I already put in my time to upgrade the 3 year old client to the latest version of Bitcoin and creating the current website.
is the port finished? ie. should people be updating already? looking at the source:
* bip16 and bip30 activate 2014-01-1. this is already past. people running the new client could get forked. p2sh transactions are currently unsafe until the majority upgrade. actually the source comment is wrong. it says '2014-01-01' but it is actually 2014-01-02 according
to epoch.
* a patch will need to be made for old clients to update lock limits for bdb. without this old clients can be forked if transactions exceeding certain limits happens on new client. this was cause of the dogecoin fork and the bitcoin fork previously.
other than that i assume all is fine?
I completely forgot about those two bip, I had set it for the beginning of the year expecting when I performed the work that it would have been deployed by then. I wlll have to move that to later then.
I will be unable to update the old clients, so the bdb limits may be problematic. Refresh my memory again as to how Bitcoin solved it when the blockchain forked back in 2013?
Correct me if I'm wrong, but if we deactivate the two bip's then there is no hard fork. If so then one strategy is to release the new client in two steps. The first release could deactivate the two bip's with the primary goal of testing the merged mining and getting everyone on the new database. I think the main resistance to upgrading is uncertanty about merged mining. So it's safer to make that transition without the fixed upgrade deadline that a hard fork produces.
This strategy has less risk of a catastrophic fork if some miners won't upgrade. If they are very confident that merged mining is working there should be little or no resistance to the second release. Better safe than sorry. Opinions?