Thanks again. I'm really not trying to conjur disagreement, as much as to illustrate the sentiment I harbor that filing a document called a provisional application can be an added source of uncertainty for inventors.
That said, please allow me to share my opinions relative to yours, viz.:
"When it comes time to claim priority, it is of course necessary to accurately assess inventiorship of the provsional but only to the point of ensuring that there is proper overlap with the non-provisional."
In my practices, I would never wait until time comes to claim priority, to accurrately assess the inventorship of a provisional application. The gap of time between provisional filing, and conversion to non-prov or re-Submission of beefed-up spec can allow memories to cloud, people to leave the company, inventors to die, etc. By filing claims in the provisional, I know exactly what the inveniton is, and who the inventors are. If someone else feels they should be an inventor, then they're asked what their contribution is to the existing claims, or new claims are added.
"At that time all of the questions about who is a proper inventor of the provisional can be answered in light of the claims in the non-provisional that are supported by the disclosure of provisional." Maybe, but not always are claims in the non-prov supported by the provisional. Your preference is a matter of style. As for me, I like to put inventorship issues to bed early, by documenting who contributed what and including at least one claim, which I find to be a simple academic exercise. I've never filed a case without a claim in it.
"If the true inventors turns out to be no one named in the provisional, then a correction of the inventorship of the provisional is in order. " True inventors - presumably are those who contributed to the subject matter of a claim that has been found allowable. Of course people can invent things, but not all inventions are patentable. Focusing on patentable inventions, the inventors of claims directed to same might not necesarily be added to a provisional after allowance. Consider, A invents X, and B, working for same company later reads the provisional spec that claims X. B modifies X to X2 and patnet applicaiton is filed naming A and B and claiming X and X2. During prosecution, X is found in prior art, but X2 is not and X2 is deemed patentable. Amending the provisional to name B would not be proper here.
Isn't this wonderful, so many potential scenarios. I did the above just for the mental calisthetics, but also to illustrate that provisional applications can add more uncertainty to an already-uncertain endeavor. For those who like to complicate their patent lives, I stand aside. Personally, I've concluded that by always including at least one claim in the provisional, that headaches can usually be eliminated. The spec is still there, just as if the claim weren't, and if the spec supports something else, having a claim won't change that a bit.
cheerio,
Chris Whewell