I agree. Reasonable approach.

That's why the rules for unpublished translations seem so silly.

Before a translation is published, the rules are very strict - I can't edit the translation at all.

After a translation is published, the rules are very loose - I can edit whatever I want.

The loose rule works well - kata get fixed / improved if necessary.

The strict rules don't work well - it takes far too long for kata / translations to get approved. Too many never get approved.

The situation is frustrating for kata authors, and it makes the site less interesting for everyone, because there are so many translations we can't use yet, just because of these silly rules.

I can't even approve a translation if there is a minor difference in the description. And if I fork it to fix that simple bug, I can't approve it anymore. Because "you can't approve your own translation". Why not? I can do whatever I want with an existing, published kata.

This is to make easier the maintenance of present languages. Once a language is approved it cannot be retired.

Easy to say.

Is that supposed to be an argument? You said "this is a feature of the website". I said "it's a misfeature". Which one was easier to say?

I don't understand what you are are refering to.

I can edit the tests, the solution and the description of a published kata. I can rewrite them if I want. And no one needs to approve my edits.

It's a misfeature...

Easy to say.

Why are the rules for unpublished kata much stricter than for published kata

I don't understand what you are are refering to.

once you publish a fork, you can also rewrite entirely the tests

I can edit the tests, the solution and the description of a published kata. I can rewrite them if I want. And no one needs to approve my edits. That's fine. Why are the rules for unpublished kata much stricter than for published kata? Doesn't make sense.

The problem is that these arcane rules mean that many translations never get approved. It's a misfeature that I can't fix trivial changes in the description when I want to approve a translation.

Merge conflicts are annoying (I have been annoyed an incalculable number of times with that), but they need to be fixed to have a coherent description. You cannot fork yourself a translation and then approve it, for the same reason as you cannot approve your own translation (once you publish a fork, you can also rewrite entirely the tests). If you find yourself with a similar case do what you have just done here (I guess): publish a new fork fixing the conflict and post a suggestion for approval. Once there is a merge conflict on a fork, it is no longer approvable. But this is a feature of the website, the way description versions are handled, and not a problem of a rule being too strict actually. Some things can be annoying, but managing a website like CodeWars is not something easy; things improve with time, but the administrators have a lot to do and I understand they can have other priorities.

You're wrong. I can't approve the pending translations of this kata – because of some trivial changes in the description. But I could easily reject them.

Please... This is a feature on CodeWars: the rules for approvals and rejection are identical. Someone can approve a translation if and only if he can reject it.

That's even more frustrating: It's easy to reject a translation (without even giving a reason or any kind of comment!), but it's hard to approve a kata or translation.

Ceterum censeo: Codewars approval rules are too strict. Too many katas remain in beta forever, too many translations don't get approved.

We cannot know who rejected it unless the person who did it leaves a note somewhere. In general the practice is to justify the rejection with a comment on the translation page or in Discourse.

Why has it been rejected? By whom? I see no reason for it, and no discussion about it.

So I mark this as resolved, since it cannot be approved anymore like I have said.

It has been rejected. It is always possible to do a new fork from it, but the person who publishes the fork cannot approve himself.

I'd be happy to approve it, but now the translation is 'closed'. :-( What happened?

(Ceterum censeo: Codewars approval rules are too strict. Too many katas remain in beta forever, too many translations don't get approved.)