1,361 karma · joined February 27, 2007
we made it last forever. i think 15k lasted 4 1/2 months for two people. our rent was 1800 in a SF. and the only reason it went that quickly is we had a month where we spent more than we realized by accident. it was really easy, the only thing i did less was go out for restaurant food and beer less.
(socialmoth)
the fb app has a different feel because of the high volume of posts. it is almost impossible to refind a post. we have notifications, but just not enough, and nothing as good as the old socialmoth.com. working towards it though.
the first week socialmoth on facebook was out, it doubled every 29 hours.
the most common feature request is "negative hearts" whatever that means.
balance=1
of the boxes, the first box always has a much higher load.
we had about 20k more users come in and use socialmoth than we expected the other day. that is MORE, not total. up until that point, the dot-com and the fb app were working on the same server and using the same database.
my focus has been entirely following the users, and the users are on facebook. also i should follow the money, and that is also facebook. i moved the fb app to a new set of servers sunday, but i haven't moved the dot com to the same servers yet.
i plan an relinking them soon, but i've been scaling again since then. and learning about proxy balancers (nginx vs mod proxy balancer vs haproxy - if anyone wants to give advice)
rkabir, if you ask a specific question about flow, i can tell you want i know and can make public. the sites popularity was due to the facebook users. nothing i made scaled well enough for the new popularity and i haven't gone back and redone all of the dot com code to match the facebook scaled/optimized code.
omouse, do you use socialmoth? new features are on the way. they are done, but not publicly shown yet. also, my facebook users are dying for new features and suggesting them, but they realize there is a lot of scaling work to do. (i love how nice my users are) sometimes you just need to take a break and make new stuff though.
scaling the first week was hell. now i'm doing more features.
they have the money to pay a couple hundred k for a domain. but it was after the fact, not before.
that being said. having the .com is a very smart move.