Proposal for new trademark approval policy

[Christoph Wickert]

Am Montag, den 18.04.2011, 10:24 -0400 schrieb Jared K. Smith: 
> On Sun, Apr 17, 2011 at 8:17 AM, Christoph Wickert
> <christoph.wickert at googlemail.com> wrote:
> > Only once or per-spin?
> 
> The proposal is to make this decision once.

Fine with me. 

> >> 1) Each appointed SIG should create a checklist or SOP for Item approval
> 
> > What does "Item" mean in this context? Is it the spin or the trademark?
> 
> An "item" in this proposal is a spin or new image format that doesn't
> fall into the definition of a spin (such as an EC2 image, for
> example).

Ah, I didn't think of this and I don't think it is obvious from the
context. Maybe we could word it more explicitly. 

While I do think that a new item needs special treatment, I'm afraid it
will slow us down dramatically, say at least by one release cycle. We
need to have a strict timeline to avoid that.

> >> 2) When an appointed SIG gives its approval for a Item, it should do
> >> so in a public and transparent manner. The SIG can decide the exact
> >> manner of approval but who and where the approval is granted should be
> >> specified in the SOP.
> >
> > Would the current spins process meet this description?
> 
> The current spins process meets this description.  The idea is to have
> the other appointed SIGs (let's say Rel-Eng, Design, and QA for
> argument's sake) also build standard operating procedures for how they
> give approval to a spin or image format, and ensure that those
> approvals happen in a transparent manner.

Great!

> >> 3) Once approvals from appointed SIGS are done, the Item will be
> >> submitted to the Board (via a ticket) for trademark approval.
> >
> > As far as I understood the current process, this is step one. Before we
> > request trademark approval from the board, the spin was already ratified
> > by the spins SIG, the design team already approved the artwork and so
> > on.
> 
> That's how it was previously done, but this is one of the major
> changes that we're proposing.  In the new system, you'd submit a Board
> ticket that says "I got approval from the Spins SIG, the Design team,
> Rel-eng, and QA.  I'm ready to get trademark approval."

I think this makes sense, but it should be outlined better in the
proposal. 

> > And isn't using the ticket system a bit overkill? Say we have 20 items
> > to check by 10 different SIGs, this means we have 20 tickets, right?
> > With the current privacy level of the board's trac, we cannot have a
> > ticket where all SIGs comment on.
> 
> I anticipate one ticket per SIG.  

But this doesn't work. If I open a ticket on behalf the spins SIG, I'm
the only member of the SIG that can comment.

> If an individual SIG wants to break
> particular items into a separate ticket for very specific reasons they
> can, but I do not think we need one ticket per checklist.  One ticket
> per SIG should be sufficient if most (if not all) cases.  

If the ticket is just to state "SIG X approves spin Y" even a single
ticket should be sufficient, but this is not possible in trac either.

> And to be
> honest, it doesn't even have to be a ticket.  If a SIG wants to come
> up with some other way of tracking approvals, that's fine with me, as
> long as it's done in a public, transparent way.  I just think tickets
> are easy to keep track of and easy to follow.

Agreed, I do like tickets, too. If this cannot be done in the board's
trac we could do it in bugzilla by adding a new component "Spin review"
similar to "Package review". However I think this is abusing bugzilla,
so I think we should better go with the spins SIG's trac instance.

> >> 4) The Board can appoint a specific SIG to keep track of "trademarked"
> >> Items and that SIG will set schedules and/or reasons for re-approval
> >
> > Spins are already approved per release only and need re-approval.
> 
> There was some confusion on that in our meeting, so I'm glad you've
> clarified that.  In most cases, I would think that the re-approval
> process would be very quick and easy.

Well, it turns out that on a technical level it's not, but this is not
the board's problem then. 

> > As I said this proposal is a step into the right direction, however I'm
> > afraid it duplicates processes and governance that is already handled on
> > a SIG level. I'm a big fan of the KISS principle.
> >
> > SIGs cannot approve the trademark usage, they can only approve what is
> > in their scope and then give green light. Once there is positive
> > feedback from all parties it is on the board to grant the trademark
> > usage. This is what has worked fine in the past until it suddenly and or
> > no apparent reason became very complicated on the board level.
> 
> Maybe you've misunderstood the proposal, but your description above
> almost *exactly* fits with what we're proposing now.  The new proposal
> is to have *less* involvement from the Board and *more* involvement
> from the SIGs on their particular areas of expertise.
> 
> I'd be happy to help explain the new proposal if you have more
> questions or concerns.

No, I have absolutely no problem with the proposal, because as you say
correctly it's basically how it has worked well in the past put down in
a policy. I like it but I am a little concerned that we need to have our
board work on a policy that basically says "Dear board members, don't
stick your nose into stuff that's none of your business but trust our
contributors to do their job and do yours instead". Sorry for putting it
so bluntly. :D 

Regards,
Christoph