedback about something that many (if not most, by another large margin) women consider vitally important. (The response amongst my group of friends have so far ranged from "unbelievable" to "Jesus Christ.") Do what you want with that information, but don't dismiss it as "wrong" or irrelevant because you believe it applies to all areas of life. You don't have to care about female users, but know that it's obviously a choice.

ETA: in response to the comments below:

1. No one suggested that risk should be completely eliminated, or indeed that this was possible. You're responding to a straw man argument. Which...also doesn't seem in good faith.

2. If you don't understand that someone may screen a group for potential danger without applying further filters ("picking and choosing"), I'm...not sure what else to say except that you obviously have never found yourself in such a position. It seems trivially obvious to me that a user who has already self selected as someone who is looking for a service to match them up with new people would be perfectly capable of deciding to continue in a group after they'd screened it for potential dangers, since that's what they signed up for. People are capable of moderation. In this case, I don't want to be in a group with someone I've seen in a violent situation, and I would rather they hadn't gotten that automatic notification email, either.

This is really very basic stuff.

Please trust that I did read your comment carefully, agree with the issues you raised, and am working to resolve them.

I sent you an email, with an apology, and my plan to fix the problem. I want all my users to feel completely comfortable and safe using the platform, and will respond swiftly to any issues raised. I believe there are certain unfortunate risks that people, and women especially, must deal with in this day in age. I will do whatever possible to minimize those risks on Krewe, but I don't know if they can be completely elimated.

Edit: Response to stonecraftwolf's edit:

I don't know what else to tell you. I've apologized. I've said that I agree with you and your point is valid. I've said that I understand the danger you're referring to and why people would want to avoid being in a group with certain people. I've said that I'm working on a fix. If there's something more I can do, then I'll do it because I want this to work for you and for anyone.

You've done an admirable job responding to drive-by badgering with open-minded interest, patience, and kindness. Kudos to you.

I think part of the problem here is the fallacy that if you can't eliminate a problem entirely then it's not your responsibility at all. This seems to be a common approach for some tech companies that deal with social interactions, and it's obviously convenient to simply wash your hands of something, but it's an abdication of responsibility. it's also just not how life works. No risk can be completely eliminated; it doesn't mean we don't take whatever steps are available to us to mitigate those risks anyway. People wear seat belts, they buy insurance, they use all sorts of tools to help them mitigate risk. If you make the decision not to provide your users with the tools they need to mitigate risk on your platform (or wherever), you are very much making a decision about what experiences and what users you value. In our current culture, that often means it's a hostile place for women. Again, that's a choice, but obviously it has consequences.

Baron, I've read your responses again, and I do think you are sincerely trying. But I also think you have massively missed the point. I don't think you are unique in this respect; in fact, I think you have a lot of company, and that's...kind of the problem. I appreciate whatever efforts you can make, and I really hope that you can try to seek out feedback from as many kinds of users as possible.

I think we may have cross-posted our replies/edits. Honestly, I think the best you can do is approach it with an open mind, which you seem to be doing now. A big part of my irritation was the dismissiveness that I think originates in the common fallacy that I described in my other reply. But it's also...I mean. People--women, especially--have been talking about these sorts of problems with social networks for as long as they've existed. I think if you want your app to be useful for everyone, you have to consider the experiences of...everyone. The only way I've found to do that in my own work is by talking to as many kinds of people as possible, and asking them specific questions where appropriate. I've tried to provide you with some of those answers already, but I would encourage you to keep asking questions.

Can you give an example of an organization that does this screening process properly, as a role model?

My church, grocery store, and local bars, to give a few examples, don't give me an advance lists of attendees to scren before I show up somewhere. If I see someone I can't be near, one of us has to leave. Thankfully, this collision happens only occasional

I think he's got a good point. In what circumstance is 'screening' or 'Picking and choosing' in the context of his app not the same thing? What you propose would completely destroy his app for the reasons he already mentioned in his post. Have you not read his comment in good faith?

The issue that stonecraftwolf is raising regarding the platform is not at all unreasonable. I think there is a change that I could make that would allow users to leave a group with people they know they don't want to be in with before those people are notified that they've joined. So I'm going to be making that change and I think it will make Krewe better.

I find that a reasonable response, but people also join groups after you. So who gets to see who first? Why does only the new person joining get to screen the existing people and not vice-versa? Or if both happen, how?

I'm struggling to figure out OP's context (not sure why they're not just saying it) but one thing I can come up with is restraining orders.

It doesn't seem too outrageous to me that the use case of 'have been in a neighbourhood for a while and already have enemies' is just outside of the scope of the program, though. It has far more utility for people who have just moved to a new town and have no contacts at all.

I witnessed an incident of domestic violence involving the other person in my group. I don't want to go into more detail than that for reasons that should be obvious.

I disagree with your limited use case. IME there are many stages of life where you might need to refresh or replenish your friend group. Aside from moving to a new place, people complain about the difficulty of making friends after college, when their friends get married or have kids, if they're stuck in a job without peers, after a divorce, etc etc etc.

Good idea. I signed up for fun. No one else in my group yet though. Too bad the network effect is so hard to overcome.

Yep. It's still very new.

People have highly evolved mechanisms to decide who they want to be friends with. I doubt that any kind of software as simplistic as dropping people into random groups will work. Yes, people will make a choice if you allow them to, and remember, their primary choice is wether or not to use Krewe.

I know that not everyone is going to will be able to get along with who they're placed with initially. But I think that's fine for a couple reasons:

1) I didn't like some of my best friends at first, but the more I spent time with them, the more we influenced each other and the more we were able to get along. I'm sure that's the case with a lot of people, and that situation will happen with Krewe.

2) A group of friends is a situation where the whole is greater than the sum of its parts. Even if you don't like any particular person very much, you can still really enjoy being with the group. You get to help build a culture and a communal identity.

3) Krewe doesn't end when you're matched with your group. You eventually get to expand your and make dozens of friends in your neighborhood. So even if you don't like they people you're with at all, and if you stick with it, you'll get to meet people you will like.

4) People get to have multiple friends. One of the great thing about Krewe is that hanging out with your group takes almost no effort. As long as you don't totally detest them, you should find it enjoyable to go out and do stuff with them.

I don't think anyone can design some sort of algorithm that will match you to someone who will inevitably be your best friend. You have to actually meet someone and get to know them before you can determine if you want to hang out with them.

Cool idea. Just signed up!

I had an idea which was related to this. My idea wasn't to make new friends, but rather to figure out how you can be relevant to people you already know and grow that into deeper, better friendships. Honestly, I don't know how I can help my friends sometimes and feel they don't know what I'm struggling with, working on, etc.

Marketing page: http://www.littlequest.me

I had to put the app on hold because my friend dropped out on the project. It was in beta for a bit. Never officially launched. Still I hope to re-write it from scratch later this year.