Good To See: Wikipedia Moves Forward With Color Coding Less Trustworthy Text

from the teaching-people-to-be-skeptical dept

More than two years ago, we talked about a great idea to deal with the (somewhat misleading) question of the trustworthiness of Wikipedia: color code new edits from untrustworthy editors. Not only would this alert people to at least double-check that particular info, it would remind people that Wikipedia is a constantly changing site. To be honest, I was a bit disappointed that I hadn't heard much about this idea since that summer of 2007. However, apparently, it's been gaining in popularity, and now Wikipedia is set to start using it across the site. Here's how it works:

Based on an person's past contributions, WikiTrust computes a reputation score between zero and nine. When someone makes an edit, the background behind the new text gets shaded orange depending on their reputation: the brighter the orange, the less "trust" the text has. Then when another author edits the page, they essentially vote on the new text. If they like the edit, they'll keep it, and if not, they'll revert it. Text that persists will become less orange over time, as more editors give their votes of approval.

While there are some concerns about how well this will work (and how much processing power it will take), it seems like a worthwhile experiment.

22 Comments | Leave a Comment..

One More Reason Not To Blindly Trust What A Computer Tells You

from the this-site-is-sooooooo-dangeorus dept

By now, you've probably heard the news that Google had a bit of a "glitch" this past weekend, whereby it warned people that every single site in existence (including Google) was rated as potentially dangerous and could put malware on your computer. It lasted for about an hour Saturday morning, causing amused chatter around the web. But, of course, it does highlight one key issue: whenever we end up with various "automated" warning systems, we tend to start believing what the systems tell us -- even when we know they're fallible. It's something worth remembering -- not to say that computer models are bad, just that we almost always underestimate how much weight people put on them once they're in place, no matter how much we intuitively understand that it's just a model.

23 Comments | Leave a Comment..

Trust Your Customers... And They Do Amazing Things...

from the treat-them-like-criminals,-however... dept

The entertainment industry has a long and sorry history of treating customers like criminals, despite plenty of evidence that suggests that treating customers like criminals makes them more likely to act like criminals, rather than less. SteveD writes in with an example out of the UK, where the proprietor of a small shop decided that his store should be open the day after Christmas, but he didn't want his employees to have to work -- and he didn't want to work either. So, he opened up the shop, put up a note and a box for people to put money in and left the shop entirely unstaffed. It actually worked out well. He made a fair bit of money and didn't find any damage or products stolen.

This reminds me, quite a bit, of the Freakonomics story about the "Bagel Man" who delivered bagels to a variety of office buildings around Washington DC and left out boxes for people to pay. On average, he ended up with around 90% of the money requested, and some interesting lessons in which types of people and companies were more likely to be honest. There's also a scene in the Kevin Smith movie Clerks where the lead character Dante does the same thing -- though his explanation for why it works is: "Theoretically, people see money on the counter, and no one around, they think they're being watched." And, as his girlfriend notes, this is "honesty through paranoia." I'm not sure which it is, but it seems that there's ample evidence that honest people don't need to be "kept honest" and treating your customers like criminals isn't necessarily a very good idea.

21 Comments | Leave a Comment..

Is Google Squandering Its Biggest Asset?

from the alarm-bells-should-be-going-off... dept

One of Google's biggest assets is trust. The company's founder has admitted that adopting the "Do No Evil" mantra purposely set a very high bar against which to judge every company action -- often making it impossible to take the easy way out, even when the "right" solution was more a lot complex. However, over the past few years, the company has repeatedly distanced itself from that mantra, and we've all been seeing it in certain actions the company has taken. Earlier this week, we noted that it would be a mistake for anyone to trust Google to look out for the best interests of users, but you need to admit that Google often made that explicit promise, with both its mantra and actions, that it would, in fact, look out for the best interests of users over its own short-term goals. The company's execs clearly stated that its own long term advantages would be best served by watching out for its users best interests, even if it sometimes went against short term advantage.

Yet, with actions such as Google caving on its book scanning project and other decisions such as paying newspapers to scan their headlines, Google seems to have compromised its core principles a few too many times -- and it appears that users are starting to notice. A recent survey of the top 20 most trusted companies in the US shows that Google has fallen off the list entirely from its spot at 10 on last year's list. Of course, these sorts of lists will always fluctuate, but it still should be setting off alarm bells in Mountain View.

Much of Google's success is based on that implicit trust. People stick with Google because of that trust. They use products like Gmail, desktop search, and Google Docs because of that trust. If people are starting to lose that trust, it creates tremendous opportunities for someone else to step in to put a serious dent into Google's online dominance. The guy who did the study claims that Google's fall is probably just due to "big company syndrome," where people just start trusting big companies less, but that doesn't make much sense. Plenty of the other companies on the list are quite big as well, and have no trouble staying on the list.

Even if it's just as blip, Google should be extremely concerned with such an early warning sign that one of its biggest assets is quickly losing value.

