Re: Status of First-Party Sets

Hey Robin, et al:



(For context, for those who don't know, I’m also a WICG co-chair.)


The WICG is intended for incubation of standards features - as are all
community groups. It is important to remember that Community Groups are not
standards-track venues in the W3C, and one CG does not lend more legitimacy
to its products than another; they simply have potentially different
communities.  Nothing that lives in a W3C Community Group should be called
a “standard”.



As per the minutes of the Privacy CG meeting on 5/26/22 [1], Tess stated
that the PCG wanted to focus on “privacy stuff”, and FPS was not, in her
opinion, a good fit for the focus of their group;  this was backed up by
Pete, who said “Seems like PrivacyCG is not the right place to continue the
discussion.”  This was not by unanimous community consensus; but, as Tess
said, removing work items is at the discretion of the PCG chairs according
to its charter.    I’ve noted before my personal disagreement with giving
the very small number of browser engine companies essentially a veto vote
over incubations in the PCG; but regardless, that’s not the case in the
WICG.



As I said during the PCG meeting, the WICG will allow for incubation as
long as there are multiple independent entities (not specifically different
root implementations) that are interested.  FPS met that bar initially [2],
and continues to pass that bar - several PrivacyCG community members who
are supportive of the proposal chimed in on [3], including at least one
other browser vendor (Microsoft Edge) who is interested in continuing
development..  This should not be unexpected; FPS was adopted into the
Privacy CG from the WICG with the understanding that PCG wanted to spend
their community time on it; the fact that the Privacy CG have since decided
to stop spending time on this exploration does not change the fact that
there are still a significant set of independent voices that think it
continues to merit further incubation.



On a separate note: Google Chrome and the broader Chromium project has not,
to my knowledge, used presence in the WICG as justification to ship any
feature.  It is true that the Blink process strongly encourages any
to-be-shipped feature to be in an open forum, to enable collaboration and
at the very least, eventual interoperability. However, having a
specification in an open forum is a baseline requirement, not a
justification, for shipping.



-Chris



[1]
https://github.com/privacycg/meetings/blob/main/2022/telcons/05-26-minutes.md

[2] http://discourse.wicg.io/t/proposal-first-party-sets/3331
[3] https://github.com/WICG/first-party-sets/issues/88

On Tue, Jun 7, 2022 at 11:53 AM Robin Berjon <robin@berjon.com> wrote:

> Hi Travis!
>
> On 2022-06-06 14:09, Travis Leithead wrote:
> >  > [..] I’m unsure how one would go about removing FPS from WICG.
> > Perhaps the WICG chairs can advise?
> >
> > The WICG is home to over 120 <https://wicg.io/> unique incubations at
> > varying stages of maturity and implementation. While I have been a
> > co-chair, we have graduated numerous proposals into other venues, and
> > archived others at the request of their owners, but we've never forcibly
> > removed any incubations (even when they appear inactive for years). I
> > think it would set a bad precedent to start now. The WICG is a field for
> > sowing ideas; for this reason our criteria for acceptance is very low.
>
> I think a potentially important point is at risk of finding itself
> buried under James's usual anti-privacy activism.
>
> The status of FPS in WICG is unusual and (to me) unexpected. The WICG is
> intended for incubation of new features and early standard proposals.
> FPS has already been incubated quite a lot, and the incubation didn't
> pan out. I'm not suggesting that FPS be shut down — as we all know,
> sometimes standards take trying more than once — but I would encourage
> WICG chairs to be particularly careful that it does not impinge upon the
> WICG's reputation. There is already significant grumbling in the
> community that the WICG is primarily a venue for the standards-washing
> of Google's plans; it would be very unfortunate if the WICG found itself
> used as justification to ship FPS in a browser.
>
> It also appears that Kaustubha and the Privacy CG chairs have a
> different appreciation of the status of implementer support. Given the
> importance that having multiple implementations holds in our processes
> and community, this is an issue that seems worth clarifying. Kaustubha:
> do you mind explaining your conclusion on this point? Again, I don't
> think that having just one implementer interested means FPS shouldn't go
> to the WICG (some things there have zero implementers interested and
> that's fine!) but we should at least be able to reach consensus on who
> is interested and how that impacts the legitimacy of shipping the feature!
>
> --
> Robin Berjon
> VP Data Governance
> Acting VP Marketing Analytics
> The New York Times Company
>