I just played a game against your AI - I should've lost but it got stuck at the very end. I had a king and 2 blocked pawns, AI had queen, king and 1 pawn. It should've moved in for a checkmate at that point, but it just kept moving the king right and left on the spot.
Great read, but I think that if the VCs can resist the temptation of investing too much too early there does exist a positive outcome.
We might just see feature development moving away from teams in big companies into smaller, independent teams. These will raise small amounts of money from angels and micro-VC's and get bought out en masse for mild sums by companies looking to add features and products.
Don't get me wrong, lambdas are really useful and elegant, I just think they should be used when there is a purpose other than simply making the code succinct - such as limiting scope. I've seen people use them instead purely to reduce the number of lines of code, which I think in the end makes it less readable and harder to debug.
For languages that support it, use lambda functions judiciously - too much of them make code really error-prone and hard to debug. (Not the most wide-spread problem, but came across this today)
I think this article is just an original argument for bootstrapping. I don't think this guy is saying: "don't reach for the stars" - at least that's not how I interpret it. He's saying: "set achievable goals, lean how to execute, then you might just build a big ship organically."
Agreed, it's just that building stuff is cheaper and easier now, and more industries/SMB are going to digital book-keeping, so there is now more data to be gathered by more people.
In the realm of software, which is diffuclt to defend, I like how much of a barrier to entry these companies create just by virtue of collecting data no one else does. Even google can't really compete here right away, because in each vertical space data has its own dimensions and requires different methods. I imagine this is one of the areas of software still really attractive to big VC money. (Not that it ever wasn't, it's just a lot easier to build now)
The economy of scale here will be huge - if the founders of Imagine K12 build vertical expertise in promoting education startups (i.e. connections with schools, VCs who invest in education) it will be a lot easier for startups to do customer development and build/implement cool ideas. Our education system will benefit big-time.
I think whether you agree with the specific technical points or not, this is a great article. I don't know if "best practices" exist, but consistent practices do, and they make it a lot easier understand other people's code. (and impose a minimum quality standard on people not familiar with the tech yet).
For example, Adobe never really bothered with specifying best practices for Flex, so now every Flex developer's code looks different, and there is a lot of unnecessary trial and error. (Although I like Flex as a framework)