The maths just doesn't add up

Good, bad, biased, paid or what-have-you. There's an endless supply.
Post Reply
User avatar
CrowsNest
Sucks Maniac
Posts: 4459
Joined: Tue Feb 27, 2018 4:50 am
Been thanked: 5 times

The maths just doesn't add up

Post by CrowsNest » Tue Apr 17, 2018 2:01 pm

What's the most common thing you notice when rooting around Wikipedia?

Lack of anyone taking direct care of an article, in the long term. As in, someone with a working knowledge of the history/content of the article reacting to changes as they occur, as opposed to someone without that familiarity reacting quickly, or someone with it only reacting some time later, and lastly nobody reacting at all.

I'm not saying it doesn't happen, lots of editors take pride in the long term curation of a page, and sometimes it's even because they didn't write/expand it, or is about something that doesn't interest them (but you probably know which of those is least likely). But everywhere? No, not even close.

This is the root cause of many quality issues with Wikipedia, alongside the other common causes - detrimental ownership (the other side of the long term care coin), and the phenomena of someone simply not having added any content in the first place (if it isn't there, then it can't be suffering from a lack of care).

Without this long term care, then it is inevitable that articles will degrade, because Wikipedia is by design, not only vulnerable to malicious edits, but well meaning changes which will also, quite unintentionally, degrade the quality of what is already there, either in a very small way (e.g., a typo or duplicated text), or a very big way (e.g., misleading or even erroneous knowledge).

The reason this occurrs, the reason it is so easy to find, is because the maths of Wikipedia's model simply doesn't add up, and likely never did, even back in the boom times of 2007-8.

If we assume that the sort of Wikipedia editor who undertakes this crucial activity is one who makes more than 25 edits a month, then on current stats, Wikipedia has just 16,000 editors in that resource pool, the shallow end of which isn't exactly doing a lot (less than one edit per day). The current article total is 5.6 million, give or take a few Pokemon.

Doing the maths, that's a whopping 350 articles for each and every editor to be looking after. That is by no means an insurmountable workload, since if you just took a random 350 page slice of the 5.6 million pie, then it is unlikely the total number of changes per day needing to be reviewed and if necessary acted upon, would be very large. I'd guesstimate you might see 20 changes a day.

But then we get into the harsh realities of Wikipedia. First, there is the time factor. It seems unlikely many of those editors will have the required time to do their fair share of the work, day in, day out, forever. Some have lives or jobs, surely. And Wikipedia being what it is, it doesn't take much for something that looks like a trivial task, to become a time sink.

Then there's the obvious issue that no such fair share allocation exists, even if everyone was prepared to do it. Wikipedia simply isn't that organised. In this crucial regard, it is freakishly inefficient. Rather than 16,000 editors watching 350 pages each, because every editor is free to choose what they watch and what they don't, and indeed are free to watch nothing at all, it is more accurate to view the workload distribution as decidedly non-uniform, but one where effort is undoubtedly not being expended in a sensible/proportionate fashion.

Wikipedia in reality has large groups of editors watching mostly the same pages, some highly controversial pages being watched by a thousand or more (2,297 for Trump), but with a long tail of articles with fewer and fewer watchers, until you get to some quantity with no watchers at all. There is a list of those, but for obvious reasons, if you're not an Administrator, you're not allowed to see it.

https://en.wikipedia.org/wiki/Special:UnwatchedPages

A lot of time is therefore also wasted when one editor begins to look into an issue on a highly watched article, which they might understandably prioritise over a less important page, only to find another watcher has independently noticed it and was quietly working away on a fix, something you only realise when it 'edit conflicts' with you're own fix.

So quite clearly, there are large swathes of Wikipedia that editors with sufficient knowledge of its rules, policies and norms, either don't have the time or the interest to maintain on the basis that they intimately know the articles in question and are in it for the long haul.

There is a sick reality to this aspect of Wikipedia - undoubtedly the cause of many an editor burning out, often in spectacular fashion, is the stress induced by the realisation that they either have to take on an enormous workload to cope with the fact others aren't doing their fair share, or they have to take the view that their own time and well being matters more than fixing an issue in a timely manner, if at all.

