@Hyperme.sh
I agree that spammers
with very large stake (and as you say, possibly little or even negative economic exposure; btw, this need not be direct shorting, etc., it could include having a positive exposure to a competing system) can disrupt by spamming. They could also disrupt by electing malicious or just poorly-performing block producers, especially given the natural and inevitable voter apathy problem where a large portion of stake does not and will not vote intelligently or at all. (It is quite plausible to me that a motivated 10-20% stakeholder could either vote all block producers or disrupt the voting sufficiently to cause other problems.) I'm not sure how much the spamming angle adds to this sort of vulnerability, but maybe it does. Overall, I'd rate this claim as unproven by both sides.
I take that is not quite being what Vitalik was claiming, but who knows.
To further demonstrate how nonsensical your stance is, consider that if the websites which feed data to the block producers are down because they can’t download historic blockchain data to clients, then in effect the block producing clients have been isolated and are not functioning. The ledger is a holistic concept.
I'm not sure what you were talking about here. The ledger was not down. Exchanges (which run their own nodes) were still able to transact. Other people running their own nodes (including myself) were able to transact. Some lesser known web-UIs running their own nodes were still functioning.
The steemit web UI (along with others that use the steemit-provided nodes) is essentially an application that uses the blockchain, and it is an important one, but it isn't the only one.