As an addendum to a claim which showed up quite a lot in the original KGS wishlist discussion. Basically it was claimed that the use of time in Go tends to follow the pattern of slower moves in the beginning, with moves speeding up as time passes (in opposition to Chess).

This is often the case, indeed, but I'd like to bring up two points.

The first is the question of how much of this behavior is dictated by the traditional time systems used: one is under less time pressure at the start of the game, so can use more time for each move. Would people play Go differently if they'd been using Fisher time with a small main time and a reasonable increment differently: fast fuseki, reasonable middlegame, and slow and careful yose? Which is more desirable?

The second issue is the inflexibility of saying the traditional pattern should dictate the flow for all players. I, and many other players I know, do not spend much time in the fuseki at all, having the opinion that our reading skills do not allow us to substantially improve our choice of move in the fuseki, over superficial reading and intuition. I'd much prefer to have the time available for reading out fights in the middlegame, or to try catch up the 10 points I lost due to my weaker fuseki, with some skillful yose. In fact, I think it is quite possible that the weak yose of many players can be traced to the fact that in most games the yose stage has to be rushed through due to time trouble.