Gavin does not seem to understand the economics of the block fee. He says that it doesn't matter if there are room for many or few transactions, but in reality it's the equivalent of a Dutch auction. Nobody has to pay more than the lowest winning bid to enter the block, so nobody really has to pay as long as there is room for all the transactions. Even if there is a cost, the cut off price will most likely be less than 1/20 if there is room for 20x more transactions. Personally I never pay unless I'm in a hurry. Unfortunately this means that a 50% attack will become dirt cheap if the price of Bitcoins continues to go down, and/or the block reward has halved a few more times.
Regarding the previous forks they made the whole community scramble in panic to fix it and get everybody to use the same branch. I really don't understand how they can be used as proof that forks are ok.
Gavin perfectly understand the economics of the block fee. You do not seem to understand that a block size of 1MB hinders and blocks services from running on top of blockchain. I have already pointed out that project Lighthouse is having issue because of the 1MB block size limit. Nobody seems to talk about the subject and I don't understand why. Are you happy that the Lighthouse project is hindered? I'm definitely not!
Also what you don't seem to understand is that there will be services/project that will gladly pay for some fees for their transactions and maybe there will be people that will gladly pay a fee for having their transaction confirmed in the next block. Don't assume that if you don't do it then nobody else will do it. The current system limits a lot the development and the free access to anyone. Gavin is trying to open the gates for everyone to have access and you people want to keep them shut down and to only open them to the highest bidder.