Hi Henning, Please take the time to read the beginning of the thread, the proposal is to drop surefire+failsafe since they are superseeded by junit platform, it literally means supporting all test engines (including junit 3, 4, testng, spock, cucumber, ....) and common reports (mainly xml in our ecosystem - but excludes the maven site). I also explained that there is already a partial junit extension reading surefire config which literally enabled me to drop surefire in a project without changing its configuration. So think both of your points are solved since more than 5 years at least.
Also I totally disagree with your last point which got proven wrong last dozen of year: if we do not change now we will never change before maven 5 discussions start, we are not able to break between major releases (which is not bad, do not get me wrong, but this is also why there is this thread *now*). I do not care much if we have the surefire bridge extension, this is a nice to have but from my experience it will not be a big blocker (surprisingly) for end users to not have it after a short period of time and we know how to make surefire active for maven 3 projects and the replacement for maven 4 native projects (based on modelVersion). So we literally have no blocker to do it, to make it smooth for end users whatever way fits, and to gain a lot in maintenance since plugin will be minimal. Romain Manni-Bucau @rmannibucau <https://x.com/rmannibucau> | .NET Blog <https://dotnetbirdie.github.io/> | Blog <https://rmannibucau.github.io/> | Old Blog <http://rmannibucau.wordpress.com> | Github <https://github.com/rmannibucau> | LinkedIn <https://www.linkedin.com/in/rmannibucau> | Book <https://www.packtpub.com/en-us/product/java-ee-8-high-performance-9781788473064> Javaccino founder (Java/.NET service - contact via linkedin) Le mar. 7 oct. 2025 à 04:52, Henning Schmiedehausen <[email protected]> a écrit : > On Mon, Sep 29, 2025 at 6:07 AM Romain Manni-Bucau <[email protected]> > wrote: > > > Le lun. 29 sept. 2025 à 14:51, Christoph Läubrich <[email protected]> > a > > écrit : > > > > > I think this thread really starts confusing. > > > > > > 1. surefire/failsafe is not part of maven so it can not be "dropped" > > > from maven 4 > > > > > > > You're partially right, let me add what was ambiguous: > > - do not port surefire/failsafe to maven 4 api > > > > That is not an option. If we start dropping plugins as part of the "3 -> 4 > transition", most serious users will simply not move to 4. > > Your options are: > > - port everything to Maven 4. Make transitioning from 3 to 4 as painless as > transitioning from 3.8 to 3.9. Or, better, 3.9.10 to 3.9.11. That will > give you the most users for Maven 4. > - do not port everything over. Some users will transition from 3 to 4 and > eat up the pain of redoing their builds. The majority (and especially > larger users) will > - either remain on Maven 3 indefinitely. This is the best case scenario > - look at the transition cost, ask themselves if they want to keep > investing in Maven if the community pulls stunts like that or move to > gradle. The gradle company will post case studies (and tools) to make Maven > 3 -> gradle transitions simple. You lose your users. Maven 4 becomes > irrelevant. > > Java is a mature ecosystem. Not a growing ecosystem. Unless there is an > actual user base for Maven 4, there is no point of releasing it. You really > do not want to split your user base at the most critical moment when you > need them to trust and follow you across a major version transition > (something we have not done in 20 (!) years). > > Once we managed to get users "over the hump" to Maven 4, we can talk about > making some more substantial changes. Like in the 4.1 or 4.2 cycle. > > -h >
