is changing algo from scrypt still on the table? my vote goes to x11 like darkcoin..asic and multipool resistant, cooler video cards and lower electric bills.
Probably not... The way I see it is as the scrypt ASIC start to become more and more available CGA will become more and more valuable (due to more hash power = higher difficulty = less CGA per block). So we almost welcome ASIC because it will only help CGA become more elusive.
Any consideration to put a smaller cap on the coin?

What for? You wont see more then 16 Million in your lifetime. It will take a
minimum of 50,000 years to mine all 10 Billion. I don't see an issue.

I've raised this question a few times, this is exactly why I think it
is an issue.
Practically the first thing anyone asks when looking at a new coin is "What is the total number possible?" In the case of CGA the answer is "10 billion". By the time you have gone on to explain why it doesn't matter and because of the algorithm really there are only going to be a few million they have glazed over and moved onto the next coin. Obviously that doesn't apply to the wise people who are already part of the community, but I think it does to the
man on the crypto street.
If, to all intents and purposes, there can't be more than a few tens of millions then why not set the limit at say 40-50 million? I personally think it would do a lot to help people realise it really is a rare coin.