lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chris Hostetter <hossman_luc...@fucit.org>
Subject Re: The state of Solr 5. Is it in maintenance mode only?
Date Fri, 09 Dec 2016 01:17:32 GMT

: On the 5.x front I wasn't expecting 5.6 release now that we have 6.x but
: was simply surprised to see fix for 4.x and not for 5.x.

As Shawn mentioned: jira issues might have inccorrect fixVersion info if 
people don't pay enough attention when resolving (especially with 
dups/invalid) but what really matters is what gets committed/released.

Wether or not there are any future 4.x "bug fix" releases depends entirely 
on the severity of the bug and the demand from users for fixes -- 
particularly for security relatd bugs, you might see more effort put into 
backporting "farther back" in the release timeline and releasing a 4.10.x 
release.  

: As for adoption levels, it was my subjective feel reading this list. Do
: we have community survey on that subject? That would be really
: interesting to see.

As a developer, my focus is on building features and fixing any bugs found 
in the current major version release branch, w/o worrying too much about 
bugs that may only affect older previous major versions.  If someone finds 
a bug in 5.x, and that bug no longer exists in 6.x, I have less 
interested/motivation to look into that bug then something else that 
*does* affect 6.x, because there is already a fix/workaround available...

	Upgrade to the latest version.

while some users might have (completley understandable) mitigating factors 
preventing them from upgrading, that doesn't really affect my 
interest/motivation in fixing bugs on older branches, because users who 
have reasons preventing them from upgrading to recent major versions 
frequently tend to have one thing in common: They have things preventing 
them from upgrading at all.

So even if i put in the effort to find/diagnose/fix an old bug, and even 
if the project as a whole goes to the effort to build/test/release from an 
"older" major version dev branch, the return on investment for that work 
is lower then putting the same amount of effort into bug fixes on a 
"newer" major version dev brach.

For example: let's say hypothetically the Solr user base was devided 
evenly into 3rds: 1/3 using 6.x.0, 1/3 using 5.y.0, 1/3 using 4.z.0.  In 
theory, if 3 diff bugs affect each of those 3 versions to the same degree, 
then the number of users impact by a 4.z.1 bug fix would same as the 
number of users impacted by a 5.y.1 or a 6.x.1 bug fix -- but in practice, 
the number of 4.z.0 users who are likely to upgrade to 4.x.1 is much lower 
then the number of 5.y.0 users who would upgrade to 5.y.1, which is less 
still then the number of 6.x.0 users who will upgrade to 6.x.1. 


-Hoss
http://www.lucidworks.com/

Mime
View raw message