For those who think some time is saved by the fact not all articles are completely open to editing by anyone, the reality is, at any one time this is only ever a tiny, tiny, fraction of the 5.6 million total. Enough to scupper Wikipedia's "anyone can edit" philosophy, without ensuring those left open do not degrade.

The reason Wikipedia took off at all, is because their editors do not have uniform interests. There is absolutely no concept of doing one's fair share as the price of entry - so you are free to concentrate on whatever you find gives you the biggest reward, until addiction properly kicks in.

So some editors busily create the content, or perform wholesale rewrites, while some speed around fixing typos and style issues, and still others guard the gates and clean house using wholesale deletion. This resource allocation profile is what produced the shit pile that you see before you. Large, undoubtedly, but with little to no uniform quality, the largest success in that regard only being on the superficial aspect of style. This merely masks issues of substance, or course.

What most readers of Wikipedia probably don't even appreciate, is that many of its editors quite happily take the pragmatic view that because these vast hinterlands are little read, and are often of little importance in the grand scheme of world knowledge, then it doesn't really matter if it takes months or even years for poor changes to be rectified. Of course, sometimes it never happens at all.

This critical lack of editors whose sole or even major interest in how they spend their wikitime is maintaining a level of quality already attained, no matter how poor, so as to ensure that overall Wikipedia is only ever improving, is what will eventually doom them.

HTD.

User avatar
Mutineer
Sucks Fan
Posts: 130
Joined: Tue Mar 21, 2017 4:11 pm

Re: The maths just doesn't add up

Post by Mutineer » Wed Apr 18, 2018 2:34 am

CrowsNest wrote:What's the most common thing you notice when rooting around Wikipedia?

Without this long term care, then it is inevitable that articles will degrade, because Wikipedia is by design, not only vulnerable to malicious edits, but well meaning changes which will also, quite unintentionally, degrade the quality of what is already there, either in a very small way (e.g., a typo or duplicated text), or a very big way (e.g., misleading or even erroneous knowledge).

The reason this occurrs, the reason it is so easy to find, is because the maths of Wikipedia's model simply doesn't add up, and likely never did, even back in the boom times of 2007-8.


I agree that articles degrade over time. I think it's like somebody did the work of researching the original content, and then somebody comes along a year or so later, doesn't bother to research, and mangles the content, because it didn't please him or her for some reason.

Giggly Drmies has an habit of just deleting large chunks of an article. I noticed him make a comment once, essentially, "we need nice prose and this stuff is a bit disjointed." So why doesn't he edit to make it more like prose? Because he expects others to react in alarm, and rush in to to repair the damage to his liking. If they do it he walks away with a big grin taking full credit. If they don't do it, the content is just lost.

Then there's political people. The ones I've noticed tend to be "conservative" (what is it they want to conserve?) that just come in and switch a word out and a word in, not reading the source and making the sentence inaccurate. In the meantime thousands of people are reading the article and believing it.
I am "Modsquad" here, and participate, but I don't want you to think we can't have an angry argument.

User avatar
CrowsNest
Sucks Maniac
Posts: 4459
Joined: Tue Feb 27, 2018 4:50 am
Been thanked: 5 times

Re: The maths just doesn't add up

Post by CrowsNest » Wed Apr 18, 2018 9:08 am

Once or twice in a blue moon I could understand, but it is insane how often I've seen an edit made that simply duplicated something already in the article. It is clear proof the editor didn't even bother to read the article they're "editing", which is surely a basic requirement of that task. But the real issue is, they're not catching it (otherwise I wouldn't find it so often!).

Drmies' carelessness is well known, he clearly isn't doing the due diligence required for the sort of hacking edits he does. He even introduces errors with his lazy edits. He doesn't care, he's already conned a tenured position off the back of it. The fact people do nothing, whether out of fear or he's just that good of a con-artist (his peers simply don't look because they trust him, the gullible fools), or they simply just don't care, is another defining quality of the encyclopedia of shit.

Post Reply