Yeah before I get a clearer answer regarding this I would assume this is a failed attempt of minting based on time without centralized identification service. The designer probably grossly underestimated the impact of sybil attack, in this case it's an attack on the minting design (aka counterfeiting). It's part of the core foundation of a currency design. Any currency that cannot effectively deal with counterfeiting is doomed to failure.
They're not counterfeit-- they're real Timekoins and the bot-net could do whatever it wants with them. The problem is that the bot-net can take over the network and outnumber other users so they will be unlikely to generate coins. If there really is a 33,000 joins-per-year limit then it only slows down that process-- how do you tell the difference between a "normal" user trying to join and one of 100,000 bot-net zombies staggered throughout the year trying to join?
Of course this is similar to normal cpu users in Bitcoin no longer being able to mine coins themselves. But in Bitcoin the benefit is that network security increases as the difficulty level rises (and consequently the mining farms' coins can be understood as a reward for that work they are doing). In Timekoin it only means you have a bot-net with a huge stash of coins that isn't adding security to the network, nor did it expend much in resources to get them in the first place.
I don't see anything in the technical document that addresses this. But again-- this isn't the same bot-net problem that Bitcoin has faced because Timekoin doesn't require the proof-of-work which provides a barrier to entry for the Bitcoin bot-net (and which makes it unlikely that the successful Bitcoin bot-net operator of one would just destroy the coins for lulz).
Still, I could be missing some feature that mitigates this threat.
