[osol-discuss] Nameclash on svn_77 because Sun is ignoring PSARC discussions

Alan Coopersmith alan.coopersmith at sun.com
Fri Dec 14 11:03:16 PST 2007


[Dropping opensolaris-discuss & ogb-discuss because this is really an ARC
 issue, and I don't need 4 copies of every e-mail in this thread.]

Joerg Schilling wrote:
> Joep Vesseur <Joep.Vesseur at sun.com> wrote:
>> On 12/14/07 12:58, Joerg Schilling wrote:
>>
>>> The real compare is 20 years older and I did _warn_ _before_ the name appeared
>>> in /usr/bin. For this reason, this is an important bug in Solaris Express.
>> I completely fail to see how your side comment on a PSARC case discussing
>> netcat (PSARC 2007/389) could have possibly given you the idea that "compare"
>> would be a name reserved for your program.
> 
> The fact that you found this text verifies that you got the message...
> Not people only need to behave the same way.

I'm afraid you still don't understand how the ARC is actually operated.

The ARC consists of engineers who spend a few hours a week reviewing the
cases brought before them (usually 1-3 hours in meetings and a bit more
in e-mail and in reading the submitted materials).    While all the ARC
members I know are very smart people, none of them are capable of keeping
the entire 15+ years of ARC cases, e-mail threads, and meetings in their
heads for instant recall, and especially not when people like you add
dozens of off-topic messages to case discussions like the netcat ones.

There is no giant database of cross-referenced issues, and no one has
time to go back and search all the threads for any possible mention of
a related issue in a completely unrelated case.   Expecting ARC members
reviewing an ImageMagick case to search the entire ARC history for the
word "compare" is ludicrous - I just did that on our internal database
and got 5784 hits, which no one has time to sort through.   Expecting
them to remember to go back and find an off-topic message made in the
unrelated netcat case is also unreasonable.

The ARCs rely on the members to all know the "big rules" (mostly those
listed on the ARC Best Practices & Policies page), and for those
engineers who are the experts in their areas to chime in if there is an
issue they see there.   If a case came forward that proposed to integrate
an XML editor named "xedit", I'd expect that I need to step up and say
"But we already ship an obscure X11 GUI editor with that name" and that
if I didn't, it might slip through - I won't blame the other ARC members
for not being experts in X11, nor for not knowing the name of every other
program in the universe.

In the end, the ARC members will vote to do what they think is best, and
they may not agree with you - that's why it's a voting system, not a
"any one person can veto" system.

-- 
	-Alan Coopersmith-           alan.coopersmith at sun.com
	 Sun Microsystems, Inc. - X Window System Engineering




More information about the opensolaris-arc mailing list