rmannibucau commented on PR #765:
URL: https://github.com/apache/maven/pull/765#issuecomment-1183036354

   > In short, IMHO we should nag users (and indirectly plugin devs as well) to 
keep up, especially if they want to sit on Maven 3.9 train that is taking them 
to Maven4.
   
   This is where this all fails IMHO, it assumes there is always a new plugin 
version but in practise it is not true and what is probably worse is that it is 
not even needed technically. Would be really great to avoid to request releases 
without any change except a version upgrade + keep one of our strength which 
was a relatively high portability of the build accross versions compared to 
concurrents.
   
   > Well, I believe they do not update their operating system, Java or 
anything, because it also "just works". Also, if it "just works" why would they 
go with Maven 3.9 (why would the go the fuss to upgrade maven in the first 
place?). These kind of statements are full of contradictions for me, so I never 
considered them.
   
   This is exactly the opposite, they upgrade their work tools (maven , java, 
os etc) but the projects are under rules which are not "on your machine" so you 
can not always upgrade and the key point is previous one, we cant assume all 
plugin writers will release a plugin version per maven version - once again 
technically there is no reason we request that.
   
   > And no, we will not break EVERYBODY, we will break (well, they will break 
themselves when upgrade to Maven4) those users who use Maven2 plugins, that's a 
huge difference: we have no resources to support compatibility across 2 major 
versions. Maven3 did support Maven2 plugins, just like Maven4 will support 
Maven3 plugins (but not Maven2). I think this is really fair expectation, but 
if not, stick with 3.8.6 as long as you want (the user I mean).
   
   Yep but the current implementation does not respect that. I only have two 
points in my comments:
   
   1. the warnings are not consistent and the heuristic used does not work so 
we should just go without another one,
   2. hope it is not a default/all-the-time check (could be a daily like 
snapshot check) - I care more about seeing the warnings all the time than the 
perf for now until the algorithm used will be very slow but it is not yet the 
case


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscr...@maven.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

Reply via email to