I'd be in his position I'd would too ask to see some code or at least some data analysis supporting the design. You can't just propose stuff and expect the people reviewing it to do all the leg work. An implementation at least proves your design is conceptually sound. It's easy to forget certain aspects when you theorycraft, and having to implement at least the PoC certainly motivates you to keep it as simple as possible.
Not sure to which extent this is criticism to me. But I believe everyone has a part to play in this world, and should be doing what he's best at. My comparative advantage is in coming up with ideas and discussing them; and in unrelated work (to those who don't know me, my day "job" is in promoting Bitcoin in Israel). It's not in coding and empirical analysis - I'll leave that to others. This methodology worked quite well at the time I helped mining pool operators with implementing DGM. Perhaps the discussion I've started will result in this or a similar idea being implemented and accepted. But if not, so be it.
I'll clarify that I think Gavin's request is perfectly legitimate. I didn't exactly expect him to be so dazzled by the idea that he'd drop everything he was doing and start working on it.
It's not criticism directed towards anyone per say. In the course of my work with Armory, I get suggestions to implement this and that, but an idea that can be summarized in a single sentence can often demand 10k LoC. I'm much more inclined to look at a pull request than just some formulated concepts. As I said, having a PoC to support the idea has several advantages, one of which is to make the task of reviewers simpler, another which is to go through the most obvious optimizations right away. An idea without a PoC is not diminished, but an idea with a PoC is certainly improved. I felt like I should share that. It wasn't even an attempt to defend Gavin.
Obviously if you can find someone to work a PoC for your proposal, that would be fantastic.
You're an idea man, I'm a nuts and bolts guy, I can't help but look at this from my perspective. Your natural stance towards people with my skill set is "you don't sophisticate enough". My natural stance towards people with your skill set is "you complicate too much". This isn't about to change anytime soon, yet that doesn't make it a personal attack. Present a patient with some general syndrome to N different medical specialists, all in different fields, and they will come up with N different diagnosis. They're not all necessary wrong.
If you think there is some underlying ad hominem in my criticism of your proposal, that is not my intent. There are plenty of other sections in this forum which are ripe for this kind of rhetoric. I'm going to defend my point of view with every opportunities I get, I don't expect less from others. The intensity of the criticism may come across as unwarranted but that's only cause I'm genuinely interested in this discussion. That should vouch on its own for the importance I bear to theoretical research.