It's good that this worked for them, but I think it's a bit questionable to assume that because it worked in this case it's a good idea (or even an idea at all instead of just a complete dismissal of planning and structure).<p>There's a lot of different ideas about how much is too much in terms of up-front design, but to me this goes a little too far to be called anything but chaos with a chance happy ending.
I'd say that your HTML prototype acted as your wire frame. It does seem to have guided the engineers hand. Nothing wrong with that though. I tend to do a similar process a lot on my own projects. Being solo, and not much of a designer, I'm anxious to jump into the HTML/CSS/backend process. Then when I'm done, I'll call in a designer to "make it pretty", and they'll usually follow a similar layout.
It was incredibly refreshing to read about this level of trust being put in the engineer/developer who needs to make it work right for the long haul. Version 2.0 is almost always the one you wished you launched with and this is a great way to get there.