A child's garden of wikipedia, part I

ScienceBlogs (blog)

I've been using Google Reader recently, following the lamented death of Planet Fleck, and I suppose I have to admit its better. Here are some "shared items"...

View the article

Connolley upset that a journalist got something wrong about his WP efforts .

Well, so what, Connolley? It can still be used as a cited source now for a Wikipedia article! Unlike your BLOG.

Wow, tough old world, huh?

Dear god, he writes the way I'd imagine FT2 and Brad do when they're drunk. Does he really expect people to read this shite?

Um, I'm afraid Connelley is (mostly) right in this instance.

The original "story" being quoted by Climate-Change deniers all over the internet is this online op-ed written by Lawrence Solomon, who is described thusly:

OK, so he's a (former) journalist. But the numbers Solomon uses are so obviously misleading and deceptive as to be laughable to anyone who understands how Wikipedia actually works. He probably used the "WikiChecker" to get the total number of articles Connelley ever touched, and then just used that as the "total number of Global Warming articles controlled by Connelley," and assumed nobody would bother to actually check.

I'd just have to guess that Solomon felt he couldn't make the case for Connelley's POV-pushing unless he presented it as far less subtle than it actually was - to the extent that he presented it as not being subtle at all. And, naturally, that people's attention spans and their willingness to check the facts on these things is so limited that he could pretty much say whatever he wanted, and it would be repeated uncritically by everyone who has any anti-environmentalist leanings whatsoever.

Connelley may have been a terrible admin, and clearly took many, many liberties on GW-related subjects, but what I've been seeing with this Solomon opinion piece, and the degree to which it's been disseminated without anyone questioning it, is really frightening.

And it pisses me off, too, because here we've spent months making the case against Connelley, with diffs and links and anecdotes from real experiences, showing that he's a terrible choice to defend the environmentalist position on WP - and then this other guy comes along, out of the blue, spouts ludicrously wrong numbers that can be knocked down by anybody with five minutes to spare, and he's copied and quoted en masse by practically every blogger on the planet. It really makes you wonder.

5,474 is actually total number of distinct pages Dr. Connolley has edited, in all namespaces. I believe most would consider this low in relation to his total edit count.

The number of distinct articles he has edited is only 1,716.

In other news the local weather-guesser says we're on pace for the coldest winter since 1988. Granted I can't vouch for this as I'm not a meteorologist and I don't have the raw numbers handy, etc.

Right. Well then, thanks for making the situation look even worse!


I can believe that, at least - where I am we're into our sixth night of subzero temps in a row, which is pretty unusual here, even for January. That's on top of two major snowfalls for a total of over 2 feet, very little of which has melted...

Well here are is his page-pwnership indexes for each namespace, if anyone cares. This may imply that Dr. Connolley is slightly more obsessive toward talk-pages than articles, but I won't form any conclusion beyond that.

ns# ns_name: edits/pages (ratio)

0 [article]: 13142/1716 (7.66)
1 Talk: 8416/863 (9.75)
2 User: 949/194 (4.89)
3 User_talk: 10000/2654 (3.77)
4 Wikipedia: 4911/598 (8.21)
5 Wikipedia_talk: 728/148 (4.92)
6 File: 221/103 (2.15)
7 File_talk: 56/14 (4.00)
8 MediaWiki: 0/0 (—)
9 MediaWiki_talk: 1/1 (1.00)
10 Template: 45/9 (5.00)
11 Template_talk: 11/6 (1.83)
12 Help: 0/0 (—)
13 Help_talk: 0/0 (—)
14 Category: 9/5 (1.80)
15 Category_talk: 6/3 (2.00)
100 Portal: 1/1 (1.00)
101 Portal_talk: 0/0 (—)
108 Book: 0/0 (—)
109 Book_talk: 0/0 (—)
------------------------
total: 38496/6315 (6.10)

Interesting note, the numbers shown per namespace on Soxred93's "Edit Counter (version 4.0.0.49)" do not add up to the total given by the same tool. Looks like that total is based on the total number of edits he has made, ever (given as 39,799), subtracted by how many exist in the `archive` table or in common parlance "are deleted" (given as 1,300). This yields a difference of 38,499.

However, double-checking firstly the namespace totals for that tool against my numbers, and secondly that he has not edited in the past few minutes, I still come up with only 38,496. This leads me to believe exactly three of his edits have been oversighted (I'm sure somebody somewhere could confirm this, not that they will), but I won't speculate as to the nature of these edits or reason for their removal.

But if somebody here wants to dumpster-diving, knock yourself out…