[Search-l] [canonizers] Fwd: Re: NPOV for Search?

[Bryan Bishop]

On Tuesday 08 January 2008, Jimmy Wales wrote:
> I wonder if my explanation in the previous email was helpful in terms
> of explaining how I think NPOV works for search.

Yes, absolutely. I agree with the examples you have been giving.

> > After all. The scoring algorithm on the search engine *demands*
> > that it is no longer a neutral point of view. That's the entire
> > idea of ranking in the first place. So they can't say they want
> > NPOV. ;)
>
> I don't think this is true, really.
>
> Suppose I go to a search engine and I search for "Barack Obama" and I
> get all positive or neutral pages.  And then suppose I search for
> "Hillary Clinton" and I get all negative pages.  I think we can say
> that neither of those search results is good and, intentionally or
> unintentionally, this search engine as a whole has a bias.

Agreed.

> Rather, neutrality would mean that I get good quality pages from all
> reasonable perspectives in a debate.  Yes, crackpot opinions might
> show up further down (page 2, page 4, lower still perhaps).  But I
> want to be given a balanced selection of good web pages.

Aha. That's it right there. "I want to give a balanced selection of good 
web pages." What is good for one person, is bad for another. Granted, I 
happen to like your taste in Google search results -- personally I have 
been living off of them for the past decade (give or take two years). 
But that's because, for my POV, those are useful and relevant.*

What about a balanced selection of 'bad' web pages? What about a search 
that can leave the moral compass (that's what good/bad is, after all) 
up to its users?

Anyway, some of the guys in #wikiasearch on freenode have pointed out 
that this is too philosophical and getting into some 'heated debate'. I 
want to say that this is not my intention. I started this thread for 
the expressed purpose of adding value to search engines. I think that 
this discussion can result in fruitful code. (But, equally, in good 
intentions, I'll readily stop if our benevolent leader asks me so.)

John's email of today @ 9:41:26 pm says these things well.

Just a quick note- I sent forwarded these emails to the Canonizer group 
because they are specifically specializing in structuring social trust 
and 'canonized' beliefs and opinions that would be valuable to a search 
engine with a 'social backend'.

- Bryan

* Most of the time. I just had a good, long chat in #physics about how 
the results have been increasingly poor on Google for scientific 
research, how much of it is inaccessible (behind paywalls), etc., but 
right now Google is the best option I see. (Maybe Scirus??)
________________________________________
Bryan Bishop
http://heybryan.org/