Hello everyone, Thank you for the discussion. As we’ve previously mentioned, the deprecation rule would be better addressed in a separate KIP for greater visibility, allowing other developers to participate in that discussion. I’d like to keep KIP-1124 focused solely on the upgrade path without introducing additional concepts.
Regarding Connect, I propose leaving it as it is. I’ve added a Connect section to the KIP, which is largely similar to the client section. The only notable difference is the bridge version range—since we’ve removed a deprecated API from 3.7, the appropriate bridge version from an API compatibility perspective should be [3.8.x–3.9.x]. Best, Kuan-Po Tseng On 2025/03/04 00:44:47 Chia-Ping Tsai wrote: > Dear all, > > Thank you for the discussion. Apologies for introducing an unrelated topic. > Here's a summary of that discussion. > > 1. A new KIP or thread will be created to define a formal deprecation policy. > This policy will apply to releases following 4.0, as 4.0 does not fully > conform to it. > > 2. We will not revert the 4.0 code. KIP-1124 focuses on client and streams > APIs, and the removed Connect REST API is outside its scope. > > If there are no objections to KIP-1124, please cast your vote in the voting > thread. Additionally, I propose adding a 'Client Upgrade Path' section to the > 4.0 documentation to highlight KIP-1124 if it passes. > > Thanks, > Chia-Ping > > > > Matthias J. Sax <mj...@apache.org> 於 2025年3月4日 凌晨1:57 寫道： > > What Chia-Ping says. > > > > To me, if we remove it in 4.0, we did not really keep it for 1 year if > > deprecated in 3.7, but it's subject to debate. At least for KS, we always > > kept stuff of the last 3 releases. > > > > I agree, that KIP-1124 should focus on clients/streams, and we want to keep > > the code as-is for 4.0 release, and remove these API in Connect, I have no > > objections at all. > > > > Thus, the question is not really about KIP-1124 directly, but more about > > 4.0 release in particular. > > > > Seems the verdict is, to keep the code as-is for 4.0 and remove these > > Connect API with 4.0.0. Works for me. > > > > > > -Matthias > > > > On 3/3/25 9:02 AM, Chia-Ping Tsai wrote: > >>> So that's 3 releases (3.7, 3.8 and 3.9) and over 1 year, no? > >> KIP-1124 highlights "we keep deprecated APIs for at least 3 prior > >> versions," but the Connect API change does not follow this rule. It is > >> valid if the deprecation happens in 3.6. > >> Best, > >> Chia-Ping > >> Mickael Maison <mickael.mai...@gmail.com> 於 2025年3月4日 週二 上午12:40寫道： > >>> Hi, > >>> For the Connect REST API change, the deprecation is in 3.7.0 which > >>> released in February 2024. So that's 3 releases (3.7, 3.8 and 3.9) and > >>> over 1 year, no? > >>> Mickael > >>> On Mon, Mar 3, 2025 at 5:31 PM Chia-Ping Tsai <chia7...@gmail.com> wrote: > >>>> hi all, > >>>>> I am also happy to follow Ismael's proposal and say "at least 3 > >>> releases > >>>> _and_ a minimum of 12 months". > >>>> +1 to this proposal > >>>>> Another example is > >>> https://github.com/apache/kafka/commit/a753172ad3e0927f412fb56e468c95a9a81ba3ad > >>>> We deprecated our log4j1 appender in 3.8.0 and it's been removed in > >>>> 4.0.0. Kafka 3.8.0 released in May 2024, so it's less than 1 year. > >>>> Yes, that's also an exception. Fortunately, this "breaking" change > >>> doesn't > >>>> affect the client, Streams, or Connect update path > >>>> I personally suggest creating a separate KIP to detail the new > >>> deprecation > >>>> rules (and create a new thread for this topic) . KIP-1124 only covers a > >>>> portion of deprecation issues, specifically API compatibility for > >>> clients, > >>>> Streams, and Connect. As Mickael mentioned, 4.0 cannot fully comply with > >>>> the new deprecation rules across the entire project. KIP-1124 should > >>> focus > >>>> on reaching consensus regarding the consistency we can achieve in 4.0. > >>>> Best, > >>>> Chia-Ping > >>>> Matthias J. Sax <mj...