I fully agree with Eric. As the specification is broken it has to be
fixed. So it makes sense to fix the spec & schema as soon as possible.
If the changes are done late then it may break content.
Regards,
Thomas
-----Original Message-----
From: Fredericksen, Eric
Sent: Wednesday, July 02, 2008 11:50 PM
To:
OVAL-DEVELOPER-LIST@...
Subject: Re: [OVAL-DEVELOPER-LIST] Oval 5.4 draft, User Sid test
Looking at the official schema bundles we see that this problem is more
serious than I first thought.
The windows system characteristics XSD requires that user_sid_item
entries contain children named
user_sid
enabled
group_sid
However, the windows definitions XSD requires user_sid_state entries to
have children named
user
enabled
group
This looks to me like a clear and serious problem in the either the
schema or the specification or both.
If the names were the same there would be no problem. However, my guess
that this requirement could easily break content in any OVAL consumer
that expects that state objects and item objects have correctly named,
that is, equivalently named, element children.
So, am I to understand that there is no mechanism for fixing a broken
specification? If there is no existing content that follows this then
why not fix the schema, amend the documentation, and move forward?
At the very least this should be fixed in a 5.5 release and not wait for
6.0.
Regards,
Eric
P.S.
Re: security_principle - I agree wholeheartedly. :) entities
(principals)
can have rights and rules (principles) are for secure behavior. I gound
my teeth over that one for some time.
There are a few other teath grinders in the same place: there are
security principal rights that are partially camel case, out of sync
with the rest.
sedenyremoteInteractivelogonright
------------^
sedenybatchLogonright
-----------^
-----Original Message-----
From: Gary Gapinski [mailto:
gapinski@...]
Sent: Wednesday, July 02, 2008 3:22 AM
To:
OVAL-DEVELOPER-LIST@...
Subject: Re: [OVAL-DEVELOPER-LIST] Oval 5.4 draft, User Sid test
Thomas_Joy@... wrote:
>
> In the user sid test (oval 5.4 draft),
> user_sid_object is having
> - "user" as element name whereas it should be "user_sid"
> user_sid_state is having
> - "user" as element name, it should be "user_sid"
> - "group" as element name, it should be "group_sid".
>
> To make the draft uniform, i think the above name changes are minor
> ones.
>
> Please let me know in case there is any difference in my
understanding.
>
While I agree that the element's name is wrong, the change will
unfortunately have to wait for a subsequent OVAL version, as 5.4 is the
official (i.e., not draft) version as of 2008/04/10. Worse, probably a
major version, as minor versions strive to maintain backward
eccentricity.
A pet peeve of mine is accesstoken_test security_principle, which has
enjoyed remarkable longevity.
To unsubscribe, send an email message to
LISTSERV@... with
SIGNOFF OVAL-DEVELOPER-LIST
in the BODY of the message. If you have difficulties, write to
OVAL-DEVELOPER-LIST-request@....
To unsubscribe, send an email message to
LISTSERV@... with
SIGNOFF OVAL-DEVELOPER-LIST
in the BODY of the message. If you have difficulties, write to
OVAL-DEVELOPER-LIST-request@....
To unsubscribe, send an email message to
LISTSERV@... with
SIGNOFF OVAL-DEVELOPER-LIST
in the BODY of the message. If you have difficulties, write to
OVAL-DEVELOPER-LIST-request@....