OSI, licenses, MPL

[Roland Turner]

Michael Stutz wrote:

> Forbidding modification of the license itself seems to make sense --
> you want parties to use _this one license_, not encourage a thousand
> or more incompatible derivatives be written and adopted instead. So
> you define the parameters of your open source corpus with a set of
> terms outlined in the proprietary, closed-source document of the
> license proper.
> 
> But this is also a problem with the current open source licensing
> model. What do you do when you find a bug in the license you want to
> use, and the steward of the license is unable or unwilling to change
> it?

You've found a limit to the whole scheme of using copyright to provide
controlled sharing (as distinct from simply placing a work in the public
domain). When you publish under an open-source license, you are
specifically intending to limit the freedoms of some people in order to
protect the freedoms of all. Given the desire to impose such
constraints, the freedom to change the license must, apparently, be
withheld or limited.

The trick appears to be to have the right steward(s) for the license.
Gnu/FSF seems a safe bet, Netscape seems a moderately safe bet. The
latter is perhaps less likely to treat its role as a matter of personal
integrity however.

You pays your money, takes your chances. Nothing is foolproof.

- Raz