What would you desire in an EA dating site?

One useful feature is maybe an option to list your institutional affiliations and block anybody else from those org(s) from seeing your profile?

But I'm also not sure how useful technical features are, really the thing that gets people to use a dating site is the other people on the site, so what you really want to do is to seed it with an initial group of people who are willing to use it (and make it either primarily gay, or have a balanced gender ratio)

I think the real struggle would be how to get anywhere near enough users to make the app usable - there are hundreds of copycat dating apps which don't place onerous restrictions on can use them and struggle to get traction, and you're talking about opening it to maybe 5-10000 people in the world.

So my first thought would be 'make the category more general'. It's not like I'm only interested in dating other EAs - and I also doubt my profile of partners is particularly typical among EAs, or that there will even be that much commonality in who we prefer to date. 

Tbh what I would really like is an app that takes the best of OKCupid before it was acquired and seemingly actively sabotaged by the group who acquired it (something with meaningful questions, the option to add more, filters that actually work, the ability to just view all people in your selected area and sort by match %), and then starts innovating (rather than making it more Tinder-like) from there. 

For eg, you could let users select the weightings of their own questions as well as specific tags that people apply to themselves (so that eg you could make the 'effective altruist' tag mandatory, a dealbreaker, or anywhere in between), that had better bot control, that let you view same-sex profiles without having to change your sexuality, that let you opt in to explicit content (while still having hard controls against people who posted it without marking their profile as explicit), that somehow algorithmically arranged singles events for people in the same rough profile bubble, and presumably much more cool stuff you could think of.

I'll answer with some technical recommendations, as I'm not personally interested in a dating app (I love the name, though👍).

I'd suggest staying away from easy-site-builders like Wix or Wordpress. These could seem very attractive to not-yet web developers and maybe you could build some kind of (really) minimal prototype in no time, but on the long run you will most probably be missing some crucial features you need. 

I might be wrong, because I've just used those tools (Wix and Wordpress) very rarely. But in my short experience, it's like building with Lego: you can build anything... as long as the pieces you need actually exists. And you'll never be able to build a Lego sphere (again, I might be wrong with this methafor. It's been some years since the last time I played Lego).

What I'd recommend is building your MVP using:

• a framework like Angular, React or Vue for the front-end: the "visible part" of your webapp.
• some service like MongoDB Atlas or Google Firebase for the back-end: to store the user data.

That approach will require coding, and if you don't know how to code (and don't plan to learn) you might need to seek help from someone who is. Otherwise, reading your "I am not a web developer (yet)", I suposse you might either already have some coding experience on a different field or feel confident to be able to learn whatever skills you need. In either case, if I was to recommend you a learning path it'll be something like:

1. Learn HTML5 and CSS. Just the basis. You don't need to be a pro to start creating something. You can go through the free tutorials at W3Schools.
2. Learn Javascript. At W3Schools you'll find high quality tutorials too.
3. Choose a framework and learn the basics. My personal favourite is Vue, but I don't have any reason to recommend it to you. Have a look, try and, when you've chosen one, learn to use it. There are so many talented people sharing their knowledge on Youtube, so I'm sure you can learn the basics there.
4. Atlas or Firebase? Not sure what'll be best for your application. I might go for Firebase for a quick prototype, as it can easily handle all the autethication stuff, wich seems to be crutial with the kind of app you're planning to build.
5. Play. Try to use all of these newly learned skills and mix them together. Start by something simple and stupid, so you can easily spot the bugs. Keep playing building increasingly less-simple  stuff (even if it's still stupid). Try combining things you've learn in interesting ways.

At some point, you might feel confident enought to start building the first prototype of your app from scratch.

I based these recommendations on my personal experience as a solo developer for the last few years. Being self-taught I might be missing some important points.

Courage!

Disclaimer: When I say stupid I don't mean to scorn. I just mean you don't need to build anything useful while learning. It's much more important you're having fun. Its usefulness lies in the fact that it helps you to learn.

I suggest exploiting Facebook's Dating App instead, roughly like so (still needs some testing; dm'd you, Affective Altruist): https://docs.google.com/document/d/1VTRO12Nsl3H9P7Zpx3mcyeQ1HWNapxkUlaf45xS5OcU/edit?usp=sharing

Updates:

• I own the domain effectivealtruism.dating, which currently points to a matchmaking bounty spreadsheet someone linked in the Facebook group "Effective Dating". Let me know if there's something better to point it to.
• I made a Discord server for those interested in this. It could either be used to matchmake people in the server or project manage building a prototype with volunteers who want to help.
• I am still strongly in favor of using a no-code builder to prototype something simple and iterate towards something better and more sophisticated. This is a passion project for me personally and I definitely don't think it would or probably should be funded by major funders.

I recommend that it be free and open-source. To quote Gwern:

A closed-source program or library can, by definition, only be worked on by its owners and those the owners permit; while a FLOSS program can be worked on by its owners and anyone in the world with bare access to the source. A closed program developed by a firm dies when the firm dies, and businesses lose interest or die all the time...

Open-source software is also more trustworthy for those who care about internet privacy, an area in which mainstream dating apps are notoriously bad.

You could probably make this happen faster and at a much higher quality by instead spending your time on finding funding (e.g. from the EA Infrastructure Fund) to pay an experienced full-stack developer to build the first prototype, under your guidance. Unless it's also your main career plan to work in web development, then you'll learn a lot from doing the project on your own. (This is from someone who has worked in web development for 15+ years.)

Interesting thread!  

I've been building sites (way to many to count) since 1995, and have coded sites that build sites, forum software, blog software, and so on, so I feel qualified to comment here.

First, yes, you are under estimating the challenge involved here.  I know I did!

Second, a key thing I learned (from many mistakes) is to not code anything until I had a solid plan for marketing whatever I was planning on creating.   I'm retired now, but if I had my career to do over I wouldn't even learn coding, but hire people for that, and focus on marketing.  These projects live or die by marketing, not coding.

That said, if you think you'd enjoy coding and want to explore it for fun, then don't worry about any of the above, and just enjoy your nerdy hobby.  Just be realistic with your expectations, and don't expect to have much traffic or many users unless somebody on your project knows how to build an  audience/user base. 

I'm way beyond dating these days, and 40 years in to a happy marriage.  From that experience I can guess that having two philosophical people in one house may be one and a half too many.   :-)  A highly EA person might find it wise to balance their energy with dates who are on different channels.    

So, you could perhaps start a dating site for EA people who are fed up with EA people.   :-)   Just kidding...