As far as I know people won't be able to invite others, so there is no invite to share. You'll only be able to recommend (8) friends, and they will have priority on the others.
BTW: I am in the developer preview but I haven't received the invite for the real thing. I asked to the others in the sandbox but it doesn't look like there's any invite being received so far.
I can't really imagine a better peer review other than hacker news to be honest. However I wish I'd got a more replies when I tried to get skilled people to read this (expecially from the english mothertongue), but that didn't happen.
You say the internet is broken because authentication is a pain? You say there should be just one social network?
I didn't say the Internet is broken because authentication is a pain. I said the Web has been created to deal (and access) hypertexts. Nowadays the WWW is used for much more. People use it to shop, keep contacts with friends living far away, some use it to share their "lives" and the current state of art is far from being functional to the scope. To my personal point of view, it is broken by design when it comes to these scopes. And far from me trying to say there should be just a social network.
The gateway I idealize is not a giant bucket (or a silo) you'd fill with your stuff, that's exactly what I want to avoid. The "gateway" would just be a place you use to authenticate and authorize apps to access subsets of your data. Some of it (anagraphical data for the most part) would be stored on the gateway, and other stuff would be in much more appropiate places. Flickr, Twitter, Blogger, Facebook, et cetera. The model I idealize would be a common layer to build thing on, nothing more and nothing less.
Yeah, that's more or less what I'd like it to work, but I think using phones or other devices that may get offline would be unworkable (at least in the present). You could, however, act as your own gateway just like you can do with OpenID right now
Normal people don't care about a lot of stuff, but this doesn't stop research and development. The world would be a really boring place if we had to think solutions targeted at the "average Joe". I do care about this and if someone offered me a solution for these issues I'd buy it right away.
In regards to 2., you are right but today the same risk is multiplied on dozens of websites that I know basically nothing about. I don't know how they store and backup my stuff and I don't know how they deal it.
As per 4. OpenID and OAuth DO work (I use them a lot), what I tried to say is that they do not solve the underlying issue.
Yup, pretty much the info I managed to find on my own. The thing is interesting (being in the EU and all), and I e-mailed him to get some more detail (waiting for a reply) -- I posted it to share and because I wanted to see if anyone in the field knew him more or less directly
BTW: I am in the developer preview but I haven't received the invite for the real thing. I asked to the others in the sandbox but it doesn't look like there's any invite being received so far.