Single point of failure = NON-centralization.
I mis-stated this, sorry; omitted the "de" from "decentralization".
OK, so now we're back to talking about reliability rather than
privacy? It's very hard to respond when the topic keeps changing.
Privacy is a feature: when it fails (due to human corruption,
hackers, et all), the effects should be limited by design. (This is
basic damage control.) The single point of failure, in this case, is
about reliability only to the extent that we can rely on "private"
communications *remaining* that way: it's still about privacy.
Akamai is a good approach (since the content is static, that works),
and local apps to import the JS to cache without contacting any
particular server is also a good mitigation.
-Shade
_______________________________________________
specs mailing list
[email protected]
http://lists.openid.net/mailman/listinfo/openid-specs
