> I'd welcome anyone interested in PIM to work on it. I'd be great to > have proper kernel support that meets OpenBSD standard and have it > enabled by default. If you're interested in such work, I can help you > getting started ;)
Joachim, In our tree, all security or front-facing code must be maintained by an interested individual/group. If it isn't maintained eventually it will become a hinderance towards other developments, or error prone and therefore a risk factor. It gets marginalized by other changes. Then it gets the axe. Every part of the tree needs believers who commit their time to keeping that part good, or even better -- improving it. Especially relevant as minimum standards of quality rises. You'll see an errata posted later today -- I think it is somewhat related to this discussion. I believe there is a minimum mindshare required to keep code "good enough", and therefore present in the tree, but a greater mindshare is required to make it high and of increasing quality. The code the mistake happened in, simply doesn't have enough mindshare to have spotted the mistakes. Expanding mindshare is a hard social problem, so I'm glad you mailed in... hopefully you take the bait mpi has placed in the trap^W^W^W^W^W^W^W
