At Mon, 15 Jan 2007 12:00:48 -0500, "Jonathan S. Shapiro" <[EMAIL PROTECTED]> wrote: > > On Mon, 2007-01-15 at 09:42 +0100, Marcus Brinkmann wrote: > > At Sun, 14 Jan 2007 07:21:38 -0500, > > "Jonathan S. Shapiro" <[EMAIL PROTECTED]> wrote: > > > Libraries are a tool that you should only reach for when the performance > > > cost of engineerability becomes prohibitive. > > > > You are conflating separate issues. If you want address space > > separation, then there is no reason why the C library can not spawn > > arbitrary helper processes to execute the algorithms desired by the > > user. I leave open the question for now if this makes sense in this > > case or not. > > Sigh. > > Marcus: when people speak of implementing libraries they generally mean > that the algorithm runs in the same address space. It is you who are > conflating issues here.
So let's separate the issues. There are two issues: 1) Should a linking (interpreter selection etc) algorithm be provided along with the program files or should it exist in the ambient space of the operating system? 2) Should the algorithm be run in its own address space or not? My main point is that the answer to question 1 should be that programs come in form of formatted data files which are interpreted by the instantiator according to an ambient protocol. At the current time, I don't care about the answer to question 2. Luckily, I don't need to care, because if the answer to 1 is as I say above, every instantiator can decide for themselves how they value the trade-offs involved. You are currently focussing narrowly on question 2, although I think I have made sufficiently clear in my last note that for me question 1 is more critical. Thanks, Marcus _______________________________________________ L4-hurd mailing list [email protected] http://lists.gnu.org/mailman/listinfo/l4-hurd
