[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: [mizar] PVS GPLed





On Mon, 15 Sep 2008, Robert Boyer wrote:

I was having a daytime nightmare about a possible world
in which some versions of a hypothetically liberated
Mizar reported that something was a theorem while
others said that it was not.

There is no need to liberate Mizar for that - it is reality (and sometimes a nightmare) - all sufficiently complicated software is (sometimes) buggy. And I also don't think that open-sourcing automatically produces more incompatible versions. Especially the Mizar library is a strong reason for staying compatible in this case.

Josef