With all due respect, Adam (big fan of your work here), if the user hasn't backed up within a certain period of time, there's a link right at the bottom that says "BACKUP REQUIRED". If you click on that, it takes you to a screen to perform a backup. What's the problem with that? And beyond that, what about the interface reminds you of a very early Mastercoin wallet?
My point was it startled me, made me more wary and set the tone for the rest of my experience using the service that first time which should have been one of exploring something cool and instead was trying not to step on a landmine because somebody just gave me a vague warning that when I blow up they warned me. The only other wallet i've ever experienced that with is the early mastercoin ones.
And your note about the BACKUP REQUIRED link is another illustration of my point, it would have been trivial to give people the option to "BACKUP NOW" by link in the aforementioned warning but instead you could only agree and close the warning. I think the BACKUP button popped up after i'd been dicking around with import wallet function for ten minutes, if they're really concerned they should have forced me through the process right then. I also found myself scared away from brainwallet when I thought the capture and verification process worked really well once I got up the courage to ignore the warnings and try. The wallet makes strange judgements.
By the way, perhaps you're not aware that this is a reference design. Many of us understand the point of it to be a usable wallet for the community to jump start the new governance model, for early adopters and investors to use more generally, and at the same time a working demonstration of the features business builders can design into their own product/service offerings built atop the Bitshares 2.0 platform and aimed at the end-user masses. Obviously it's not perfect, but I think it's delivering on that promise.
The wallet is a bare implementation of a crazy featureset. It took almost a year after counterwallet was released as a bare implementation of a less crazy featureset with more time spent on UI no question, before they even had other people running copies of the servers and it was about that same time (a year after release) that my project developed
http://pockets.tokenly.com largely to be able to use the counterparty system without having the problems that had become apparent in the Counterwallet approach.
What i'd like to do is put a simplified, normal user friendly (no crazy exchange features or much of the other high level stuff) version of bitshares into our Tokenly Pockets application and let people carry and use Bitshares based tokens alongside and interchangably with bitcoin based ones. Will Bitshares be easy to work with by looking at the reference implementation for our needs? From what I can tell, no it won't.
As for the API, like anything else clearly there were/are some kinks to be worked out. But I really doubt it will be overly difficult for you to integrate Bitshares 2.0 into your platform.
Theres another thread on the forum talking about how many of the structural inconsistencies that were present in the 1.0 implementation were replicated or carried over into the 2.0 implementation. I'm thinking about trying to fund a 'worker' who could come from the community and be paid by the blockchain to do and maintain the integration work, then I don't have to care as much about how difficult the protocol is or isn't to work with because the protocol will be providing the labor and expertise to make it not my problem and get it done. I'm hoping Daniel will be able to point me in the right direction on that, i'm not really "getting" the process for doing this very intuitively