Hey Uwe et al,
thanks for bringing this up.

I understand the wish to align Lucene majors with Java LTS releases, that
sounds like a good idea.

I do wonder how many users moved to Lucene 10 since its release, and I
worry that a new major this year or early next year may make it harder for
Lucene users to keep up with it and upgrade, especially as all data indexed
in Lucene 9.x will require reindexing.
At the same time, I am not super convinced that a new major so far provides
enough value compared to its costs. Is there a risk that releasing Lucene
11 early will delay its adoption?

If the main motivation is aligning with LTS java releases, and we'd do a
new major just a year after the last one due to that, I think we should
consider broadening the write compatibility as a one-off, so that Lucene 11
requires no reindexing of data indexed in Lucene 9.x. Is that technically
doable and sustainable, in special circumstances?
When it comes to reading from older versions, could we leave the same
versions that are supported in Lucene 10 readable in Lucene 11 (hence all
the way to Lucene 8.x)?

Otherwise, if some ad-hoc data compatibility policy is not considered for
Lucene 11, what would be the consequences of releasing it later than what
you proposed? I am not necessarily advocating for this but I think it would
be good to consider it as an option, and enumerate pros and cons.

Cheers
Luca






On Wed, Jul 9, 2025 at 4:41 PM Dawid Weiss <dawid.we...@gmail.com> wrote:

>
> You're right - there are still some out there.
>
> On Wed, Jul 9, 2025 at 3:14 PM Uwe Schindler <u...@thetaphi.de> wrote:
>
>> Oh I missed that one. Did we also remove all "catch SecurityException", I
>> have the feeling I saw one yesterday.
>>
>> Uwe
>> Am 09.07.2025 um 15:00 schrieb Dawid Weiss:
>>
>>
>> +1.
>>
>>
>>>    - cleanup of all AccessController/SecurityManager/SecurityException
>>>    (to be done...., will open PR soon
>>>
>>>
>> This is already done, I think.
>>
>> https://github.com/apache/lucene/pull/14801
>>
>> D.
>>
>>> --
>> Uwe Schindler
>> Achterdiek 19, D-28357 Bremenhttps://www.thetaphi.de
>> eMail: u...@thetaphi.de
>>
>>

Reply via email to