4 ms·
I really liked OKC when I was using it and think it's one of the best dating sites, but I have something of an anti-OKCupid story: I met my current SO in perso
by msluyter 16y ago
I really liked OKC when I was using it and think it's one of the best dating sites, but I have something of an anti-OKCupid story:
I met my current SO in person, at a party. It turns out that we have opposing political and religious views (we're both fairly moderate, so we're not polar opposites, but we do categorize ourselves as Dem/Rep, Atheist/Christian). Personally, I'm a little surprised that things have worked as well as they have, given that, but love is mysterious.
Surprisingly, we discovered that we were on OKC at exactly the same time, but due to the above discrepancies our match percentages were so low that we never saw each other. I'm not sure what moral one should take away from that, other than that your mental checklist of dating requirements may be less relevant than you imagined.
- JeffL 16y agoI always searched for "atheist only" in my online searches. Then a Muslim girl wrote me and told me I had good bone structure. 3 years later, we're engaged. (She's also very moderate in her beliefs.)
- rue 16y ago> I'm not sure what moral one should take away from that, other than that your mental checklist of dating requirements may be less relevant than you imagined. The take-away is that there is no "special someone". Just a fairly large pool of "soulmates" with suitable chemistries. The less discriminating you are about the more superficial stuff (biologically speaking), the better your chances.
- dkarl 16y agoUnfortunately, unrestricted searches generate way more hits than you can process by hand. I worked around this by applying subsets of search criteria at a time. For instance, instead of searching for an Atheist/Agnostic 25-35 making over $50k with an Average or Athletic body type who likes backpacking, exercises 3-4 times per week, and has a graduate degree, I would search for athletic Atheist/Agnostics making over $50k, then search for women with graduate degrees who like backpacking, then maybe widen the age range to 20-40 but apply the rest of the criteria, and so on. The challenge was to be as inclusive as possible without generating more hits than I wanted to process in one sitting. It was also an interesting odyssey into my own preferences, prejudices, and thought processes. Sometimes I wasn't interested in getting a lot of results. Instead, I'd sit down with a desire to explore a very specific search, which often resulted in zero hits. I'd fiddle with the search until I got a few hits. One day I met a really pretty girl (not available, unfortunately) who was Indian and kept thinking about how great it would be if she was going on my upcoming backpacking trip with me. So I did a search for South Asian women who listed backpacking as an interest, and got zero results, none at all in my area, even after I removed all other criteria (18-60+, 4'0"-8'11", all body types, etc.) I was intrigued by the result, so I tried other ethnic groups (no black women into backpacking, either, IIRC) and found that East Asian + backpacking produced a set of results that I could process in a reasonable amount of time. Voila, I found a few interesting profiles and ended up getting a date with one of them. Small result sets always intrigued me. Searching for women making over $150k turned up a very small set for my area, with a couple of interesting and surprisingly approachable-sounding profiles. When I broadened my search radius, I found one girl I would have messaged in a heartbeat if she lived closer. It was strange how often a small result set had a higher density of interesting and compatible profiles, even if the parameters I chose to create it seemed unrelated to my preferences. The >$150k search surprised the heck out of me; I assumed it would turn up a bunch of women I had no chance with and nothing in common with. But that was pretty typical for small result sets, at least the ones I tended to produce. Large result sets contained lots of interesting people, too, but the payoff of interesting profiles per time spent slogging through results was much lower.