Michael McCandless

This sounds reasonable to me @pawel-bugalski-dynatrace but I'm not familiar with Monitor/Luwak's code. It looks like there are conflicts -- is this PR still relevant? Thanks @pawel-bugalski-dynatrace!

Thanks @gokaai -- I'll try to review soon! If possible please try not to force-push: it removes the history of the past commits and makes it harder to see what...

> > Thanks @gokaai -- I'll try to review soon! > > If possible please try not to force-push: it removes the history of the past commits and makes it...

Hi @gf2121 -- did we ever performance test this change? It sounds compelling!

This sounds like a nice optimization @wuwm! Is it still relevant? Lucene's nightly benchmarks include [somewhat sparse documents (NYC taxi database)](https://home.apache.org/~mikemccand/lucenebench/sparseResults.html) -- maybe this could be used to test the...

Oooh I missed this @uschindler -- it looks like a nice possible opto for the costly `BytesRefHash` methods, and it looks like (on the issue) you and @rmuir came to...

> @uschindler pushed 0 commits. Huh, how do you do that? Mike McCandless http://blog.mikemccandless.com On Fri, Nov 3, 2023 at 10:42 AM Uwe Schindler ***@***.***> wrote: > @mikemccand : If...

Thanks @uschindler! Removing vShort and switching to LE (or native -- I didn't understand the problem with that -- this is never (directly) serialized to a Lucene index) short seems...

@jimczi -- is this still relevant? Thanks.

Thanks for the idea @ldkjdk! It looks like we are unsure this is helpful in the general case ... I'll close the PR for now. Please re-open if you feel...