Definitely +1 (even) for non-MR patches.

Otis 

--
Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch

----- Original Message ----
> From: Grant Ingersoll <[EMAIL PROTECTED]>
> To: [email protected]
> Sent: Thursday, February 14, 2008 3:57:33 PM
> Subject: Re: Community development was Re: [jira] Updated: (MAHOUT-4) Simple 
> prototype for Expectation Maximization (EM)
> 
> Right, the question is more along the lines of would people rather see  
> patches committed even if they aren't in M/R form?  It is a given in  
> my mind that at a minimum they have to compile and have reasonable  
> unit tests, etc.  Taking this approach would allow people to put up  
> initial standard patches, and then give others room to improve  
> separately with M/R implementations, etc.
> 
> Just trying to get at what people will find the most useful for  
> getting code that people can work on, extend, etc.
> 
> -Grant
> 
> 
> On Feb 14, 2008, at 3:20 PM, Ted Dunning wrote:
> 
> >
> > Patches should be for the continuous integration system and for the
> > committers, not normally for users to use for building the code.  They
> > should either download a release or they should download the trunk.
> >
> >
> > On 2/14/08 6:58 AM, "Goel, Ankur"  wrote:
> >
> >>> This brings up an  ... find patches?
> >>
> >> My vote is in favour having people check-out code directly instead of
> >> finding patches.
> >
> 
> 
> 


Reply via email to