Side note: dont think toolchain enhancements is a requirement at all, lot
of users keep rejecting this additional work (and to be honest I can agree
1nd it does not help more things than using properties to switch the
env/executable in plugins) so only question for me is the baseline and
minimum requirement when 4 will be final.
It is probably wise to start fresh since it will be frozen for years so the
lts at that moment does not sound crazy to not cumulate debts before
starting.

Le sam. 3 févr. 2024 à 16:43, Tamás Cservenák <ta...@cservenak.net> a
écrit :

> Similarly, this topic (to me at least) somewhat correlates with another
> "wrong reflex" (that somehow get roots in project):
> - compile against maven-core 3.2.5
> - compile with Java 8+ "as we support Maven on Java 8"
> Funnily, both are from 2014, just 10 years old.
>
> IMHO,
> We SHOULD compile against the latest, to pick up the latest deprecations,
> and not be "detached", but follow changes in core (and resolver).
> We SHOULD compile with latest to benefit from latest improvements (how many
> of us said just one thing: is FASTer to build on 21.0.2 than 8.xxx).
> Javadoc is the third (important) thing, and more will just come...
>
> And so on...
>
> T
>
> On Sat, Feb 3, 2024 at 4:36 PM Tamás Cservenák <ta...@cservenak.net>
> wrote:
>
> > The non problems:
> > - members building project with ancient java versions and calling -1 on
> > release votes (as turns out,by mistake)
> > - javadoc inconsistencies: what is allowed and what not
> > - being stuck in a 20 year tech stack
> > ...
> >
> > Again, _build time requirement_ has nothing to do with _runtime
> > requirement_.
> >
> > Or in other words, a bit of rephrasing:
> > For example, desktop app devs who develop apps "certified to work on
> older
> > OS-es (than current)" _develop those apps on "lower end" of supported
> OSes
> > version_?
> > So, if a macOS app CAN work on macOS 12 (current is 14), it MUST be
> > developed on macOS 12?
> >
> >
> >
> >
> > On Sat, Feb 3, 2024, 16:25 Michael Osipov <micha...@apache.org> wrote:
> >
> >> Am 2024-02-03 um 15:16 schrieb Martin Desruisseaux:
> >> > Hello
> >> >
> >> >  From the replies in this thread, it seems to me that there is a
> >> > consensus for moving Maven 4 to some Java version after 8. I see:
> >> >
> >> >   * 0 in favour of Java 11
> >> >   * 1.5 in favour of Java 17 (the .5 is because I split a vote between
> >> >     Java 17 and 21)
> >> >   * 2.5 in favour of Java 21
> >> >   * 4 seem neutral (including myself)
> >> >
> >> > Do we take that as an agreement to require Java 21 for building Maven
> 4?
> >> >
> >> > On a related question, what should be the minimal Java version for
> >> > *running* Maven 4? Keeping in mind that if Java 21 (for example) was
> >> > required, users would still be able to compile for an older Java
> >> version
> >> > using the --release option.
> >>
> >> I still don't understand what non-problem you are trying to solve here?!
> >> I think that your time and our time would be better invested in solving
> >> real problems, just look into JIRA how many issues have piled up.
> >>
> >> M
> >>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
> >> For additional commands, e-mail: dev-h...@maven.apache.org
> >>
> >
>

Reply via email to