1. Forum
  2. FidoNet
  3. aus.computers

  • Microsoft open sources DOS 1.0

    From Axel@3:633/10 to All on Tuesday, May 05, 2026 10:04:47


    <https://www.zdnet.com/article/microsoft-open-sources-dos-1-0-much-more-than-the-code/>

    --
    Linux Mint 22.3


    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From JJ@3:633/10 to All on Tuesday, May 05, 2026 13:37:54
    On Tue, 5 May 2026 10:04:47 +1000, Axel wrote:
    <https://www.zdnet.com/article/microsoft-open-sources-dos-1-0-much-more-than-the-code/>

    All Microsoft's released MS-DOS source codes are not true open source. All
    of them are missing the source code for the boot sector bootstrap code.

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Tuesday, May 05, 2026 12:57:01
    JJ <jj4public@gmail.com> wrote:
    On Tue, 5 May 2026 10:04:47 +1000, Axel wrote:
    <https://www.zdnet.com/article/microsoft-open-sources-dos-1-0-much-more-than-the-code/>

    All Microsoft's released MS-DOS source codes are not true open source. All
    of them are missing the source code for the boot sector bootstrap code.

    Well, strictly speaking the boot sector bootstrap code is not part of
    MS-DOS, because it could boot any OS. But the fact that most of the time
    the boot sector bootstrap code will be *generated* by the relevant
    MS-DOS command (AFAIR 'SYS' in the old days) the point is rather theoretical/moot.

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Tuesday, May 05, 2026 13:50:39
    A liitle earlier, I wrote:
    JJ <jj4public@gmail.com> wrote:
    On Tue, 5 May 2026 10:04:47 +1000, Axel wrote:
    <https://www.zdnet.com/article/microsoft-open-sources-dos-1-0-much-more-than-the-code/>

    All Microsoft's released MS-DOS source codes are not true open source. All of them are missing the source code for the boot sector bootstrap code.

    Well, strictly speaking the boot sector bootstrap code is not part of MS-DOS, because it could boot any OS. But the fact that most of the time
    the boot sector bootstrap code will be *generated* by the relevant
    MS-DOS command (AFAIR 'SYS' in the old days) the point is rather theoretical/moot.

    Oops, that probably should be FORMAT (or FDISK for a multi-partition
    disk), not SYS. SYS puts the (MS-DOS) io.sys and msdos.sys files (and -
    version dependent - command.com file) on the already formatted disk/
    partition.

    Anyway, as I said, the point is theoretical/moot, because other OSs
    can generate a boot sector, MBR, etc..

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From JJ@3:633/10 to All on Wednesday, May 06, 2026 15:02:10
    On 5 May 2026 12:57:01 GMT, Frank Slootweg wrote:

    JJ <jj4public@gmail.com> wrote:
    On Tue, 5 May 2026 10:04:47 +1000, Axel wrote:
    <https://www.zdnet.com/article/microsoft-open-sources-dos-1-0-much-more-than-the-code/>

    All Microsoft's released MS-DOS source codes are not true open source. All >> of them are missing the source code for the boot sector bootstrap code.

    Well, strictly speaking the boot sector bootstrap code is not part of MS-DOS, because it could boot any OS. But the fact that most of the time
    the boot sector bootstrap code will be *generated* by the relevant
    MS-DOS command (AFAIR 'SYS' in the old days) the point is rather theoretical/moot.

    DOS is not an assembly compiler. It does not generate the bootstrap from scratch. The bootstrap binary had to be at least precompiled, and that precompiled bootstrap binary must come from somewhere. Or unless you're suggesting that, the bootstrap binary was made using a hex editor?

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Wednesday, May 06, 2026 14:09:18
    JJ <jj4public@gmail.com> wrote:
    On 5 May 2026 12:57:01 GMT, Frank Slootweg wrote:

    JJ <jj4public@gmail.com> wrote:
    On Tue, 5 May 2026 10:04:47 +1000, Axel wrote:
    <https://www.zdnet.com/article/microsoft-open-sources-dos-1-0-much-more-than-the-code/>

    All Microsoft's released MS-DOS source codes are not true open source. All >> of them are missing the source code for the boot sector bootstrap code.

