Status as of Monday, May 4

[Steve Sokolowski]

Here's today's status:

• I spent a lot of the weekend fixing bugs in the mining server, as usual. For the first time ever, the error_log table does not have any mining server exceptions reported in it during the past 12 hours.
• We resolved an issue that caused some coins to be paid at the wrong rates (or paid the wrong coins) yesterday. The errors averaged out and no money was lost, although we apologize that some miners received bitcoins that they did not want.
• We need a way to post documentation that resolves the bfgminer hashrate issues that customers will see and read. The problem with bfgminer is that the system does not send new blocks to merge miners when the benefit is less than the money lost due to the work restarts. bfgminer reports (correctly) that these are partially stale shares. The problem is that bfgminer takes inappropriate action because it doesn't have access to the information the system has (that these partially stale shares are indeed optimal).
• Multiple merge mining was working in development yesterday, but then the system changed in some way that it stopped working. Getting single merge mining working took almost 120 hours, so I expect multiple mining to take at least 20 more.
• Chris continues to contact coin developers and add new coins. To do that, he needs to delete coins that nobody has requested for payouts. Keep an eye out for an upcoming deletion notice if you want to contest a coin that will be deleted. In the future, we may buy a cheap, desktop-grade computer for the purpose of hosting more daemons, since we don't need things like error correcting codes and can just reinstall the daemons and copy in our wallet backups if the cheap computer fails.
• Comkort will go live within one or two days.
• We are investigating the concept of "sell-only" exchanges for those exchanges that do not allow automated withdrawals. Bter and ccex are two examples. It isn't practical to offer payouts in coins only offered at either one of these exchanges, because the profits we would make are lower than the hours it would cost Chris to have to manually withdrawal the coins. These exchanges have indicated that they will never allow automated withdrawals, but manually withdrawing only bitcoins is feasible.
• Something interesting seems to be happening: there are some coins that were not being traded because their blockchains were stuck. We are now mining about 20 coins during any given 5-minute period, and the small hashrates given to each coin are keeping those chains moving along and increasing prices and trade volume. It seems that, at least with these small networks, the coins are reaching a value based on their merits rather than seesawing difficulties.

I will begin posting requests for comment in this forum tomorrow for non-profitability features. The first is likely to be a "worker dashboard," which means we need input from customers on what information about individual mining rigs they want to see. Keep an eye out for that.