Minimap - stranded?

[rubjo]

Background:

A lot (hundreds) of users have indicated this as a desired feature (characteristics and considerations discussed in #5308). Based on reactions, it's up there among the top 10-15 issues. The issue was closed by @SomeoneToIgnore with a request for PRs.

A preliminary PR (#20852) by @wazawoo was rejected with a request for other contributions / more discussion.

So, where does this stand now? Indefinitely cancelled? I'm one of the people who would like to see this (along with a little more powerful built-in Git add / commit / push integration) before I feel I can commit (heh) to Zed and maybe switch permanently from Sublime.

Anyone with the skills and time to pick up the baton and work towards making Minimap happen?

[SomeoneToIgnore]

To fix your vocabulary a bit:

The issue was closed by @SomeoneToIgnore with a request for PRs.

It was locked, not closed — it cannot be commented, but open in the tracker still.
While I was asking for actions in that issue, the main pledge was to stop spamming — which was ignored by a notable chunk of the commenters there.

A preliminary PR (#20852) by @wazawoo was rejected with a request for other contributions / more discussion.

The PR got closed by that person without a real discussion, it was not rejected as there's been nothing functional in it and it was very far from being completed — hence, nothing to reject to start with.

I'm still happy to review whatever PR comes, as I do now with inline diagnostics PR, for example.

Ironically, inline diagnostics are getting 2nd good, large PR from the community (without a single spammy message in the tracker), yet this feature is not even in the constructive discussion phase.
Anything I can do from my side, to improve this?

---

Now, to the answers:

So, where does this stand now?

Same place as before, there's an open issue and useless noise around it.

The most constructive action from the community so far was the PR, and even that did not get far from the start, unfortunately.

Indefinitely cancelled?

This is a very weird, exaggerated claim, as nothing is cancelled, the issue is there still, open.

Yet, there are much more demanded features like git integration that the team works on now.

[rubjo]

Apologies for the imprecisions: yes, locked issue (with admittedly a lot of spam, but at the same time a lot of constructive discussion), not closed. And closed PR, not rejected.

"Indefinitely cancelled?" is more of a "weird, exaggerated" question than a claim, so I am pleased to hear that the feature isn't at least ruled out for the future. It has more comments and reactions than the git integration issue, by the way.

[TYTheBeast]

Anything I can do from my side, to improve this?

Hiya, popping my opinion as a new user trying to switch from Cursor, a fork of VSCode. I believe the best you can do as a maintainer is outline the underlying work to get a start on the PRs, and see if anyone has their interest piqued by the possibility of having a minimap, or in other words, just make a little roadmap for others to follow, based on what work you believe needs to be done. Had I known Rust better, I would've probably given the PR a go myself, but alas I do not have the skills, yet.
I'll subscribe to both the Minimap issue and this discussion to see how things go, and I really hope to see Zed take off, it seems like a new revolution that could finally dethrone VSCode!

Quick edit: On a second note, not to be rude, SomeoneToIgnore, but I believe another thing you could do is perhaps try to change your approach with contributors asking for guidance. I'm mostly referring to this:

I trust you didn't mean anything rude and was just being playful, as even I which don't contribute much do know GitHub has a Discussions section, but your approach completely turned off a potential contributor, and while others will still continue their work or develop on Zed, with others editors still having so many more users, it's important for Zed to develop and mature, and it needs everyone's help for that.
TL;DR Try to keep the sarcasm to a reasonable amount with newcomers, not everyone's mind is unshackled enough to get a good chuckle out of your playfulness :)

[SomeoneToIgnore]

For the roadmap, the last message in the issue has some resemblance to it: #5308 (comment)

I'm afraid I cannot make it better before anyone goes and actually tries things to understand what's missing or wrong in that text.

---

As for being rude, I sure do agree that I'm not the best person sometimes, I'm not happy with it and hopefully getting better these days.

Ironically, that particular bit screenshotted is the time where I've tried to be nicer than my former average, and even my first reply in this discussion looks more stern to me than the bit.
And of course I was not trying to be bad in that PR, as that PR tried to be much more compared to the rest of the noise on the topic so far.

Another odd take: I'm ok with that PR got closed so fast, as a "potential contributor" with no Rust knowledge trying to adjust the relatively complex rendering code is usually bad news — being so sensitive on top is for sure not good for the process.
This usually ends up with a ton of comments, PRs dragging for months and, in the end, the maintainer doing all the work through the comments, implementing the feature, just very painfully and slowly, taking a lot of everyone's time.
The worst part — such people rarely come back with more improvements after their "education".

People come and go, thinking of their "potential contributions" and other things, while totally overlooking that there are maintainers having to deal with all this, over and over and over, and those maintainers have to be good and happy to help reviewing and whatnot, just because.
My time is relatively sparse and also allocated to certain commitments of my own (which could lead me to getting fired if I pivot into contributors' caretaking overly; or will take off my free time; or, worse, undermine my mental health slowly).

So, if you have any respect for this side of the fence, prefer being productive or doing nothing over being ignorant and always doing something, "potentially".
I'm not saying that you always should and not implying on anything — but somewhat very tired to emphasize for Nth time in the minimap topic, that it's not getting anywhere with all this.

Ok, my 1 hour of ventilation (it takes a while to compose these messages, and what for??) is over and I'm going to do something useful, sorry for the long writeup.

[0x2CA]

The reason is that the underlying layer needs to be refactored to support it, and the refactoring is a lot of work that the zed team doesn't have time for.