Slashdot Mirror

← Back to Stories (view on slashdot.org)

New International Standard: ISO/IEC 9945:2002

Posted by michael on from the one-posix-to-bind-them-all dept.

An anonymous reader writes "ISO/IEC and The Open Group announce international approval of the joint revision to POSIX® and the Single UNIX® Specification. More info here."

16 comments

  1. Do standards cost to much for the open source? by Anonymous Coward · · Score: 4, Interesting

    One of the problems with standards is the fact that they cost too damn much especially for free or open software users. I usually try to collect the last publicly available draft or copy that was used for voting and do my work with that. This has worked pretty well the the IEEE 802 standards and the SCSI and FireWire standards. Anybody have a link to such a copy of this standard?

    1. Re:Do standards cost to much for the open source? by Confuse+Ed · · Score: 5, Informative

      If you follow through a few levels of links from the artice, the standard is available online (for free) at:

      http://www.unix.org/version3/online.html

      You have to register (your name and email address) and agree to some terms and conditions to be allowed to read it.

      Interestingly they also thought it important to answer the seemingly pointless question : how many API's are there? with a bar chart showing the number in a variety of specifications!

    2. Re:Do standards cost to much for the open source? by steveheath · · Score: 3, Insightful

      Open standards are normally cited as the basis of OSS. Closed standards are the problem (see Word format), where there is no way to be sure you are actually going to be compatible. Open standards allow freedom of choice for the app-user, encourage competition on the basis of functionality and are generally a good thing(tm).

      Anyway, the link on the page can be followed to the spec, free registration required.

    3. Re:Do standards cost to much for the open source? by Anonymous Coward · · Score: 0

      Why not go to the opengroup's web and download the Single Unix Specification ? (Yes its free)

    4. Re:Do standards cost to much for the open source? by NeonSpirit · · Score: 1

      This may be because the common unix specification was goiing to be called SPEC1170, until it went well over 1170 API's when they gave up :-(

      --
      I will not be pushed, filed, stamped, indexed, briefed, debriefed or numbered.....my life is my own.
    5. Re:Do standards cost to much for the open source? by Dazza · · Score: 1
      As far as the increasing number of API's goes, I suspect this is simply more of the previous approach where the way to a 'single' standard was to include every other standard, leaving you with the same basic functions existing as several different API's.


      This is *not* standardisation


      Of course, maybe they've cleaned it all, and I'm talking out of my arse...

      --
      -- "I know that this is vitriol, no solution, spleen-venting, but I feel better having screamed, don't you ?"
  2. Excellent by drdink · · Score: 5, Insightful

    Personally, I think this is great. Bringing SUS and POSIX together will make things much more portable once people actually meet the standard. The alternative is to have 47,000 different standards that all specify different things and then when you, the programmer and/or user, build your system, you must choose which one you want to meet.

    --
    Beware, Nugget is watching... See?
    1. Re:Excellent by Anonymous Coward · · Score: 2, Interesting

      this is good, however, Linux (kernel) has never been entirely POSIX compliant. And worse, lots of programs written for linux aren't POSIX compliant and may require a lot of minor work to compile on other (posix) systems. Things like not including the proper header files (pre gcc 3.0 let you skip a lot of header files), or making assumptions (posix doesn't define stdin as fd 0. It does require that the STDIN_FILENO macro exist, though), etc can cause programs to compile happily, but fail to work as expected.

    2. Re:Excellent by Anonymous Coward · · Score: 0

      Everyone on IRC hates you! We in #trolls won't stop until you're off slashnet forever, smegmabreath!

  3. sus certified distro by Anonymous Coward · · Score: 0

    Any chance we'll see som SUS certified linux distros soon ?

  4. Well... by RAMMS+EIN · · Score: 1

    I think it's great that Unix flavors are still converging. However, with MicroSoft's big stakes in the desktop, server, and PDA market, many systems do not even come _close_ to being compliant. This might be good for big Unix vendors, because people can't switch to Windows, on the other hand, it's one competitor less with many remaining. My take is that GNU/Linux will take over the Unix marketplace step by step, which will lead to one new and glorified standard for Unices: GNU. However, that still leaves application developers with two platforms to support: GNU and win32 (and possibly Mac, if they don't adopt GNU). I wish corporations who didn't abide by standards would thereby push themselves out of the market, but in the MicroSoft case that seems in vain. Us developers are faced with the choice: boycott (reducing portability) or support (approving their violation of the standards). I'm hoping that people will see the light and ditch MicroSoft someday...

    --
    Please correct me if I got my facts wrong.
    1. Re:Well... by larry+bagina · · Score: 1
      openvms is certified posix compliant.

      Windows NT/2000 with interix or USF is ertified posix compliant.

      Sure, linux is open source, but the documentation is piss-poor. Check the BSD man pages. Check the new posix standard. then check linux documentation. A standard requires strict definitions of inputs, outputs, dependencies, side effects, etc. Linux can't provide that. glibc is slightly better, but it's not a standard.

      --
      Do you even lift?

      These aren't the 'roids you're looking for.

  5. To be published in mid December by andkaha · · Score: 2

    For those interested: The four parts of the ISO/IEC 9945:2002 standard will be published on the 15:th of December. But then again, those interested would probably already have seen this on the Austin group mailing list

    Hmmmm... I wonder why my submission of this thing was rejected and why it still showed up the next day as submitted by "An anonymous reader". I'm not anonymous.

    --
    It's 11pm, do you know what your deamons are up to?
    1. Re:To be published in mid December by Anonymous Coward · · Score: 0
      Hmmmm... I wonder why my submission of this thing was rejected and why it still showed up the next day as submitted by "Ananonymous reader". I'm not anonymous.

      It's probably because my anonymous submission was better than yours. :-)

      Of course, my first submission was rejected as well...must have been my bad grammar.

  6. Re:YAATHL by Anonymous Coward · · Score: 0
    that's actually a good point.

    Posix specs didn't appear overnight when a committee decided what system calls a "portable" OS needed -- they were based on a union of system calls from BSD and SysV.

    That means someone (some OS) needs to be a test-bed of innovation, and create new system calls (like jail(2) in NetBSD). Once their proven in the real world, they can be accepted as part of the standard.

    If linux becomes POSIX compliant, we lose this innovation.