    Well, strictly speaking the boot sector bootstrap code is not part of MS-DOS, because it could boot any OS. But the fact that most of the time the boot sector bootstrap code will be *generated* by the relevant
    MS-DOS command (AFAIR 'SYS' in the old days) the point is rather theoretical/moot.

    DOS is not an assembly compiler. It does not generate the bootstrap from scratch. The bootstrap binary had to be at least precompiled, and that precompiled bootstrap binary must come from somewhere. Or unless you're suggesting that, the bootstrap binary was made using a hex editor?

    All true, but my point is that the boot sector bootstrap code is not
    really a part of MS-DOS, so that code is not really "missing" from what Microsoft released.

    All you need is the binary code and you can get that from anywhere.

    Also other OSs can create (probably 'generate' was not the best word)
    a boot sector on a disk and some of them - noteably Linux - *will* have
    the source code for (their version of) the boot sector code.

    So I fail to see that there is really anything "missing" here.

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From JJ@3:633/10 to All on Thursday, May 07, 2026 03:29:42
    On 6 May 2026 14:09:18 GMT, Frank Slootweg wrote:

    All true, but my point is that the boot sector bootstrap code is not
    really a part of MS-DOS,

    Yes it does. You can't use boot sector bootstrap code from e.g. IBM PC-DOS
    or FreeDOS, to boot MS-DOS.

    You're probably referring to the MBR bootstrap code, which is OS
    independent.

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Keithr0@3:633/10 to All on Thursday, May 07, 2026 10:49:45
    On 7/05/2026 6:29 am, JJ wrote:
    On 6 May 2026 14:09:18 GMT, Frank Slootweg wrote:

    All true, but my point is that the boot sector bootstrap code is not
    really a part of MS-DOS,

    Yes it does. You can't use boot sector bootstrap code from e.g. IBM PC-DOS
    or FreeDOS, to boot MS-DOS.

    You're probably referring to the MBR bootstrap code, which is OS
    independent.

    https://thestarman.pcministry.com/asm/mbr/DOS50FDB.htm


    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From Frank Slootweg@3:633/10 to All on Thursday, May 07, 2026 12:24:43
    JJ <jj4public@gmail.com> wrote:
    On 6 May 2026 14:09:18 GMT, Frank Slootweg wrote:

    All true, but my point is that the boot sector bootstrap code is not really a part of MS-DOS,

    Yes it does. You can't use boot sector bootstrap code from e.g. IBM PC-DOS
    or FreeDOS, to boot MS-DOS.

    If you say so, but that still doesn't explain why you need the source
    code.

    Anyway, Keith' reference sort of gives the source code, at least for
    MS-DOS 5.0! :-)

    You're probably referring to the MBR bootstrap code, which is OS
    independent.

    No, I was/am talking about the boot sector. (I mentioned the MBR and
    FDISK in side notes.)

    (AFAIC,) EOD.

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)
  • From malxau@3:633/10 to All on Saturday, May 23, 2026 04:09:34
    JJ <jj4public@gmail.com> wrote:
    All Microsoft's released MS-DOS source codes are not true open source. All
    of them are missing the source code for the boot sector bootstrap code.

    I've been working with the 4.0 release. src\boot\msboot.asm should
    generate boot.inc which is the code included by format. I think it's
    all there.

    (...it's also quite nuts. It checks that the system files are the first
    two in the root directory, but loads the first cluster on the disk
    to boot from whether the files point to it or not. sys.com works by
    defragging to ensure the first few clusters are free, poking the global
    DOS memory to tell the allocator to look at the first cluster, and
    creating a file.)

    --- PyGate Linux v1.5.14
    * Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)