Posix.1e?
Peter
I understand that the Posix.1e (1003.1e) draft standard on security
interfaces (and it's related standard on security utilities) was
withdrawn before it was formally completed. My question is: why? It seems
like many systems have implemented parts of the draft anyway and there
seems to be plenty of interest in the facilities Posix.1e was attempting
to standardize. Were the reasons for its withdraw technical or political?
I can't seem to find any information on-line explaining why the withdraw
occured.
Peter
Casey Schaufler
>
> I understand that the Posix.1e (1003.1e) draft standard on security
> interfaces (and it's related standard on security utilities) was
> withdrawn before it was formally completed.
That is correct. I was the final technical editor of the document,
and had the unpleasant task of requesting its withdrawl after
the completion of Draft 17.
> My question is: why?
In the end, only SGI and IBM cared enough about it to continue
working on it, IBM would not pay for travel, and twice in
Poughkeepsie was all I could handle.
More to the point, standards development fell off
of the list of important things for computer companies
right about 1995, and the security effort fell victem
to that.
There where a number of issues with the Draft itself that
didn't help. It should have been five seperate efforts
(ACLs, Audit, Capabilities, Information Labels, MAC)
rather that a single integrated document. The source
for the draft disappeared for a year and was only
partially recovered. Some sections where too ambitious
for their intended purpose. Too much was designed by
the working group.
> It seems
> like many systems have implemented parts of the draft anyway and there
> seems to be plenty of interest in the facilities Posix.1e was attempting
> to standardize.
In that aspect, the standard effort was a success,
even though the standard was not.
> Were the reasons for its withdraw technical or political?
Yes.
> I can't seem to find any information on-line explaining why the withdraw
> occured.
There it is, strait from the (old, bent backed) horse's mouth.
--
Casey Schaufler Manager, Trust Technology, SGI
ca...@sgi.com voice: 650.933.1634
cas...@pager.sgi.com Pager: 877.557.3184
Peter
In article <3CCECAB8...@sgi.com>, ca...@sgi.com says...
> There where a number of issues with the Draft itself that
> didn't help. It should have been five seperate efforts
> (ACLs, Audit, Capabilities, Information Labels, MAC)
> rather that a single integrated document. The source
> for the draft disappeared for a year and was only
> partially recovered. Some sections where too ambitious
> for their intended purpose. Too much was designed by
> the working group.
Given that a number of systems are implementing the draft... or at least
parts of it... do you think the matter might be revisited later? With
some real implementation experience behind us (as a community), the
creation of a standard might go more smoothly the next time.
> > I can't seem to find any information on-line explaining why the withdraw
> > occured.
>
> There it is, strait from the (old, bent backed) horse's mouth.
Thanks a lot! I appreciate your comments.
Peter
Casey Schaufler
> Given that a number of systems are implementing the draft... or at least
> parts of it... do you think the matter might be revisited later? With
> some real implementation experience behind us (as a community), the
> creation of a standard might go more smoothly the next time.
It's possible, although I believe the era of
standards-as-we-know-them is behind us. I don't
know what mechanism would be used, open source
adoption may be the best bet.