Probably not the best to release to JCenter now

# feed

[Kirill Grouchnikov]

Probably not the best to release to JCenter now

[Hamza GATTAL]

What's the main reason

[Rob Elliot]

https://jfrog.com/blog/into-the-sunset-bintray-jcenter-gocenter-and-chartcenter/

[Rob Elliot]

Discussion here: https://kotlinlang.slack.com/archives/C0BJ0GTE2/p1612372634063700

[Vampire]

Well, it's probably still a good idea to release updates to JCenter for existing consumers.
And for new ones there is already the alternative repo on Space.

[gammax]

As a side note: you can publish to JCenter only till the end of the month (Feb 28th). Afterwards it will become “read-only”.

it’s probably still a good idea to release updates to JCenter for existing consumers

I’d like to disagree here. Given the short timeline, it’s probably better to just push new releases to Central only and make sure consumer migrate.

[Kirill Grouchnikov]

Indeed, I don't think you're serving existing consumers well by pretending business is as usual with JCenter

[tddmonkey]

I agree. Use new versions as a carrot to get people to eagerly move.

[altavir]

It is still working. And I am not happy with central release process (they are still did not request for my registration request for my university address. kotlin.link repo is a nice substitute and I am using it as a proxy for the Space repo.

[altavir]

And of coures, I can't push this one to central since kscience domain does not exist. I've bought kscience.space just in case though.

[Kirill Grouchnikov]

You're just postponing the inevitable by three weeks

[altavir]

Maybe. But I have to do breaking change and switch package again to publish to central, so I will wait and see if there are some other ideas.

[Dominaezzz]

What's this about the alternative repo on Space?

[Vampire]

What is your question actually?
In his announcement he present an alternative repo on Space additonal to JCenter.

[Dominaezzz]

Ohhh, didn't click the link. Thanks.

[Robert MacLean]

Use new versions as a carrot to get people to eagerly move.

Assumes people check this; so many projects get the bits and don’t check until it breaks.

I think it is a wise move but don’t think it will have any major impact. Actually need something like gradle to start raising a warning

[Vampire]

Which would assume people update Gradle more often than their libs. :-D

What you actually need is an intentionally broken artifact, so that at least people looking in JCenter or via some plugin for lib updates are getting "notified" :-D

[Robert MacLean]

fair shout 😆