3 ms·
The basic argument is "design a protocol (damn-it)". I agree with this point, I've made it here and I don't think I'm the only one. But everyone knows the Dia
by joe_the_user 16y ago
The basic argument is "design a protocol (damn-it)".
I agree with this point, I've made it here and I don't think I'm the only one.
But everyone knows the Diaspora folks went the "product first, protocol later" route and a fair number of people agree this approach. So it's a disingenuous little to act "oh, they made a basic mistake so it just won't work." They went a route, at least, that lots of people think will work. It's not like they used hard coded eight-bit node identifiers, had each node multi-cast its information or some other design decision would guarantee things couldn't work.
- joe_the_user 16y agoThe other thing with protocols is... Naively, on first glance, to me Facebook looked like just a nice interface on top of standard Internet functionalities. And there are standard protocols for all of these. * Messengering we have * Micro-blogging we have. * Selectively sharing information is pretty simple. Give your "friends" a password to your website and let them connect via SSL or Oauth or something. ... Nothing seemed that new. My first thought was, "why couldn't you just slam a UI/Application on top of this existing stuff and call it an open Facebook implementation"? Ah, but there is a funny thing I realized looking at the situation (and discussing it here). The unique selling point of Facebook and its open competitors boils down a "share and revoke" function. The umpteen Facebook users want to imagine they are "safe" when "privately" sharing their photos. They don't look too carefully into what "safety" or "privacy" involves and they probably don't want to look that carefully but they really want to have it. Facebook has flattered them into thinking they can just carry their real world intuitions into virtual space. Of course, this isn't quite "real world privacy" as a number of fired employees can atest. Thus the "competing Facebook protocols" aim to provide something like "personal DRM". Without going into the details of the inherent problems, I'll just ask, "why would you want to?". I would argue that rather than trying to simulate something that's impossible, that Facebook doesn't and can't give you, why not educate people. Allow secure sharing of information but start people learning about what makes sense to share or not. And sure, that's a big job but the Internet itself is kind of doing that. Looking at the situation as a whole, Facebook is leveraging people's lazy, foggy and "intuitive" notions of privacy taken from the real world. Anything Facebook-like is going to "leak-like-a-sieve" and the more people learn about the Internet, the more they will realize this.