[ogb-discuss] [opensolaris-summit] My comments (very subjective) on proposed Summit topics

[Alan DuBoff]

On Wed, 26 Sep 2007, Garrett D'Amore wrote:

>> That would be an error.  IF the process were so heavy weight, then 
>> something is wrong.
>> 
>> I've slipped many changes into a single putback ... *but* I try to make 
>> sure I have only one CR. 
>
> This didn't come out right.  I meant to say, "one CR per bug".    So one 
> putback, many bugs, and hence many CRs.  Sorry for the confusion.

Garrett,

I guess my point is that I have heard engineers say they have done that 
because the process is so difficult. I won't name who, but I think that to 
some extent this does go on some, today. I was more using this as an 
example of how the engineers have learned to work around some of the 
limitation.

I don't know that this is a bad thing, just that it can and does happen. 
It makes the engineer more aware so in the case where something does 
break, they know they're on the line for it.

I wasn't pointing any finger at you or anyone else specific, just pointing 
out that there are many cases where files are going back on a daily basis. 
You really need to look at the changes to fully understand what is going 
back.

I agree with what you said about workarounds, examples, contact info, 
etc...Sun is moving towards getting the bug database out to the community 
though, time being the only real process holding it up.

I was also more using that as a case where the community claims foul, that 
Sun has not opened up the bug database. I was only pointing out that it 
wouldn't magically help people understand every CR that gets putback, they 
would still need to look at and evaluate the files and fixes to fully 
understand each case.

Again, I don't know how you could easily transition such a large code base 
to a community and not cause some issues as the bug database, unless it 
just all magically happened together at the same time (source code 
management, bug database, sources, ARCs, etc...). In the case of Solaris, 
as it was, there was just too much process for that to happen cleanly. As 
it is they did a good job, IMO, and our community is moving towards a 
pure, open model.

--

Alan DuBoff - Solaris x86 IHV/OEM Group