The problem for me was that the example you gave (yourself) was way too meta for the typical end user. It needs to be more obvious what the problem is. For this reason "Does this solve your problem?" is a horrible opener.
Here's the process: (1) State the problem, (2) Propose your solution, then (3) ask for feedback.
Overall: Don't make me think. You're never going to get replies otherwise. A big textbox is daunting.
I imagine this almost as a streamlined "proposal" process. Lets say you get an RFP from Widgetia, a company that creates and sells widgets. You get the RFP (problem), then you write a proposal (comprehensive or not) to address the the problem. Because this is on the web, you can use text, photos, video, whatever. Just communicate the solution as efficiently as possible.
Then your job is to ask people to identify. "Does this sound like you?" "Would this work for you?" Don't put just one textbox. Put a couple -- in strategic spots in your presentation. You'll gather better contextual feedback as well as gather analytics on when/where people are commenting to gauge how effective your presentation was.
One answer: The built-in structure (the questions). I would make it so that there are 3 pre-built questions, with the option of adding more, but that all apps use those 3 questions. Or 3 questions and an open comment field. And then make it easier to answer them, and to go from product idea to the next product idea (so I can give feedback on 5 products in a minute).
But the main answer will be if you get some traction, ie. launch with a bunch of apps in there (email people who are building apps) and a bunch of people ready to review. I think that should be possible, loads of people love to opine on startup ideas.
It seems like HN echo chamber response to me. People are never going to tell you a potential solution works or doesn't. HNers will give you high level, and if the app is done will often go through it and nit-pick, but they are NOT anywhere close to paying customers for a vast majority of people (as valuable as their advice is).
As for the regular users (the ones who would be your clients' potential customers on the site), quite frankly they don't have the ability to extrapolate from a picture or even an alpha version if the product will serve their needs.
This seems a lot like feedback army at best.
What I would like to see instead:
Feedback army has the informal aspect of this market tackled pretty well. I'd like to see something a little closer to focus groups. My major issue with feedback army was that of the 25 reviews I got I think all but 2 were "yes this is good, yes I would use it, yes this was easy to use" - they have no stake in the project and often times seemed to be trying to tell me what I wanted to hear. Of the 95% or so who would "definitely" use the site, not a single one continued (no shock, but makes me question the validity of their responses).
What I would like to see is a way to pick 15-30 people as "evangelist testers" or something, people who you would perhaps pay, or would be offered some other kind of incentive to really think about the problem they are trying to solve and how your app fits into it. This would give those 15-30 people a real way to be involved in the app's creation process, and the app creator much more high quality feedback. But the focus has to be on quality not quantity IMO. I would be willing to pay a monthly fee for that.
But I wanted to focus this specifically on answering Steve Blank's questions in the "4 steps to Epiphany"
Update: Regarding KISSMetrics, I've been paying attention to Hiten's startup. They started off with tracking David McClure's conversion funnel, I thought I'd tackle it from the other end, using Steve Blank's customer development questions.
One thing I've learnt is that a rough mockup is not going to do to solicit good quality feedback. I'm going sleep on this and try something different.
Have a play with it. Here's a url to set up your own survey:
http://www.seotomata.com/beta.php
If this doesn't really help you with your startup, tell me what would, and I'll pivot.
By the way, since all the comments are coming from HNers, it's only fair to share their comments here:
http://www.seotomata.com/response.php?code=ab3kc