> I'll guess that when we're done all the nice features we like in BeOS > will be found in every other major OS anyway. You may be right, and if Linux/FreeBSD/etc. were to magically become as cool a desktop OS as BeOS, I might be tempted (only tempted, mind you) to go that way. The only other viable routes, in my mind, are Windows and OSX. The invasive registration process that MS is implementing for Windows XP and beyond rule that out for me -- that thing is *never* going on my home system. If that means I don't get to play the latest games, etc. then so be it. I'll be damned if I buy into the Big Brother system they want to build. OSX seems, in at least a few ways, to be sort of cool, but I don't want to be stuck buying unnecessarily expensive hardware that doesn't perform as well -- not to mention, tossing out the considerable hardware investment I already have. If they ever port to x86, maybe I'll think about it. ;) > We need to keep BeOS innovative. The best way to do that is to work > from the existing innovative code. - if possible. Which is why reimplementing on a kit-by-kit, server-by-server basis makes sense. Once you've got a kit/server at 1.0 (i.e., fully R5 compatible), you can immediately start extending it. And some improvements will naturally present themselves as reimplementation is happening. For example, it would be pretty daft to reimplement the net_server and *not* make sockets file descriptors, or reimplement the VM and *not* support memory-mapped files. > If not possible... I really don't know what I'd prefer. But I hesitate > to try to rewrite the existing BeOS. I afraid we'll spend too much time > just making a compatible opensource version and not have resources to > move the OS forward, keeping up with the rest of the world. I'm willing to cross that bridge when I get to it. If nothing else, there's an awful lot to be learned in the process of trying to do it. =) e