[mbell]

> Bad patents can get submited as many times as the holder wants, but the USPTO can only mistake a bad patent for a good one once, and it's granted.

The problem is that the USPTO doesn't even seem to want to make the decision that early in the process. They seem to take a stance of "unless it is blatant crap, grant it, and let the courts sort it out and/or wait for enough complaints".

I think its just a matter of man power, they don't seem to be able to dedicate the time/talent up front to do a thorough job. Instead they make a bet that most patents are meaningless anyway so grant liberally and only put in the real time when/if a patent becomes an issue.

The problem with this is that it strikes a firm blow to one of the key goals of the patent system, to protect the 'little guy'. Making a patent a big enough issue to draw review is prohibitively expensive for most little guys. It seems like it takes two or more behemoth companies locking multi-million dollar horns to draw their attention.

[Symmetry]

I think its just a matter of man power, they don't seem to be able to dedicate the time/talent up front to do a thorough job.

That's pretty much it. Congress sees the USPTO as a revenue source, so they aren't even allowed to keep all the money they collect in fees (and they aren't allowed to raise their fees either). And they aren't allowed to develop a backlog (they started falling behind in the 1980s, but Congress slapped passed new laws preventing that). And if they reject an application the application can be amended and resubmitted, and almost always is until it's accepted.

So basically the USPTO doesn't have the authority by itself to solve the issue, and we need a solution from Congress.