Re: [PATCH v13 0/2] checkout: --track=fetch
- From
Junio C Hamano <gitster@pobox.com>
- Date
- Jun 18, 2026, 02:50 UTC
- Message-ID
- <xmqqtsr0tvci.fsf@gitster.g>
- In-Reply-To
- <CAHwyqnXLceLXzRrW_7TB8JM+Ur92gw5QkYeKjzOGbWX+f_yLjw@mail.gmail.com>
Harald Nordgren <haraldnordgren@gmail.com> writes:
Show 13 quoted lines
> But can I offer some (unsolicited) feedback on this review process in > particular? Given that it seems unlikely to hit 'master' at this > point, I want to say that it's the wrong order of things to dig into > code specific feedback, before deciding if we even want the feature at > all. We are wasting each other's time. I have pushed on despite > initial negative feedback, that's on me. But I also cannot lay flat, I > like the idea so I keep pushing. Now we have v13, maybe soon v14, of a > topic that has slim chances of passing. > > I would have been much happier if you shut this topic down directly. > > Imagine all the review time spent on this that could have been better > spent elsewhere.
You are forgetting that there are two things a topic author can and should do: addressing higher-level design issues (e.g., Counter "is it a good idea to begin with?" with "it is, because in this context you haven't thought about, it would make the user experience better", to help reviewers understand and hopefully agree with your viewpoint) and addressing mechanical implementation issues (e.g., Respond to "it is wasteful and adds maintenance burden to dupplicate the logic here and there" with "Here is an updated implementation based on better refactoring"). The latter is easier in a sense, especially when the reviewers say something like "I am not sure if this is a good idea to begin with, but assuming it is, here are the things I find in your implementation". But the former needs to be addressed eventually before the topic can gain wider support (remember, it does not have to be sold to _me_ 100% and make me say "oh, I cannot believe Git did not have this feature for 20 years!"; you only need to convince me that it is OK to have it as an opt-in feature, that is a lot lower bar to cross. In addition, I do not have to agree with it at all, as long as others whose judgement I trust well enough support the feature).