Thanks Ryan! With that, is it necessary to start a new vote or can we consider this one as successful?
I'm counting so far: - 5 binding +1s: Eduard, Matt, Prashant, Daniel, Ryan - 5 non-binding +1s: Dmitri, Christian, Ajantha, JB and myself. - No -1s Thanks, Alex On Wed, Mar 25, 2026 at 8:23 PM Ryan Blue <[email protected]> wrote: > > Yes, this look better now so I'll update my vote to +1. > > Thanks, Alex! > > On Wed, Mar 25, 2026 at 1:37 AM Alexandre Dutra <[email protected]> wrote: >> >> Hi Ryan, >> >> That's a fair point, I've updated the spec to remove the mention of >> the Java library and the associated removal timeline. >> >> Does this address your concerns enough for you to reconsider your vote? >> >> Thanks, >> Alex >> >> On Mon, Mar 23, 2026 at 11:42 PM Daniel Weeks <[email protected]> wrote: >> > >> > For migration purposes, this new endpoint should be included with >> > supported endpoints config response, so newer clients should know if the >> > catalog supports signing through this new path or should default to the >> > old behavior. >> > >> > I think deprecating the old spec is fine (though agree that maybe the >> > removal timeline should be reconsidered or simply removed). >> > >> > -Dan >> > >> > On Mon, Mar 23, 2026, 1:32 PM Ryan Blue <[email protected]> wrote: >> >> >> >> -0 >> >> >> >> I think the addition to the REST spec is fine, but I don't think the >> >> changes to the old signer spec are correct. First, the old spec now >> >> references the Java library versions and states that support will be >> >> removed in 1.12.0. I think it should be independent from Java versions >> >> since the REST spec is not tied to Java releases -- it's a bit unclear >> >> how we want to handle this with secondary specs, but I doubt that the >> >> solution is to rely on Java library versions. Second, is there a summary >> >> of the discussion where we decided to deprecate this so quickly? I >> >> thought that there were projects that implement remote signing, so how >> >> can we expect people to move in a Java minor release timeframe? What is >> >> the plan for falling back to the old API and for how long? >> >> >> >> On Mon, Mar 23, 2026 at 12:37 PM Daniel Weeks <[email protected]> wrote: >> >>> >> >>> With the updates, I'm changing my vote to +1 >> >>> >> >>> I believe the vote was already called, so for procedure purposes, we >> >>> should probably just start a new vote. >> >>> >> >>> -Dan >> >>> >> >>> On Thu, Mar 19, 2026 at 9:39 AM Eduard Tudenhöfner >> >>> <[email protected]> wrote: >> >>>> >> >>>> +1 >> >>>> >> >>>> On Wed, Mar 18, 2026 at 6:07 PM Alexandre Dutra <[email protected]> >> >>>> wrote: >> >>>>> >> >>>>> Hi all, >> >>>>> >> >>>>> Gentle reminder to review the revised spec changes: >> >>>>> https://github.com/apache/iceberg/pull/15450 >> >>>>> >> >>>>> Thanks, >> >>>>> Alex >> >>>>> >> >>>>> On Fri, Mar 6, 2026 at 5:21 PM Alexandre Dutra <[email protected]> >> >>>>> wrote: >> >>>>> > >> >>>>> > Hi all, >> >>>>> > >> >>>>> > FYI the required changes were implemented: >> >>>>> > https://github.com/apache/iceberg/pull/15450 >> >>>>> > >> >>>>> > Thanks, >> >>>>> > Alex >> >>>>> > >> >>>>> > On Thu, Mar 5, 2026 at 9:49 AM Alexandre Dutra <[email protected]> >> >>>>> > wrote: >> >>>>> > > >> >>>>> > > Hi all, >> >>>>> > > >> >>>>> > > With one binding -1, the vote does not pass. I will prepare the >> >>>>> > > requested changes and start another vote thread when we're ready. >> >>>>> > > >> >>>>> > > Thanks, >> >>>>> > > Alex >> >>>>> > > >> >>>>> > > On Wed, Mar 4, 2026 at 11:12 PM Daniel Weeks <[email protected]> >> >>>>> > > wrote: >> >>>>> > > > >> >>>>> > > > -1 (but I think we can address the concern easily) >> >>>>> > > > >> >>>>> > > > I just added a comment to the PR that's a blocker for me. We >> >>>>> > > > introduced an explicit enumeration of cloud providers which I >> >>>>> > > > strongly oppose codifying in the spec. >> >>>>> > > > >> >>>>> > > > That limits other providers from leveraging the signing portion >> >>>>> > > > of the spec without a spec change and is unnecessarily strict. >> >>>>> > > > >> >>>>> > > > This should be a simple update to address, but I can't support >> >>>>> > > > this change until we remove that. >> >>>>> > > > >> >>>>> > > > -Dan >> >>>>> > > > >> >>>>> > > > >> >>>>> > > > >> >>>>> > > > On Wed, Mar 4, 2026 at 8:44 AM Jean-Baptiste Onofré >> >>>>> > > > <[email protected]> wrote: >> >>>>> > > >> >> >>>>> > > >> +1 (non binding) >> >>>>> > > >> >> >>>>> > > >> Regards >> >>>>> > > >> JB >> >>>>> > > >> >> >>>>> > > >> On Mon, Mar 2, 2026 at 7:33 AM Alexandre Dutra >> >>>>> > > >> <[email protected]> wrote: >> >>>>> > > >>> >> >>>>> > > >>> Hi all, >> >>>>> > > >>> >> >>>>> > > >>> This is a second vote attempt in order to adopt the promotion >> >>>>> > > >>> of the >> >>>>> > > >>> remote signing endpoint to the main REST spec. >> >>>>> > > >>> >> >>>>> > > >>> Related links: >> >>>>> > > >>> >> >>>>> > > >>> ML thread: >> >>>>> > > >>> https://lists.apache.org/thread/2kqdqb46j7jww36wwg4txv6pl2hqq9w7 >> >>>>> > > >>> PR: https://github.com/apache/iceberg/pull/15450 >> >>>>> > > >>> >> >>>>> > > >>> Please vote within the next 72 hours. >> >>>>> > > >>> >> >>>>> > > >>> [ ] +1 Adopt the promotion of the remote signing endpoint to >> >>>>> > > >>> the main REST spec >> >>>>> > > >>> [ ] +0 >> >>>>> > > >>> [ ] -1 Do not adopt, please explain why >> >>>>> > > >>> >> >>>>> > > >>> Thanks, >> >>>>> > > >>> Alex
