lucene-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Grant Ingersoll <gsing...@apache.org>
Subject Re: Factor out a standalone, shared analysis package for Nutch/Solr/Lucene?
Date Wed, 03 Mar 2010 14:06:51 GMT
I think it should be clarified, that those who are using Lucene would not be affected at all
by this proposal, other than they would probably get some things from Solr that they wish
they had anyway (Analyzers, likely Faceting, etc).  I would encourage people to go read what
Mike wrote again.  There would still be Lucene jars.  There would still be Solr jars.    All
of those third party projects would still build exactly as they do now, unless of course they
want to add new jars.  Most of the merging is behind the scenes, like a single dev list for
coordination.  I would suspect in practice that most Solr committers would still focus on
Solr, but...

What you would also be getting is less friction (and I don't mean that in a negative way)
about where things should go.  The reason there is often duplication of efforts is mainly
due to the arbitrary boundary put up by the fact that most Solr committers are not Lucene
committers.  So, when a Solr committer comes up w/ something that may belong in Lucene proper
(an Analyzer is just one example) they don't bother to make the effort to put it in Lucene,
so Lucene loses out.
  
-Grant


On Mar 3, 2010, at 1:21 AM, Ard Schrijvers wrote:

> On Tue, Mar 2, 2010 at 6:39 PM, Uri Boness <uboness@gmail.com> wrote:
>> 
>> I disagree here. I believe Lucene still has larger install base than Solr.
>> Think of Jackrabbit which uses Lucene directly and all the CMSs that use
>> Jackrabbit. Think of frameworks like Compass and Hibernate Search (that use
>> Lucene directly) which are used in a lot of JEE deployments around the
>> world. And certainly there are a lot of large infrastructures that use
>> Lucene directly as well (as in LinkedIn for example). Solr is great in what
>> it does but it is certainly not everything when it comes to open source
>> search or Lucene.
> 
> I have been involved in the Lucene implementations of Jackrabbit (and
> before Slide). With respect to repositories (jsr-170 / jsr-283), where
> the backing persistence are some database and the storage is mainly
> key-value's, we use Lucene to do all relational queries (a subset of
> xpath/sql is translated to Lucene queries). This custom Lucene
> implementation (see [1] for overview) can imo never be replaced by
> Solr afaik, the only thing it has in common with Solr is that they
> both use Lucene.
> 
> I agree with Uri that Lucene has a much larger install base than Solr
> 
> Regards Ard
> 
> [1] http://jackrabbit.apache.org/index-readers.html
> 
>> Cheers,
>> Uri
>> 


Mime
View raw message