Couldn't agree more with everything said in this article. Well done.
I want to reiterate on the fact that multi-tier demoing just does not work. While the prizes and judging are important pieces of every hackathon, they really are not the main focus. These events are about enabling hackers to show off their skills and experiment in a fun, safe enviornment. When you start grouping hacks into "the best" and "the rest", you end up with the focus being on the product rather than the people. And that fundamentally sucks.
A number of hackathons have done this to great success. If you want to expand a hackathon beyond ~40 teams, you do need to figure out a way to structure demos better, since less 2 minutes per demo is too little time to really show off what you built, in my opinion. And I don't think it's a poor idea to go beyond 40 teams for a hackathon like this (because then you have to start being super selective of who you let in, which sucks for teams who are new to hackathons).
I totally agree that short demos don't do these hackers justice and that being selective on admission also not an attractive option either. In fact, I'm not sure if there actually is a best answer for how to handle the situation. On some level, you always end up having to pick the lesser of three evils and somebody always ends up getting left out because of it.
If I do come up with a good system, I assure you, you'll be the first to know!
I want to reiterate on the fact that multi-tier demoing just does not work. While the prizes and judging are important pieces of every hackathon, they really are not the main focus. These events are about enabling hackers to show off their skills and experiment in a fun, safe enviornment. When you start grouping hacks into "the best" and "the rest", you end up with the focus being on the product rather than the people. And that fundamentally sucks.