Re: [rdf4j-dev] Solr and Elastic Search

[Jeen Broekstra <jeen.broekstra@xxxxxxxxx>]

> On 29 Jul 2017, at 20:57, James Leigh <james.leigh@xxxxxxxxxxxx> wrote:
> 
> On Sat, 2017-07-29 at 08:23 +1000, Jeen Broekstra wrote:
>> Well, don't forget that many people may be using it through maven.
>> I'm pretty sure Marc himself uses these backends, and there have been
>> several stack overflow and mailing list questions about them. Also
>> they haven't been around that long, and I think they're a nice
>> proposition: we give people a choice of text indexing tool, something
>> few other frameworks do I think.
>> 
>> Is there a particular reason you want to get rid of them?
> 
> I think we should either include them in the sdk or drop them
> altogether.

I think include then, at least for now. I heard via private mail from Mark (he’s having difficulties getting through to the mailing list, for some reason his mail keeps being marked as spam), and he/his employer is definitely still actively using the ElasticSearch Sail. 

> It can be difficult to find a balance between features and
> size. I don't want to see this project get too big to manage so iff
> they aren't being used then that seems like an easy choice.

I understand the concern. Perhaps an alternative would be that we split the distribution, into, say, rdf4j-core and rdf4j-extras? If we go all out we could make rdf4j-extras a subproject with its own release cycle, even. That way we can keep the core project lean, but also support contributions that have useful addons. 

Cheers,

Jeen