On Thu, Mar 7, 2024 at 1:32 PM Tom Lane <t...@sss.pgh.pa.us> wrote: > > John Naylor <johncnaylo...@gmail.com> writes: > > On Thu, Mar 7, 2024 at 1:08 PM Tom Lane <t...@sss.pgh.pa.us> wrote: > >> Hm ... do we know that this is not patent-encumbered technology? > >> Commits citing such specific prior art make me nervous. > > > There are several open source implementations in a variety of > > languages, so I assumed not. > > Are any of those implementations used in places that might > entice a patent troll to come after them? > > (If you think Postgres isn't an inviting target for patent > trolls, you're wrong. We've avoided getting sued so far, > but man this topic scares me.)
Understandably so. FWIW, the use case proposed by the authors was for secondary indexes within in-memory databases, not as a type of associative array. I'm unable to find patents for the thing itself, but IANAL. I believe I've been in contact with some of the same authors about a different subject, and they seemed open to people trying to implement their ideas (it was a different paper, to be sure, and unfortunately I no longer email account).
