9 ms·

I don't know why this is so hard or requires such significant investment up front. How is making a decentralized Twitter service more of a challenge than, say,

I don't know why this is so hard or requires such significant investment up front. How is making a decentralized Twitter service more of a challenge than, say, IRC or Jabber? I know that the problems aren't exactly the same, but I would hazard to guess that they are on the same order of magnitude in terms of implementation difficulty.

thinkingisfun 14y ago
Especially since it's already been done: http://friendica.com/ http://friendica.com/ ^ And that is hardly the only project, and it could use improvements like everything, but it's one that surprised me when I first heard of it, because it really does a lot and, well, nobody ever mentions it..! One link of note, not that much there yet but surely something I will watch eagerly: http://www.w3.org/community/fedsocweb/ http://www.w3.org/community/fedsocweb/ If you wanna do it right, do it right.
sprobertson 14y ago
That was my biggest concern -- it would seem that if you're building a pay-supported social network, your scaling issues will be business-side long before they are server-side.
icebraining 14y ago
App.net is not actually decentralized.
thezilch 14y ago
So they have a runway before their maximum 10K paying users is only enough to keep a part-time operations team and servers up.</sarcasm> I have to agree with you though. This may be more than <insert weekend project> we always hear about, but not _much_ more. I'd imagined the product would be mostly built already. Maybe they wanted an injection of cash before getting a handful of servers, and we should see everything up and running within the week. Maybe they think they need to be Twitter-sized day one, which would be a mistake on their judgement.