56% of People's 1st Wikipedia Edits Are Good

ReadWriteWeb
If you thought Wikipedia had seen its heyday, you'd have thought wrong. A small study performed by Wikipedia staff and published today found that new Editors are signing up and making edits …

...and getting blocked for it.

This is a blog referencing a blog... at some point I'm going to have to work out a better RSS feed.

The original blog post shows a pie chart in which a few dozen edits are graded (by whom?), showing that the number of "acceptable" edits is essentially unchanged from 2004 (what about 2006? 2008?). It then claims this is the key statistic, when in fact the chart also shows that the percentage of "vandalism" edits for "new" editors has risen from 2 percent to 23 percent.

But clearly, that can't be significant...

I fixed the apostrophe mistake … maybe Greg won't notice …

Jon

Of course these "new editors signing up" are (by definition) nameuser confirmed accounts. If a quarter of them are now doing vandalism first thing, that only means that these days enough interesting school articles are semiprotected that vandals can't get to them except this way. So the vandals have changed tactics.

But of course this is a harder road for vandals, as nameusers are blocked for vandalizing much more readily, and thus quicker in their "vandal careers" than IPs. So the study needs to look at the natural history of what happens to a new nameuser who starts out of the gate vandalizing sprotected articles (what else?), vs. an IP account that starts off vandalizing first thing. How long does one last, vs. the other? how many average vandal edits before indef block for one vs. the other?

It's still very labor-intensive to do 10 edits and wait 4 days in order to vandalize, and I think it's a deterent. We await the definitive study to show that it is, due to the fact that the WMF "staff" have not done the study yet.

Incidentally, I'd like to see a comparison of what kind of vandalism creeps into academic subjects like zirconium that have a strong weekly page-view cycle (indicating that people have better things to do on weekends than read about zirconium), versus stuff where no such cycle shows up at all, like Justin Beiber or Pokemon, or where (at best) an inverse tendency where slightly MORE viewers show up on weekends. Split WP on the basis of weekly page-view cycle type: with low weekends, vs. not (or perhaps a third category where there are MORE page views on weekends). In a sense you then find you have two or three different encyclopedias coexisting, due to two or three fundamentally different types of readers (and probably writers). Has anybody looked to compare these? Not so far as I know.

The "WMF staff" has never particularly been interested in looking at their various sets of editors and users. It's been a decade since WP started and the study in the news is one of just a handful on the en.wiki since then. Either they don't want to know (prefering to think of their viewers and editors as everyman+everywoman), or they're not used to thinking like researchers. Sometimes it's very hard to tell the difference, of course.

All very good points, but I think it also indicates that they realize, just as much as you and I do, how important vandalism is to them. They have to strike just the right balance between condemning it and encouraging it, so that they'll continue to attract both vandals and vandal-fighters concurrently without making it appear that they're favoring the one group or exploiting the other. So far it's been pretty easy for them, but I get the impression that they're starting to worry a little bit these days.

The more I think about this, the more suspicious - and telling - it seems that they chose to go back to 2004 for the comparison set. That was a year before the explosive-growth period of late 2005 and early 2006; the stakes were much lower then, and that (not coincidentally) was before people started to really game the system, manipulate, POV-push, or whatever you want to call it.

Most people, including those who would later become "vandals," still hadn't heard of Wikipedia in 2004. By late 2006, nearly everything about the site, other than the software and the people in the so-called "inner cabal," had changed... If the WMF even looked at that period at all, I suspect they'd find that edit-quality (however they define it) went down significantly - and since it coincided with a heavy influx of new users, that's the last thing they'd want to publicize.

Forgive my ignorance, but is vandalism down significantly since the 2005-2008 period when Wikipedia went mainstream? If so, I agree that cherrypicking 2004 and the past year for comparison was quite deceptive.

I'm not sure I agree that it matters, but nevertheless I suspect it is down somewhat - if only because they've made it more difficult for "vandals" to get started. Another thing to consider is that WP's automated responses are far better now than they were in 2006 - at which point they themselves were only getting started with anti-vandal bots, high-volume range-blocking, and so on. These days, the really obvious stuff (page-blanking and/or replacing entire articles with obscenities, etc.) is almost guaranteed to be reverted quickly, which is mostly a good thing. Personally, I'm surprised the number is still that high, assuming the 23 percent figure is close to being accurate, which may not be a safe assumption.

The thing they're trying to do here - and what they've been trying to do for years - is put aside the fundamental issue, which is that edits to an encyclopedia should always be reviewed by at least someone before they're published, and simply tell people that because some statistic or other hasn't really changed that much "over time," golly, that must mean that all is well. In most cases, such as this one, they rely on the idea that people these days tend to lack critical-thinking ability - and to some extent that can be relied on. But not everyone lacks that ability, and the overall trend in society is clearly towards tighter control over the internet by governments, law enforcement, etc.

For a long time, I was convinced that Wikipedia would be one of the central focus points of an effort by various authorities to clamp down on internet "freedoms." At the moment it looks like they're focusing more on Google, which is okay since Google is far more powerful, and therefore more scary (at least potentially). Regardless, the law is coming to the frontier, just like it did in the Wild West days, and it's not going to be quite as much fun for certain people once it gets there. It's essential, from Wikipedia's (and the foundation's) perspective, that they not be seen as part of the overall lawlessness problem, which is why we're seeing so much PR from them now. They have to establish themselves as an institution, and they have to make "open editing" seem like a perfectly normal and acceptable thing. So it's just as essential, from the perspective of academia, journalism, traditional publishing, and so on, that they fail in that regard.

Yes! This is the issue!

Thanks...?

I should also say (if only for the sake of completeness) that this is also why many of us have started to "reach out" more lately, showing up in comments sections of blogs, news sites, etc., to challenge people who are saying that Wikipedia is somehow the way a modern encyclopedia is compiled or the way information should be processed for public consumption. Most of the points that can be made about administrative corruption, bad policy, irresponsible management, etc. have already been made here - though I'm certainly not saying we shouldn't continue to monitor and comment on that stuff on an ongoing basis. I'm just saying the bigger danger right now (or maybe the "big-picture danger"?) does seem to be acceptance of the status quo, especially in governmental and corporate/media circles.

In the past, many Wikipedians have actually seen us here at WR as "iconoclasts," fighting some sort of "revolt," as if peer-review and professional accountability are new and radical concepts, rather than integral parts of traditional publishing that have been around for centuries. These days, making sure that particular misconception doesn't spread too much outside of Wikiland is probably just as important as having a "base" like WR for exposing new problems and figuring out what the heck is actually going on over there. I'm not super-worried about it yet personally, because most people outside of WP do seem to have this properly figured out. But hey, complacency is almost always bad.