Not surprised, but had different priorities.
I agree with you that getting this kind of feedback and resolving the
critical items that we find in it should be a priority towards getting
Maven 4 out. Good to see such problems are found, this is a strong
indication we should keep encouraging users to test drive the release
candidates of Maven 4.
Apart from that, I think there are some open discussions around the API
of Maven 4 and the use of XML namespaces. Without expressing an opinion
on either topic (there are separate email threads that cover them) I
think we need to somehow close these threads. Ideally by vote and
registering the _why_ of the decision that we collectively make. The
latter would hopefully prevent the discussion from opening again in a
few months (with Maven 4.1) or years (with Maven 5) - timelines are of
course exemplary, not a real estimate.
Maarten
On 27/03/2026 08:40, Hervé Boutemy wrote:
no feedback: does it mean everybody is surprised? agrees/disagrees?
I worked on establishing the status of ASF Maven plugins:
https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=406620656#Maven4.0.0GAchecklist-Maven4API
I all starts by establishing if existing Maven 3 plugins work with Maven 4 or
not: I did it for plugins that went to a Maven 4-specific branch
I think doing that check on ALL ASF Maven plugins would make sense, to establish how our
"Maven 3 plugins compatibility plan" works or not
and when a plugin fails initially, establishing why, and if a fix can be done
without creating the Maven 4 branch, would be useful
Help welcome
Hervé
On 2026/03/22 03:35:46 Hervé Boutemy wrote:
in https://cwiki.apache.org/confluence/pages/viewpage.action?
pageId=406620656#Maven4.0.0GAchecklist-Build/ConsumerPOM
I added a section on Build / Consumer POM
in short:
1. testing Maven 4.0-RC5, consumption of resulting artifacts published to
Maven Central failed for Maven 3, Gradle, sbt, ...
2. this one is a strong blocker
Guillaume already worked on describing the issue, and preparing a PR: need
help on getting it reviewed and merged
During a conference, I promoted testing of Maven 4, and got real feedback from
real users working on non OSS projects, and trusting us when they saw the new
simplified model that looked so great:
- description of failure: https://github.com/apache/maven/issues/11767
- fix proposal: https://github.com/apache/maven/pull/11768
To me this is where we should focus our common energy on getting Maven 4.0.0
out.
We're not that far, but need to have a common focus
Regards,
Hervé
Notice: the guy also started to propose me very good description of how we
could enhance dependency XML to be shorter. I loved his very concrete and good
proposals, but chose to not share that publicly yet as I don't want us to open
a new debate that does not match Maven 4.0 but more Maven 4.1+
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
