Erik Hatcher wrote:Completely unofficially since I'm just a lurker on the Slide lists, but +1 for Stefano and Pier as Slide committers! I'm ecstatic to see the activity here pick up and all this talk of a 2.0 release, Lucene searching, etc.... nice!
Stefano and Pier have not contributed very much to Slide, yet, but both are well known and respected not only in the Jakarta community.
First of all, we should ask them if they wanted to be Slide committers in the first place! I hereby do it: do you want to?
It would make my life easier, as I'm not planning to stop contributing until I have a polished and stable and fast content repository that I can use for my stuff and have fast and scalable DASL search on top of it.
p.s. I don't have time to really get deep into DASL within Slide just yet, but I am in the midst of co-authoring a book on Lucene - so if anyone needs any assistance I'd be more than willing to answer questions or assist in some minor ways. I do have big plans with Slide and Lucene in the near future once the book is done, so this is definitely an area of keen interest for me.
We were discussing if it was possible to use Lucene not only for full text, but also for property indexing.
I don't see how that's possible given how lucene works (is uses vector-space analysis... how would metadata properties look like in vector space?)
Maybe using fields? Then for properties like size and "date of last change comparators" like < and > would be great. I do not think this is supported by Lucene, is it? Could it be possbile / reasonable to extend Slide for this?
I think it makes *much* more sense to do what Catacomb is doing: put the content on file system stores, put the properties on database store and translate the DASL query into SQL and let the database do the nasty stuff.
that's what I plan to do.
-- Stefano.
smime.p7s
Description: S/MIME cryptographic signature
