mike@tuvie (Inst.f.Techn.Informatik) (06/14/90)
I have been told that HP's aim is to have DomainOS, HP-UX (and possibly OSF) converge into one operating system. How is this supposed to work? Will they simply kill DomainOS and create some compatiblity box for DomainOS programs to run under either HP-UX or OSF? What I fear is that two of the features I like most about DomainOS will fanish, namely the //directory (its cleaner than mounting all hosts in the / directory) and the possiblity to have variable size swap space. DomainOS unfortunately still has lots of incompatibilities with UNIX (which from a UNIX user's point of view are simply bugs, not ``features''), but I hope that those will be gone within the next year or so (that i have been hoping for 2 years now :-(. In the meantime, I have been mulling over an idea which I hope will alleviate the problems with Apollo's incompatibilities: an unmoderated newsgroup for DomainOS patches to publicly available programs (comp.sources.apollo or comp.sys.apollo.sources ???). Talking about porting problems, can the latest version of Larry Wall's latest version of Configure correctly diagnose DomainOS (especially whenn running bsd)? bye, mike ____ ____ / / / / / Michael K. Gschwind mike@vlsivie.at / / / / / Technical University, Vienna mike@vlsivie.uucp ---/ Voice: (++43).1.58801 8144 e182202@awituw01.bitnet / Fax: (++43).1.569697 ___/
rees@dabo.ifs.umich.edu (Jim Rees) (06/14/90)
In article <1640@tuvie>, mike@tuvie (Inst.f.Techn.Informatik) writes: > I have been told that HP's aim is to have DomainOS, HP-UX (and > possibly OSF) converge into one operating system. How is this > supposed to work? Will they simply kill DomainOS and create > some compatiblity box for DomainOS programs to run under either > HP-UX or OSF? My guess is that eventually HPUX (pronounced "hippucs") and Domain/OS will both go away in favor of OSF. This won't happen for a while because there are plenty of big vendors (Mentor, for example) that depend on the old operating systems. They may go into maintenance mode, and in fact Domain/OS already has, more or less. Sr10 was the last big change in Domain/OS and it's two years old now. > What I fear is that two of the features I like > most about DomainOS will fanish, namely the //directory (its cleaner > than mounting all hosts in the / directory) and the possiblity > to have variable size swap space. > DomainOS unfortunately still has lots of incompatibilities with > UNIX (which from a UNIX user's point of view are simply bugs, not > ``features''), but I hope that those will be gone within the next > year or so You can't have both new features and compatibility. For example, the //netroot that you like so much breaks plenty of programs. Even the simplest things, like widening the mtime in the stat struct to 64 bits so you can have microseconds, will break old programs. The sad thing is that in most cases, the program was broken to begin with, but its defects only show up when you try to add a new feature to your operating system. People don't care about this, though. The only thing they care about is "your operating system broke my program." Even fixing bugs can be dangerous. Lots of programs depend on operating system bugs! One of the slogans for Domain/IX (remember?) was "Bug-for-Bug compatible with Unix" (that one never made it to marketing!) I'm personally optimistic. Although OSF/1 doesn't have many of the things that make Domain/OS special to me, I think OSF/2 with DCE will come pretty close. OSF/2 is probably about two years off (my guess, not OSF's), so that should give you a clue as to how long Domain/OS will be around at a minimum.
kgallagh@digi.lonestar.org (Kevin Gallagher) (06/17/90)
In article <1640@tuvie> mike@tuvie (Inst.f.Techn.Informatik) writes: >[stuff deleted] >Talking about porting problems, can the latest version of Larry Wall's >latest version of Configure correctly diagnose DomainOS (especially >whenn running bsd)? > I installed the latest version of Perl, patch level 18, on a 3550 with 8 meg, which has DomainOS, Aegis, and BSD 4.3 installed. I installed Perl while running under BSD. The Configure utility worked fine. You will need your Apollo BSD Command Reference manual ready so you can answer several of the questions correctly. I even turned on the Apollo C compiler optimizer (-O) and did not have problems, except it took about 45 minutes to compile the file eval.c, which has a very large switch statement. (The Apollo C compiler optimizer is very sloooooooow at handling large switch statements!) It generated an executable which passed the entire Perl test suite. (This is almost true. One test in sleep.op, I believe, sometimes passes and sometimes fails, but this should happen on ANY Unix system, I should think, because of the way the test script is written. It assumes that invoking a Unix sleep function with an argument of n seconds will always result in sleeping for n or more seconds. On the Apollo, my tests show it will sleep for a period greater than n-1 seconds and less than n+1 seconds. I find this to be nothing to get excited about.)