There's been a lot of discussion about scaling bitcoin lately and a lot of different proposals have been made. One idea I haven't seen mentioned is to increase the frequency of blocks. There's probably a very good reason why that is a terrible idea, but I would like to understand why. It seems simple enough, just divide the difficulty with say ten so that blocks are created on average every minute and also divide the block reward with ten. You would increase the throughput tenfold and blocks would be smaller. Of course this would be a hard fork and require a very large consensus, but it seems like quite a straightforward solution. Plus you would receive confirmation on your transactions much faster.
Why wouldn't it work from a technical standpoint?