On Thu, 2003-05-22 at 06:18, Tom Scott wrote: > Barry deFreese wrote: > > Mark L. Kahnt wrote: > > > >> > >> > >> I'm hoping that between questions you ask and ideas you spark, maybe it > >> is just one or two outstanding epiphanies (not the new browser being > >> developed on gecko) that break the logjam on the Hurd and prove my fears > >> wrong about its current condition. I wonder if they do need outside eyes > >> and some fresh curiosity to inspire the next big step. > >> > >> > > It's kind of a curious group. I'm still confused quite a bit. I'll be > > the first to admit that I know very little about OS design (which is one > > of the reasons I'm tryin to get involved in the Hurd in the first place. > > But, they are working off of a kernel that seems to get little to no > > development while waiting for L4. Which from outside eyes almost seems > > to be being developed in a vacuum? The Cathedral?? :-) > > Fresh curiosity plus old tricks might get something going. enjoying the fruits > of his ill-gotten gains, bill convinced Yuri Gurevich, the creator of ASMs, to > move from Ann Arbor to Redmond. why not see if some of yuri's students > are interested in the model?: http://www.eecs.umich.edu/gasm > > Back somewhere in this thread someone mentioned that hurd was (supposed to be) > microkernel neutral. could someone explain that? > > -- TT My fingers offered up the "microkernel neutral" comment. It arises from the comments and nature of phrases I've encountered about the microkernel, and the general migration that has been made from GNUmach to L4, and that rather than the traditional nature of issuing calls into a kernel to do system tasks (scheduling, dealing with hardware, etc.) these are messages sent to various services handling those tasks. As such, so long as the microkernel provides a certain set of needed tasks (primarily task scheduling and switching, maybe memory management) with a certain syntax of interface, it should be able to be dropped in without any headaches. Conceptually, a front end translator could even permit Windows 95 to be used as a microkernel, but I don't think it would be a *desirable* choice ;) -- Mark L. Kahnt, FLMI/M, ALHC, HIA, AIAA, ACS, MHP ML Kahnt New Markets Consulting Tel: (613) 531-8684 / (613) 539-0935 Email: kahnt@hosehead.dyndns.org
Attachment:
signature.asc
Description: This is a digitally signed message part