lucenenet-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Granroth, Neal V." <neal.granr...@thermofisher.com>
Subject RE: Porting Automation - Sharpen
Date Wed, 17 Nov 2010 19:49:12 GMT
Thanks for repeating this George.

Everyone's situation is different I suppose.  For me, adding the IKVM framework would make
Lucene.NET unsuitable for distribution in the context in which I work.  In addition to software
quality/reliability concerns and performance degradation, its use adds another layer of licensing
complexity that I would have to justify and defend.

- Neal

-----Original Message-----
From: George Aroush [mailto:george@aroush.net] 
Sent: Tuesday, November 16, 2010 10:09 PM
To: lucene-net-dev@lucene.apache.org
Subject: RE: Porting Automation - Sharpen

I don't mean to discourage folks by posting this, but since this topic
(IKVM) keeps coming up (we need a FAQ), here is my comment on it.

>From Lucene.Net homepage (http://lucene.apache.org/lucene.net/): 

"Lucene.Net is a source code, class-per-class, API-per-API and algorithmatic
port of the Java Lucene search engine to the C# and .NET platform ..."

IKVM will not give you the above (or the other advantages of line-by-line
port which have been pointed out already as well as in the mail archive).

If anyone feels IKVM will meet your need, you are welcome to use it or even
start a project for it.  Producing a C# usable version of Java Lucene using
IKVM is a snap.  In addition, you can use IKVM to do the same for "contrib".

PyLucene (http://lucene.apache.org/pylucene) is such a port example.  Once a
Java Lucene is released, in few days a PyLucene is release.

-- George


-----Original Message-----
From: Hans Merkl [mailto:hm@hmerkl.com] 
Sent: Wednesday, November 10, 2010 12:52 PM
To: lucene-net-dev@lucene.apache.org
Subject: Re: Porting Automation - Sharpen

I remember reading that IKVM is about 10% slower than Lucene.NET. Somewhere
else I saw that the performance was even with Lucene.NET. For me definitely
not a showstopper.
And if the performance is a showstopper I'd rather put the resources into
improving IKVM.

On Wed, Nov 10, 2010 at 10:47, Ryan Hoffman <rhoffman@tntp.org> wrote:

> I would say that performance is paramount, and I'm assuming that everyone
> else would agree.  I think it's a great idea to try both IKVM and Sharpen
> and then make a few benchmarks that we compare both.
>
> Ryan Hoffman
> Software Architect
> The New Teacher Project
> www.tntp.org
>


Mime
View raw message