[sldev] [POLICY] Development by consensus (Re: Question regarding upcoming maintenance on 11/27-11/28)

[Rob Lanphier]

Several replies below:

On 11/22/07 4:07 PM, Argent Stonecutter wrote:
> On 22-Nov-2007, at 17:04, Rob Lanphier wrote:
>> On 11/22/07 2:12 PM, Matthew Dowd wrote:
>>> More fundamentally, LL needs to make up its mind whether it really
>>> does want to adopt a open community development model as it appears
>>> to aspire to or a closed development model as it tends to end up
>>> doing in practice!
>
>> Do you think we should put all of our architectural decisions up for
>> vote?
>
> I think that's an unfair characterization of Matthew's message.
>
That wasn't a characterization.  It was a question.

Argent Stonecutter also wrote:
> There's a lot of room between "going ahead without discussion" and
> "putting everything up to a vote".

There's also a lot of room between "open community development model"
and "closed development model".  We're about as open as any market
leading company out there.  Moreover, we've always said that opening up
Second Life would be a gradual process.  Matthew's statement quoted
above doesn't give us any credit for doing anything other than to
"aspire" toward openness when we are far, far past that.  You all still
wouldn't know a thing about this if we weren't more open company than most.

There's also a lot of room between "no discussion" and "going on and on
and on".  The initial conversation period may not have been as long as
people here wanted, but it was as long as our engineers felt they
needed.  The goal our engineers had was to put what we could out there
for review to make sure we weren't missing anything.  There were some
interesting considerations in the feedback, but nothing that convinced
them that the costs outweighed the benefits.  So they moved forward.

On 11/22/07 6:43 PM, Hamncheese wrote:
> I don't think everyone is asking for a vote. There was considerable
> discussion and feedback from the community and then developers went
> off and
> did their thing seemingly without consideration to the concerns raised.

The ideas were considered.  Just because there wasn't a protracted
debate on the mailing list, that doesn't mean they weren't considered.

Hamncheese also wrote:
> The
> problem is not even that. LL as the ultimate architect has a right and
> the
> responsibility to pull rank when necessary. But without an explanation it
> comes off as arrogant. As an enterprise architect, I cringe when I see
> this
> behavior in other architects because it backfires without fail.

Part of the reason for pulling rank here is because we can't divulge all
of the reasons we're doing this.  Therefore, the engineers here are
working with more information, and therefore had reasons to believe that
the community review did not factor in all of our requirements. 
Moreover, the stated requirements were not treated as requirements by
the reviewers.

Given the tight timetable that the team has been working on, they could
have spent that time trying to debate this prior to having a concrete
implementation, or they could have spent it implementing.  They spent it
implementing.

I'm not in a position to dive into great depth on the pros and cons.  If
I were personally the one responsible for the engineering effort, I
might have done things differently, but assuming that I even know what
I'm talking about (doubtful), I believe the difference between my
approach and the approach taken is at most a matter of taste.  I believe
they have a defensible design.

 
On 11/22/07 4:21 PM, Jesse Barnett wrote:
> From Oct 2:
>
> "Jesse Barnett wrote:
> > Basically I take that to mean that:
> >
> > Viewer security is the least of what they are working on. LL is
> going to do
> > what they said they were going to do anyways and we no longer have any
> > real input into the matter.
>
> Aw, the very next thing he said was:
>
> "This is still early in the conversation, so don't panic about us being
> slower than you might like to discuss this."
>
> -RYaN"
>
> So it was too early then and definitely too late now. This would be
> what people would be mad about.

Nothing is carved in granite.  This is software.  Yes, a fair amount of
engineering effort has been spent, but I think it's going to be much
easier to have an informed debate once we get some test software out there.


On 11/22/07 7:54 PM, Jesse Barnett wrote:
> I never speak for anyone but myself. But I did make the mistake once
> of being too
> adversarial in my position. Found that all it really does is push the Lindens away 
>
> instead of opening up the way to more discussion. 

I certainly appreciate that, and I'm sure my coworkers do, too.  We
insist on a certain level of civility inside of Linden Lab, and while we
can't enforce anything like that out on the wild, wooly Internet, we
will be more inclined to work with the more civil people.

Rob


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 249 bytes
Desc: OpenPGP digital signature
Url : http://lists.secondlife.com/pipermail/sldev/attachments/20071122/ad73eb57/signature.pgp