Hi Sergey, Thanks for your thoughts on this. I guess we could go some sort of retrospective to understand how consistently the 2/3 majority was applied. It is probably better to clarify the words – as you suggest. My take is we should look to have the minimal admin and process to get things through in a timely manner; where this is not appropriate is when some proposal fundamentally changes Flink – these changes need proper consideration, and the 2/3 majority would apply. I suggest the 2/3 majority should only be used when necessary and should not be overused.
Maybe a clarification could be along the lines of - If any member of the PMC thinks that a Flip is sufficiently disruptive, they should propose that the 2/3s majority be used. Would this help? Kind regards, David. From: Sergey Nuyanzin <snuyan...@gmail.com> Date: Monday, 7 July 2025 at 09:21 To: dev@flink.apache.org <dev@flink.apache.org> Subject: [EXTERNAL] Re: [RESULT][VOTE] Flip-532 Donate GetInData HTTP Connector to Flink Hi Robert and David Thank you for your clarifications. I'm not going to block or veto it, I have just raised a concern that from one side there is an example with CDC connectors and from another HTTP connector and they are handled differently. If you think it is ok, I'm fine with it. Maybe then we should add more clarifications about that in Flink bylaws or some other place? On Fri, Jul 4, 2025 at 6:34 PM David Radley <david_rad...@uk.ibm.com> wrote: > > Hi Sergey and Robert, > In the Community Health Initiative workgroup we had discussed this and Robert > mentioned his line of thinking that this would need 3 votes rather than the > 2/3s of PMC. > I agree with Robert when he says : > > I don't think a new connector changes the shape and direction of the > project ? > It's "just" a major change, requiring a FLIP. > > Saying that, we have only received support this Flip. I would love to hear > any specific concerns about this connector. It is not perfect and could be > more complete and will be further enhanced, but is it a big value add for > many use cases as-is. > Kind regards, David. > > > > From: Robert Metzger <rmetz...@apache.org> > Date: Thursday, 3 July 2025 at 07:27 > To: dev@flink.apache.org <dev@flink.apache.org> > Subject: [EXTERNAL] Re: [RESULT][VOTE] Flip-532 Donate GetInData HTTP > Connector to Flink > Hey Sergey, > > I thought about this question as well, but in my opinion, a regular 3 > binding PMC vote is sufficient here: > > The description of a new codebase adoption is: > > > Adoption of large existing external codebase. This refers to contributions > > big enough that potentially change the shape and direction of the project > > with massive restructuring and future maintenance commitment. > > > I don't think a new connector changes the shape and direction of the > project ? > It's "just" a major change, requiring a FLIP. > > For me, the biggest concern with any new connector will be the lack of an > active community around it. People willing to fix, review, release stuff. > However, I did not bring this up, since two companies (IBM, GetInData) seem > to be actively using it, and the GH community is fairly active. > I understand the desire of companies to make contributions to repositories > that are maintained by an independent organization (Apache), instead of a > company (GetInData), that's why I support this FLIP. > If you feel otherwise, you as a PMC member have the power to stop this > effort, if you think it requires further discussion in the PMC. > > Best, > Robert > > > On Tue, Jul 1, 2025 at 8:17 AM Sergey Nuyanzin <snuyan...@gmail.com> wrote: > > > Thank you for your volunteering here David > > > > I don't want to be a devil's advocate however I'm not sure we can > > consider it as a regular FLIP and accepted. > > The reason I think this way is the current Flink Bylaws[1] and > > especially telling that in case of "Adoption of New Codebase" > > we need to achieve 2/3 majority of votes in order to accept it. The > > similar thing happened while accepting CDC Flink connectors [3]. > > Also binding votes in this cases are considered not the same way as > > for regular FLIPs > > > > Please correct me if I'm wrong > > > > [1] https://cwiki.apache.org/confluence/display/FLINK/Flink+Bylaws > > [2] > > https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=120731026#FlinkBylaws-Actions > > [3] https://lists.apache.org/thread/sq5w21tcomrmb025tl820cxty9l0z26w > > > > On Mon, Jun 30, 2025 at 3:48 PM David Radley <david_rad...@uk.ibm.com> > > wrote: > > > > > > Hi devs, > > > > > > The vote[1] on FLIP-532[2] concluded with the following results: > > > > > > Approving(+1) votes: > > > > > > Ferenc Csaky (binding) > > > Gyula Fora (binding) > > > Leonard Xu (binding) > > > > > > > > > > > > PoorVank Bhatia (non-binding) > > > > > > Sharath (non-binding) > > > Nic Townsend (non-binding) > > > > > > Lajith Koova (non-binding) > > > > > > Tom Cooper (non-binding) > > > > > > Mark Nuttall (non-binding) > > > > > > Anu K T (non-binding) > > > > > > Mehdi Deboub (non-binding) > > > > > > Ammu P (non-binding) > > > > > > Sebastien Pereira (non-binding) > > > > > > > > > > > > > > > There were 3 binding, 10 non-binding and no disproving(-1) votes. > > > > > > I'm happy to announce that FLIP-532 has been approved. > > > > > > Thanks, > > > David > > > > > > [1] https://lists.apache.org/thread/ft7cw2cy7g5cj3hyx6l9r1twnoy0wr0q > > [2] > > https://cwiki.apache.org/confluence/display/FLINK/FLIP-532%3A+Donate+GetInData+HTTP+Connector+to+Flink > > > > > > > > > Unless otherwise stated above: > > > > > > IBM United Kingdom Limited > > > Registered in England and Wales with number 741598 > > > Registered office: Building C, IBM Hursley Office, Hursley Park Road, > > Winchester, Hampshire SO21 2JN > > > > > > > > -- > > Best regards, > > Sergey > > > > Unless otherwise stated above: > > IBM United Kingdom Limited > Registered in England and Wales with number 741598 > Registered office: Building C, IBM Hursley Office, Hursley Park Road, > Winchester, Hampshire SO21 2JN -- Best regards, Sergey Unless otherwise stated above: IBM United Kingdom Limited Registered in England and Wales with number 741598 Registered office: Building C, IBM Hursley Office, Hursley Park Road, Winchester, Hampshire SO21 2JN