@apache.org> 於 2025年3月4日 週二 上午12:25寫道： > >>>>> Thanks Mickeal, > >>>>> I guess the question is, if we think we need to revert these removals, > >>>>> or if it's more reasonable to make an exception from the rule? > >>>>> I cannot really judge it, as I am not familiar with the details for > >>>>> Connect. Any suggestions from your side? > >>>>> -Matthias > >>>>> On 3/3/25 7:44 AM, Mickael Maison wrote: > >>>>>> Hi, > >>>>>> Another example is > >>> https://github.com/apache/kafka/commit/a753172ad3e0927f412fb56e468c95a9a81ba3ad > >>>>>> We deprecated our log4j1 appender in 3.8.0 and it's been removed in > >>>>>> 4.0.0. Kafka 3.8.0 released in May 2024, so it's less than 1 year. > >>>>>> Thanks, > >>>>>> Mickael > >>>>>> On Mon, Mar 3, 2025 at 4:40 PM Matthias J. Sax <mj...@apache.org> > >>> wrote: > >>>>>>> From > >>> https://cwiki.apache.org/confluence/display/KAFKA/Time+Based+Release+Plan > >>>>>>>> We break compatibility (i.e. remove deprecated public methods > >>> after a > >>>>> reasonable period, and typically wait 1 year after deprecation). > >>>>>>> To me, given that we do 3 releases per year, "1 year" as stated > >>> above > >>>>>>> and 3 releases, is just the same thing. > >>>>>>> I am also happy to follow Ismael's proposal and say "at least 3 > >>> releases > >>>>>>> _and_ a minimum of 12 months". > >>>>>>> -Matthias > >>>>>>> On 3/3/25 6:48 AM, Ismael Juma wrote: > >>>>>>>> Hi Chia-Ping and Bruno, > >>>>>>>> Right. Matthias stated that the 3 releases rule is the source of > >>> truth > >>>>> and > >>>>>>>> I don't recall that being the case. The source of truth is 12 > >>> months - > >>>>> I > >>>>>>>> was one of the people who was part of that discussion when the > >>> Scala > >>>>>>>> consumer was removed. I also disagree that the 3 releases rule is > >>>>> strictly > >>>>>>>> better since we can sometimes have shorter release cycles (like the > >>>>> intent > >>>>>>>> with the 3.9 release). I am ok with adjusting the rule to be "at > >>> least > >>>>> 3 > >>>>>>>> releases _and_ a minimum of 12 months" as part of this KIP, but we > >>>>> should > >>>>>>>> be clear that we're proposing a change as part of this KIP (vs > >>>>> following an > >>>>>>>> existing rule). > >>>>>>>> Ismael > >>>>>>>> On Mon, Mar 3, 2025 at 1:24 AM Bruno Cadonna <cado...@apache.org> > >>>>> wrote: > >>>>>>>>> Hi, > >>>>>>>>> I suspect that the three-release-rule was a derivation from the > >>>>>>>>> 1-year-rule since we usually have three releases in one year. > >>>>>>>>> IMO, a three-release rule is easier to reason about, because you > >>> don't > >>>>>>>>> need to know when the release took place. > >>>>>>>>> However, I recognize that the 1-year-rule seems to be the official > >>>>> rule. > >>>>>>>>> Best, > >>>>>>>>> Bruno > >>>>>>>>> On 03.03.25 09:58, Chia-Ping Tsai wrote: > >>>>>>>>>> hi Ismael > >>>>>>>>>> The thread[0] contains a brief discussion about the one-year > >>> rule. > >>>>> I've > >>>>>>>>>> also updated the KIP page[1] to highlight this rule. However, > >>>>> declaring > >>>>>>>>>> [3.7-3.9] as API compatible with 4.0 can be unrelated to the > >>> one-year > >>>>>>>>> rule. > >>>>>>>>>> We can do this for consistency, ensuring clients, Streams, and > >>>>> Connect > >>>>>>>>> have > >>>>>>>>>> the same version range. Additionally, we can address this by > >>>>> reverting a > >>>>>>>>>> minor commit. If we don't agree on consistency, we can update the > >>>>> KIP to > >>>>>>>>>> include different API compatibility versions for Connect. > >>>>>>>>>> [0] > >>> https://lists.apache.org/thread/j7n4qqsvxz84f5cg89kdm9foby36j28n > >>>>>>>>>> [1] > >>> https://cwiki.apache.org/confluence/pages/diffpagesbyversion.action?pageId=65867320&selectedPageVersions=9&selectedPageVersions=8 > >>>>>>>>>> Best, > >>>>>>>>>> Chia-Ping >
