James Carlson <james.d.carlson at sun.com> wrote:
> Joerg Schilling writes:
> > Unplanned is simular to planless and this is what I would call the Sun tar
> > extensions I have in mind. They are based on deprecated methods 6 years
> > after
> > the deprecation was official and 8 years after the fact of the deprecation
> > was
> > known. We live in the time POSIX.1-2001 and no longer in POSIX.1-1988 time.
>
> At a guess, you're talking about the trusted extensions.
Not only about Truested extension but also about ZFS ACLs.
> If so, then I think it's rather presumptuous to say that the team that
> did this was "planless." It seems to proceed from an assumption that
> supporting the latest POSIX standards is the only worthy goal, and not
> also supporting existing customers with existing data.
Using the term "the latest" is a bit missleading here. The POSIX TAR archive
format extensions have been added 1988 and did not change since then until
the standard approved in 2001.
So we are talking about something that is nearly 10 years around compared to
something that is more than 20 years in use.
Given the fact that the POSIX.1-1988 based extensions are not vendor tagged,
they cause a high probability of miss-interpreting in special as the Sun tar
implementation does not add a "finger print" (similar to what star does) that
would allow to identify even a POSIX.1-1988 based vendor unique extension in a
definite way. Another problem is that the rules from POSIX.1-1988 require to
unpack all "unknown filetypes" as if they were plain files while POSIX.1-2001
requires to ignore unknown POSIX.1-2001 based extensions.
> In any event, this claim is completely off-topic here. You need to
> approach the people who made those changes and talk to them directly.
You seem to missunderstand OpenSolaris. I would expect that the related
people who do things that I cannot predict rather approach me as the expert
for tar archive format related questions before deciding on something.
> > > The whole point there is reaching consensus, not fighting.
> >
> > I was in hope that this will be already possible. Unfortunately a fight was
> > started that will not bring us towards an OpenSolaris that is created in
> > cooperation of poeple from Sun and the external community.
>
> I don't know of anyone who was trying to fight with you, and I can't
> imagine how calling someone else's work "planless" is a useful way to
> seek cooperation.
>
> If there's a fight started somewhere, I don't think it's here.
If someone does not consult an expert, how would you call his acting?
> > We need cooperation and dicsussions between experts.
>
> I agree with that part.
Well, then let us wait whether/when the related people approach me as expert....
J?rg
--
EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
js at cs.tu-berlin.de (uni)
schilling at fokus.fraunhofer.de (work) Blog:
http://schily.blogspot.com/
URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily
