[WikiEN-l] An expert's perspective - Tim Bray on editing the XML article

[Abd ul-Rahman Lomax]

At 05:01 PM 8/14/2009, you wrote:
>The problem comes not in finding sources, but in establishing due 
>weight, convincing anyone that a crank idea is a crank idea and so forth.

That something is a "crank idea" is rarely found in the highest 
quality reliable sources. So trying to establish it is divisive, bad idea.

>  Most usually, it's when an expert is arguing with a crank and the 
> crank won't be satisfied until the expert proves a negative - 
> there's no great sources that the crank idea is a crank idea 
> because only the cranks even bother talking about it.

Then it might not be a crank idea at all, merely an unknown one. 
Editorial consensus can handle this easily, if the goal is consensus, 
figuring out what we can agree on, instead of getting stuck on what 
we can't agree on.

Jimbo pointed out long ago that cranks know that their ideas are not 
accepted, it's often a big deal to them, in fact. So say that. "There 
has been little acceptance of this idea among mainstream scientists." 
-- and you only say that if there is *some* acceptance.

>  Often, the expert goes "bugger this, I have better things to do."

Sure. The expert gives the advice, and then leaves it to the rest of 
us. An expert should not go on arguing once the facts have been 
stated. I believe that if we establish that certain editors claim to 
be experts, we take some special care with what they contribute and 
don't expect them to "prosecute the case," both the experts and the 
project will be better off.

>  Even quite patient experts have a limited tolerance for idiocy. 
> For an extreme case, look at the first global warming arbitration 
> case, where the cranks got together to try to get one of the UK's 
> top climate scientists voted off the island. Fortunately, the AC 
> had the presence of mind to point out that peer-reviewed scientific 
> papers are rather better encyclopedia sources than Rush Limbaugh 
> show transcripts. And the expert in question also happens to be a 
> rather good Wikipedian. Abd's proposed rule is pathologically 
> anti-expert and would be disastrous for Wikipedia's content and its 
> production process. - d.

Oh, no, not that article, the site of long-running, unresolved 
disputes that never go away! They don't go away precisely because 
decisions were made that favored one side. Happens to be the most 
sensible side, my opinion, but you don't resolve disputes by crushing 
opposition. Absolutely, you give more weight to the highest quality 
sources, but if the other side is in reliable source (are Rush's show 
transcript's reliable source? They might establish notability, 
*maybe*, but not for an article on the science). There really are two 
topics here: science and politics, and we mix them up; they have 
different standards for sourcing. *Maybe* Rush's transcripts would be 
appropriate for a fork, an article like "Global warming denial," and 
I think there is such an article, something like that. Rush's 
transcripts tell us nothing about the science, but they may tell us 
much about the political controversy.

I can easily understand that my proposal may seem strange, but it is 
not at all anti-expert. As matters stand, experts tend to get blocked 
with high frequency, experts are only protected if they attract some 
protection. I've seen true experts harassed and even blocked because 
they wrote things that non-experts thought were "fringe." They 
weren't. They were within the mainstream, simply not the view that 
another editor was familiar with.

Yes, experts can become uncivil, rather easily. I'm saying we should 
protect them, respect them, and require them to find consensus if 
they want to push something, or stick with clear advice, leaving it 
to the rest of us to interpret it and use it. The rest of us also 
need to find consensus, as well. If we do that, if we value consensus 
instead of being "right," in a way that excludes others and makes 
them "wrong," we will do much better.




>  _______________________________________________ WikiEN-l mailing 
> list WikiEN-l at lists.wikimedia.org To unsubscribe from this mailing 
> list, visit: https://lists.wikimedia.org/mailman/listinfo/wikien-l