All these questions have been thoroughly discussed on [EMAIL PROTECTED] Please search the archives for detailed information. I'll add quick notes as I'm waiting for something to finish right now and have a quick moment with nothing better to do.
It's worth noting that I'm not a developer, so I'm certainly not speaking _ex_cathedra._ On 2005 May 9, at 8:16 AM, Sascha Retzki wrote: > 1.) Do you plan to distribute several MTAs, like NetBSD currently > does? Or > do you already (first check said no, but maybe I missed something)? OpenBSD ships with Sendmail. Other MTAs are available as ports. Unless Sendmail is replaced by something written from scratch (and I personally have a completely uninformed hunch that this will happen, hopefully sooner rather than later), it's the only MTA that OpenBSD will ship with. > 2.) Do you plan to distribute PAM in base? No. OpenBSD uses (surprise!) BSD authentication, which solves the same basic set of problems. See bsd_auth(3) for details. > 3.) Did anybody look close enough on TenDRA yet? I'd like to know how > much > GNUism is in OpenBSDs base, building mechanism, so basicly it adds up > to: > How long do we/you have to wait till TenDRA can be used? Replacing the toolchain is a mammoth undertaking. Even if TenDRA is up to the task, it's not something to be undertaken lightly. I expect it to happen eventually, but not for a long, long time. Even then, it might not necessarily be TenDRA that replaces GCC/EGCS, but something else. > 3.2.) Even if the compiler/debugger/linker stuff runs on BSDLed code > from > the TenDRA people, do you roughly know how much software in base is > GPL/GNU, > and how much of that must be there? There's not much left; see /usr/src/gnu. Of course, even though the list it short, it still includes things like GCC and Perl, so ``not much'' is quite a relative term. > 4.) I saw lkm-stuff in your tree, do you want the same situation as > linux has, > like that "too much" is modulized, or do you want that API for > situations > where kernel mods are the only bearable solution? It's my impression that adding stuff to the kernel while it's running is anathema. Heck, even just compiling custom kernels is rather frowned upon. It's really a solution to a problem that rarely, if ever, exists in the real world. It's also a solution that causes far more problems than it ever solves. > 5.) What do you think about devfs? Will it be there in the near or far > future? I doubt it. There's little or no need; the traditional way works fine. Virtual filesystems of any kind are easy to slap together and very hard to get right. > 6.) Do you guys like X11R6? Would you remove it if $somebody comes up > with > some basic window-manager-alike basing on something simple like > svgalibs? > Or, rather, would you distribute that in base, too? Unix uses X for its windowing environment. If and when this ever changes, I'm sure The Powers That Be will consider the alternatives. Until then, this is like asking if people like using spoons for their soup, and if they'd consider using some spoon-like tool instead if it were ever invented, worked better than a spoon, and everybody else started using it. Cheers, b& [demime 1.01d removed an attachment of type application/pgp-signature which had a name of PGP.sig]
