commit requests

[Erik Hatcher <er...@ehatchersolutions.com>]

I'm in my last ditch procrastinating efforts to get my code4lib pre- 
conference organized where I'll be teach Solr and Lucene, and guiding  
a hands-on workshop to help 85 library geeks come up to speed with  
Solr.  In order to make Solr work out as well as it could there are  
two things I'd love to have in a nightly build from Solr (rather than  
providing my own custom build):

	* SOLR-79: to allow launching a common schema/config with various  
data directories.

	* A newer version of Lucene's JAR: to have *:* syntax.

The *:* syntax is important in being able to select all documents  
since [* TO *] isn't guaranteed.

I think SOLR-79 is close - I can commit if folks think it is fine.

What about updating the Lucene JAR?

	Erik

Re: commit requests

[rubdabadub <ru...@gmail.com>]

> By the way, for many speakers its a classic dilemma to have folks,
> especially those that didn't attend the event itself, to request
> slides.  The best speakers speak way more than their slides say, so
> you miss a lot by simply viewing slides.

I fully agree. But something is better then nothing :-) Beside I am in Asia and
all the fun stuff are in US :-(

Looking forward to it anyway!

Re: commit requests

[Erik Hatcher <er...@ehatchersolutions.com>]

On Feb 12, 2007, at 2:48 AM, rubdabadub wrote:

> On 2/12/07, Erik Hatcher <er...@ehatchersolutions.com> wrote:
>> I'm in my last ditch procrastinating efforts to get my code4lib pre-
>> conference organized where I'll be teach Solr and Lucene, and guiding
>> a hands-on workshop to help 85 library geeks come up to speed with
>> Solr.
>
> Will the transcript/presentation or slides be available afterwards
> somewhere? Hope so :-)

I'll make all zero slides I use for this workshop available to the  
world, sure!

Seriously though, I doubt I'll have much in the way of slides for  
this workshop, we'll dig right in to firing up Solr and working with  
it live rather than walking through slides.

I am, however, packaging up a binary distribution of Solr, the  
example application in both source data .xml files and pre-built  
index, and other sample datasets, such that attendees will be able to  
jump in regardless of their level of experience with Solr and begin  
integrating it into their own environments.  This preliminary  
information will appear here: <http://code4lib.org/wiki/preconf-docs>

I will be making slides for my keynote presentation on Solr Flare,  
and those will certainly be made freely available (and announced here  
when they are posted).

By the way, for many speakers its a classic dilemma to have folks,  
especially those that didn't attend the event itself, to request  
slides.  The best speakers speak way more than their slides say, so  
you miss a lot by simply viewing slides.

	Erik

Re: commit requests

[rubdabadub <ru...@gmail.com>]

On 2/12/07, Erik Hatcher <er...@ehatchersolutions.com> wrote:
> I'm in my last ditch procrastinating efforts to get my code4lib pre-
> conference organized where I'll be teach Solr and Lucene, and guiding
> a hands-on workshop to help 85 library geeks come up to speed with
> Solr.

Will the transcript/presentation or slides be available afterwards
somewhere? Hope so :-)

Cheers

Re: commit requests

[Mike Klaas <mi...@gmail.com>]

On 2/12/07, Bertrand Delacretaz <bd...@apache.org> wrote:

> >...         * A newer version of Lucene's JAR: to have *:* syntax...
>
> +0, don't know enough about the issues to comment.

FWIW, I'm currently using the current lucene trunk with no apparently problems.

2.1 appears extremely close to release, however.

-Mike

Re: commit requests

[Bertrand Delacretaz <bd...@apache.org>]

On 2/12/07, Erik Hatcher <er...@ehatchersolutions.com> wrote:

>...         * SOLR-79: to allow launching a common schema/config with various
> data directories...

+1

>...         * A newer version of Lucene's JAR: to have *:* syntax...

+0, don't know enough about the issues to comment.

-Bertrand

Re: commit requests

[Mike Klaas <mi...@gmail.com>]

On 2/12/07, Yonik Seeley <yo...@apache.org> wrote:

> So I'm fine with updating to the latest Lucene nightly build now... It
> would give us a chance to make sure it works well for us too.  Lucene
> 2.1 shouldn't be too far off... but I'm sick yet again, and it's hard
> enough writing email in this state that I'm not going to go near any
> code :-)

I wouldn't mind waiting until the current bug I just submitted is
fixed in lucene before revving.

-Mike

Re: commit requests

[Yonik Seeley <yo...@apache.org>]

On 2/12/07, Chris Hostetter <ho...@fucit.org> wrote:
> :       * A newer version of Lucene's JAR: to have *:* syntax.
>
> +0 ... i don't really have any opinion about reving Lucene as far as the
> nbightly builds go ... I think yonik menitoned before that it would be
> good to make sure Solr 1.2 used a stable version of Lucene,

Specifically, that Solr 1.2 should use Lucene 2.1 (or later).
There were so many index format changes in Lucene lately, I wanted
some assurances about back-compatibility in the future.

> but i don't
> think it would be bad if nightly builds of Solr (which are by definition
> supose to be considered developer builds and therefore unstable) used
> unstable Lucene builds.

Right.

For Solr official releases, we should prefer offlcial Lucene releases,
but I wouldn't want to make it a "rule".

So I'm fine with updating to the latest Lucene nightly build now... It
would give us a chance to make sure it works well for us too.  Lucene
2.1 shouldn't be too far off... but I'm sick yet again, and it's hard
enough writing email in this state that I'm not going to go near any
code :-)

-Yonik

Re: commit requests

[Chris Hostetter <ho...@fucit.org>]

: 	* SOLR-79: to allow launching a common schema/config with various
: data directories.

+1

: 	* A newer version of Lucene's JAR: to have *:* syntax.

+0 ... i don't really have any opinion about reving Lucene as far as the
nbightly builds go ... I think yonik menitoned before that it would be
good to make sure Solr 1.2 used a stable version of Lucene, but i don't
think it would be bad if nightly builds of Solr (which are by definition
supose to be considered developer builds and therefore unstable) used
unstable Lucene builds.


-Hoss