info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
[OT] Yet another Microsoft documentation inconsistency
24 views
Skip to first unread message
JJ
Jun 11, 2016, 3:23:01 PM6/11/16
to
These two TechNet pages describes the VERIFY command of CMD.EXE, but one of
them lies.
<https://technet.microsoft.com/en-us/library/bb491029.aspx>
<https://technet.microsoft.com/en-us/library/cc754978.aspx>
them lies.
<https://technet.microsoft.com/en-us/library/bb491029.aspx>
<https://technet.microsoft.com/en-us/library/cc754978.aspx>
R.Wieser
Jun 12, 2016, 3:19:00 AM6/12/16
to
JJ,
> These two TechNet pages describes the VERIFY command of
> CMD.EXE, but one of them lies.
Both can be considered to be true: The one one tells you the command exists
and what the setting it gouverns its *ment* for, the other one tells you the
setting is ignored. :-)
But yes, I would, from the larger description, also assume that the setting
would have its described effect. I would call it "lying by omission",
though that would not hold in court, as both pieces are clearly available
(as you've shown). :-\
There remains a question though: does the OS now never, or always check if a
write-action has succeeded (rather important when writing to a medium like a
floppy).
Regards,
Rudy Wieser
-- Origional message:
JJ <jj4p...@vfemail.net> schreef in berichtnieuws
1wfvvff87ktcc$.21sn6pqzqw5$.dlg@40tude.net...
> These two TechNet pages describes the VERIFY command of
> CMD.EXE, but one of them lies.
and what the setting it gouverns its *ment* for, the other one tells you the
setting is ignored. :-)
But yes, I would, from the larger description, also assume that the setting
would have its described effect. I would call it "lying by omission",
though that would not hold in court, as both pieces are clearly available
(as you've shown). :-\
There remains a question though: does the OS now never, or always check if a
write-action has succeeded (rather important when writing to a medium like a
floppy).
Regards,
Rudy Wieser
-- Origional message:
JJ <jj4p...@vfemail.net> schreef in berichtnieuws
1wfvvff87ktcc$.21sn6pqzqw5$.dlg@40tude.net...
JJ
Jun 12, 2016, 7:27:49 AM6/12/16
to
On Sun, 12 Jun 2016 09:20:49 +0200, R.Wieser wrote:
>
> Both can be considered to be true: The one one tells you the command exists
> and what the setting it gouverns its *ment* for, the other one tells you the
> setting is ignored. :-)
But the conflict remains. One states that it's ignored in Windows XP, while
>
> Both can be considered to be true: The one one tells you the command exists
> and what the setting it gouverns its *ment* for, the other one tells you the
> setting is ignored. :-)
the other one doesn't. The one that states it's ignored... makes it look
that the command may actually exist on Windows 2000/NT4, kaput on Windows
XP, then magically pop ups back in Vista and later.
> There remains a question though: does the OS now never, or always check if a
> write-action has succeeded (rather important when writing to a medium like a
> floppy).
failures. I've checked all disk related device drivers and there's no
reference to string "verify" except 3 kernel I/O functions which aren't
related to disk-write verify. The File API doesn't seem to provide any
support for it too - *including* transaction-enabled file operations
(ouch!).
Hopefully, nowadays storage ECCs are reliable enough.
R.Wieser
Jun 12, 2016, 8:01:19 AM6/12/16
to
JJ,
> One states that it's ignored in Windows XP, while the other
> one doesn't.
The latter one is what I ment by my "lying by ommision" remark: It makes you
*assume* that the setting it gouverns has an(y) effect, while it actually
doesn't.
> makes it look that the command may actually exist on Windows
> 2000/NT4, kaput on Windows XP, then magically pop ups back
> in Vista and later.
Pretty much, yes.
.. though more "is ignored on OS ***" webpages might be floating around
there ...
(personally I get the feeling the setting is ignored on all of them)
> On WinNT, written data aren't verified. All disk write failures are
> hardware failures.
...
have error checking, repairing and bad-sector management build in), but for
(something like) a floppy ? Not so much ... And those floppies where, when
XP was young, still in use.
2mbl0mksq4id$.17gsy0z9rcgth$.dlg@40tude.net...
> One states that it's ignored in Windows XP, while the other
> one doesn't.
*assume* that the setting it gouverns has an(y) effect, while it actually
doesn't.
> makes it look that the command may actually exist on Windows
> 2000/NT4, kaput on Windows XP, then magically pop ups back
> in Vista and later.
.. though more "is ignored on OS ***" webpages might be floating around
there ...
(personally I get the feeling the setting is ignored on all of them)
> On WinNT, written data aren't verified. All disk write failures are
> hardware failures.
> Hopefully, nowadays storage ECCs are reliable enough.
I can imagine that for the harddrives (which nowerdays are rather smart and
have error checking, repairing and bad-sector management build in), but for
(something like) a floppy ? Not so much ... And those floppies where, when
XP was young, still in use.
2mbl0mksq4id$.17gsy0z9rcgth$.dlg@40tude.net...
foxidrive
Jun 12, 2016, 10:00:29 AM6/12/16
to
On 12/06/2016 05:22, JJ wrote:
> These two TechNet pages describes the VERIFY command of CMD.EXE, but one of
> them lies.
>
> <https://technet.microsoft.com/en-us/library/bb491029.aspx>
This one is right. Back in MSDOS days we knew that verify didn't actually
> These two TechNet pages describes the VERIFY command of CMD.EXE, but one of
> them lies.
>
> <https://technet.microsoft.com/en-us/library/bb491029.aspx>
test that the data was written correctly either. In Batpower forum on
Fidonet there was discussion of batch files and command.com galore.
A MUF V1.7 file exists from Mr Bat Lang in that forum and explaining the
Microsoft Undocumented Features.
It was great fun then too.
> <https://technet.microsoft.com/en-us/library/cc754978.aspx>
0 new messages