[RAS] test on ras.openlib.org

['Christian Zimmermann']

On Thu, 17 Nov 2011, Thomas Krichel wrote:

>  Thomas Krichel writes
>>   Ivan Kurmanov writes
>>>>
>>>> I did a search for my documents. default should be on 'yes' for items
>>>> corresponding to name variations, 'no' for the others. All are 'yes'
>>>> currently. Or maybe put no choice by default?
>>>
>>> Thomas, any additional configuration needed for that?
>>
>>   yes. I tihnk the default is to use learning.
>
>  The way the software works now is taht it uses learning
>  to order the
>
> above-me-propose-accept = 0.7
> below-me-propose-refuse = 0.699999999
>
>  You can set these in the configuration.
>
>  You can certainly bulid  the "no choice" option by setting
>
> above-me-propose-accept = 1
> below-me-propose-refuse = 0
>
>  But I believe this is suboptimal.
>
>  A crude rule that comes from name variation matching
>  is no longer supported, but it could be supported again
>  with code changes.

Do I understand right that name variations are now useless?

>
>  Cheers,
>
>  Thomas Krichel                    http://openlib.org/home/krichel
>                                      http://authorprofile.org/pkr1
>                                               skype: thomaskrichel
>
> _______________________________________________
> RAS-run mailing list
> RAS-run at lists.openlib.org
> http://lists.openlib.org/cgi-bin/mailman/listinfo/ras-run
>

-- 
Christian Zimmermann                          FIGUGEGL!
Economic Research
Federal Reserve Bank of St. Louis
P.O. Box 442
St. Louis MO 63166-0442 USA
http://ideas.repec.org/zimm/