See

https://www.apache.org/legal/resolved.html#weak-copyleft-licenses

Reuse is permitted in binary form only but MPL 2.0 is incompatible for
reuse or derivative works in source form. This means to use any of
such code as a starting point for new source code development in
Apache Arrow is not possible.

On Mon, Feb 17, 2020 at 8:02 AM Raúl Bocanegra Algarra
<[email protected]> wrote:
>
> Hi Wes,
>
> As far as we understand it we use MPL2 because you can use it privately for 
> your projects, but if you modify it you must contribute to our library, 
> keeping the rest of your project private. I wonder what kind of situation 
> could you find that prevents you from using the library in any Apache 
> projects.
>
> Raúl Bocanegra Algarra. C++ Software Engineer.
>
> ________________________________
> De: Wes McKinney <[email protected]>
> Enviado: jueves, 13 de febrero de 2020 22:41
> Para: [email protected]
> Cc: Raúl Bocanegra Algarra
> Asunto: Re: Arrow & Time series analytics
>
> Raúl -- it's an aside, but could you comment on your choice of the MPL
> 2.0 license? It puts you in an odd no-man's land between GPL and
> permissive licenses, neither satisfying GPL proponents nor allowing
> reuse in Apache 2.0 projects.
>
> Thanks
>
> On Thu, Feb 13, 2020 at 10:27 AM Simon Dumke <[email protected]> wrote:
> >
> > Thanks Joris, Wes and Raul for the feedback. I'm still working on 
> > convincing other people in our organisation that we would benefit from 
> > integrating artow into our systems and gradually replacing differen 
> > proprietary, internal data structures ;-)
> >
> > The Khiva project sounds like a good starting point for integrating time 
> > series analytics into the arrow-verse™, although i can not measure from 
> > some shallow looks into the code, whether or how much thosealgorithms are 
> > working on arrow structures. I'll keep an eye on it.
> >
> > Great thanks and kind regards,
> > Simon
> >
> > Am 7. Februar 2020 13:07:06 schrieb Raúl Bocanegra Algarra 
> > <[email protected]>:
> >>
> >> Hi Simon,
> >>
> >> I am actually working in a Time Series data analytics platform using 
> >> Apache Arrow called Shapelets.
> >> We have an open-source library called Khiva which is a collection of 
> >> algorithms optimized for GPU-CPU for time-series analysis.
> >> Currently we use Arrow to share in memory time series between different 
> >> processes.
> >> You can DM for more information.
> >>
> >> Thanks,
> >>
> >> Raúl Bocanegra Algarra
> >>
> >> Software Engineer
> >>
> >> M: + 34 617 83 64 45 -  E: [email protected]
> >>
> >>
> >> ________________________________
> >> From: Simon Dumke <[email protected]>
> >> Sent: Tuesday, February 4, 2020 12:58 PM
> >> To: [email protected] <[email protected]>
> >> Subject: Arrow & Time series analytics
> >>
> >> Hi all,
> >>
> >> I have not yet found anything in that regard - hence my asking here:
> >>
> >> Do you know of any application of, any library about, architecture
> >> around... Apache Arrow for time series data analytics?
> >>
> >> Thanks a lot for the great work!
> >> Simon
> >>
> >>
> >
>

Reply via email to