I'm +1 on removing (effectively) retired products.
While deprecating in the next release and
removing in after the next might be usual way,
I think it would be ok to remove them in 1.5.0.
Each product in Bigtop has dedicated subdirectories.
Removing (and reverting the removal) should be easy
unless it is depended by other products.
Thanks,
Masatake Iwasaki
On 3/16/20 15:13, Konstantin Boudnik wrote:
Thanks for bringing this up, Evans!
I think this trimming makes total sense and we were through similar
discussions before. In fact there's a consensus that we simply can't
support all the projects that seem to be nice, especially those that
weren't been active in 3+ years.
I am in favor of trimming the stack down. Unless, there's a maintainer
in the community who wants to step up and help.
--
Regards,
Cos
On 3/16/20 1:05 PM, Evans Ye wrote:
Hi folks,
I'd like to raise the discussion about the cleanup of supported
component
matrix.
From [1] the latest release of Tajo is 0.11.3 which was out on
2016-05-18.
Since then there're no further release. Similarly on Hama side the
latest
release is on Jan 28, 2016 [2]. Apex is on Attic so there's no chance to
get new release further [3].
Though we'd like to support as many projects as possible, there's simply
not enough effort and resource to do it. Dropping them can also let
us more
focus on what's valuable to the users. What do you think?
Kengo Seki
As 1.5 RM what do you think? I think if the discussion is positive this
will be in 1.5 release.
[1] https://tajo.apache.org/
[2] https://hama.apache.org/
[3] https://apex.apache.org/
Evans
