Seems like it would be better for the entire community, if you simply worked with other pool servers to accomplish the same goal.

If you support long polling, then the default upstream poclbm (m0mchil's) -- and soon, every other miner -- will only ask for work once every ~60 seconds.

Community standards like this benefit everyone.

Long polling protocol description

1) Miner initiates connection to the mining pool just as usual, requests getwork and starts working on it.

2) If mining pool does supports Long Polling, it should include a special header:
X-Long-Polling: /long-polling-url
where /long-polling-url is a path for long polling connection.

3) Miner starts a request to long polling URL with GET method and basic authorization (the same as on main connection).
This request is not answered by server until new block is found by bitcoin network. The answer is the same as getwork on the main connection. Upon receiving this answer, miner should drop current calculation in progress, discard it's result, start working on received data and make a new request to a long polling URL.

4) If all the nonce space is exhausted during calculation or 60 seconds passed since receiving the data, the miner should request new one by means of main connection. 60 seconds limit is set to allow adding new transactions into the block.