it's rational to want to do something about the situation. To remove digging altogether would remove Clams' best and unique feature, so it is not a reasonable option. As Bitcoin and most other cryptocurrencies have built-in reductions in their distribution methods, I believe it is reasonable to propose reducing dig rewards to half at (about) 1.5 years from the release of Clams, and continue to halve the dig reward every 1.5 years from now on. I'm not a programmer, but I imagine reducing the dig reward could be done by imposing a 50% (and later 75% and so on) fee on transactions from the original distribution outputs that gets paid to a burn address. Assuming the software is ready to go, this can be announced a month or so ahead of the fork to make things fair for everyone, including the large digger.
I also feel that something should be done, if CLAM is to remain something that people want to use. The current massive ongoing dig appears to be worth something like 500k CLAMs, and started when the active supply was something like twice that, and so it will end up adding 50% to the active supply.
Note that although it appears significant, that 500k CLAMs represents only around 3% of the initial distribution. There is another 12 million CLAMs out there still waiting to be dug up. There is no guarantee that the current 500k CLAM dig is the last big dig, or even the biggest one. There are very likely huge numbers of CLAMs waiting to be dug from other old wallets - think of the MtGox wallet, the SilkRoad mixer wallet, various other mixing services and other black market sites, BTC-e, satoshidice, and many others. So while it's possible that we get through this current dig and still have some investors brave enough to hold CLAM throughout, will they stick around the next time it happens, or the time after that?
I'm inclined to support the opinion of dooglus that something needs to be done. If it were to come down to some kind of a vote, the clamspeech part of staking txs could be used. (I say this as someone who stakes.)
Here's another idea to put on the table. Maybe digging could be slowed down in some way. For example, a coin from a block before X (where X is when the initial distribution stopped) can only be spent in a tx that stakes the coin. This would place a limit on the velocity of digging without actually taking away the ability to dig the coin.