I think the biggest thing against Ethereum is that it doesn't exist.
One could say the same thing about TruthCoin. I've actually made the same argument about all kinds of whitepapers. Normally I would've rolled my eyes and said "yet another whitepaper...", but TruthCoin was personally interesting/valuable enough that I started thinking about how I might actually implement it (a line of thought which led to Ethereum).
I don't even see reliable way of forecasting when it will exist (needless to say, the fact that their website says it will be available in Q4 means almost nothing to me). It may never exist, or it may change forms.
Three months ago, I would have agreed with you. When Ethereum was first announced I paid little attention because I don't have the patience for other peoples' vaporware (I have enough of my own).
Since then, they have produced highly functional alpha clients on testnets. There are two testnets currently, one for go-ethereum and one for cpp-ethereum. But they should be merged soon, as compatibility is planned for the PoC5 release, at which time the binary builds will be available for download on code.ethereum.org (until then, it's not too hard to build the PoC5 dev branches from the github repos). Scripts compile, the EVMs execute contract bytecode, account states on the Patricia tree get updated, and the block chain grows. There's a contract debugger, decent UI, WebKit view with hooks.. real honest-to-goodness progress.
That said, Q4 still seems quite optimistic. But it's unusual for an open-source project to build the kind of momentum that Ethereum has (a real leap in progress, not empty hype for a re-branded clone), and I'm hopeful.
Hard to even assess something which doesn't exist. Even the testnet can't be assessed, because there would be no reason for anyone to attack it.
After the pre-sale/fundraiser, there will (supposedly) be bounties for attacking the testnet, implementing a hardened mining algorithm, etc. Further changes/hardening will mostly be lower-level, and won't affect the way DApp authors write contracts. That's the advantage of using a well-designed stack, "separation of concerns" means that DApp authors won't have to worry about lower-level stuff, giving them a significant advantage over programmers working with an entire alt-coin codebase.
I'm as uncomfortable as you when it comes to fund raisers, in fact one reason I gave TruthCoin benefit of the doubt was because there was no imminent fundraiser (unlike many other alt-coin projects/whitepapers, including Ethereum). If they aren't very careful then the Ethereum fundraiser will likely be a big ol mess steeped in controversy. But if MasterCoin, BitShares, MaidSafe, etc can successfully pull off fundraisers than Ethereum ought to be capable as well.
I do have some degree of anxiety about it, but at least it will pay for a crowdsourced security audit before the mainnet launch. I mean, if there were a TruthCoin implementation on its own special-purpose blockchain, who would pay for it to be extensively audited/attacked (before the coins start carrying real value)? It would be much easier to audit a contract script, because you're not worrying about lower-level blockchain stuff.