27 Comments | Leave a Comment..

Garbage In, Financial Crisis Out

from the so-much-for-the-quants dept

With everyone trying to figure out just what went wrong to cause the rather spectacular financial mess Wall Street finds itself in these days, Saul Hansell over at the NY Times wanted to find out why all the sophisticated risk management quant algorithms that Wall St. has been so big on lately failed to warn of impending doom. His answer, basically, is that people on Wall St. were lying to the algorithms, coming up with ways to purposely enter data such that the risk seemed much less than it actually was -- in order to let them keep pushing the boundary. Then, it became a situation where people start relying on the computers just because the computer says so -- even though the data is bad. This happens time and time again. Even when people know that computers make mistakes, it's just so convenient to have a computer "confirm" your thinking that you start ignoring other warning signs.

44 Comments | Leave a Comment..

How Much Does The Job Of Corporate Email Reader/Snooper Pay?

from the sort-of-like-a-spy,-but-not dept

It's no secret that plenty of big firms do track emails that are being sent by employees, to be able to look for leaks or spot questionable email behavior. And, indeed, a new study finds that 41% of the largest companies surveyed do employ people to analyze outbound emails, though that could just be looking over stats for anything suspicious. However, the report also notes that 22% of the companies employ people for whom this is their primary job -- which suggests that at least some of those are basically sitting there all day scanning and reading the email of employees, looking for anything questionable. This seems fairly extreme. While I can understand the idea of having a system to go back and spot questionable emails if an investigation requires it, having full time staff scanning emails seems to be a clear indication that these companies simply don't trust their employees. I recognize in a large corporation that you can't trust all your employees, but that doesn't seem like a good reason to spy on all of them. Do these companies also record all of their phone calls and listen to them? Or track what all the employees do when they leave the office?

23 Comments | Leave a Comment..

Debunking The Wikipedia 'Brain Surgery' Myth

from the please,-please,-please-make-it-stop dept

Back in 2004, I ended up getting into an argument with a "technology columnist" from a newspaper over just how "repugnant" Wikipedia is (his viewpoint). After a series of emails back and forth, he trotted out the "brain surgeon" question, that seems to be standard fare among Wikipedia-haters. It goes something like this: "If you needed brain surgery, would you trust someone who was trained as a brain surgeon, or someone who learned brain surgery from Wikipedia?" An alternative version of the question is "would you allow a 'crowd' of people to perform brain surgery on you."

Since then the brain surgery "proof" of Wikipedia's problem has shown up plenty of times, with the latest one being pointed out by Slashdot quoting a professor who dislikes Wikipedia so much that she fails to see the problems in what she (supposedly an "expert") is saying. As she notes: If you are faced with the prospect of having brain surgery, who would you rather it be performed by - a surgeon trained at medical school or someone who has read Wikipedia?

So let's debunk this once and for all. First off, no one would want a brain surgery based on someone who just learned how to do brain surgery from Wikipedia, but that proves absolutely nothing. No one would want brain surgery done by someone who just learned how to do brain surgery from Encyclopedia Britannica either -- but you don't see this professor freaking out and trashing Britannica, do you? Wikipedia is a tool, just like Britannica, and it's not designed to be a reference on how to do brain surgery.

The second problem with the "brain surgery" example is the suggestion that experts and the folks working on Wikipedia are somehow mutually exclusive. It's this idea that no one who actually knows anything inputs information on Wikipedia, and the only people who do contribute know nothing. That's pretty clearly been proven untrue, so it's difficult to take this complaint particularly seriously.

As for the professor in question, let's take a look at some of her other statements:

"People are unwittingly trusting the information they find on Wikipedia, yet experience has shown it can be wrong, incomplete, biased, or misleading."

This has to be one of the funniest statements she makes, because every point that she makes can be equally applied to so-called "expert" resources or publications. And, there's a pretty big difference with most of those publications and Wikipedia: with those other sources, most of them can't or won't be changed when the "wrong, incomplete, biased or misleading" info is found. That's not the case with Wikipedia. Furthermore, in a bit of pure irony, this professor doesn't seem to realize that by making all of these incorrect statements, she's showing just how little you can trust supposed "experts" in the first place. After all, she's going on and on about trusting "experts" over the masses, while showing that she doesn't even understand how Wikipedia works at all, showing her own wrong, incomplete, biased and misleading positions.

This isn't to say that Wikipedia is perfect. It's not. It's got plenty of problems. But the lesson that this professor should be teaching is that you can't trust any source by itself, and you should double-check and confirm any information you find, whether it's from Wikipedia, a supposed "professor" or anyone else. It's not brain surgery to understand such a lesson.

69 Comments | Leave a Comment..

You Have No Privacy Anywhere: Administration Pushes Forward With Domestic Spy Satellite Program

from the don't-mind-us... dept

While everyone's arguing over the legality of warrantless wiretaps, it looks like the administration and the Department of Homeland Security have moved on to even bigger fish: using spy satellites to spy more thoroughly throughout the US. For rather obvious reasons, this has many folks concerned. The government has shown a decided lack of ability to "stick to the rules" when it comes on domestic spying programs, and given what some of these spy satellites can do, it has people concerned about why the US is using such military hardware, normally reserved to spy on enemy countries, to spy on our own citizens. Somehow, DHS boss Michael Chertoff's response to his critics isn't particularly reassuring:

"There is no basis to suggest that this process is in any way insufficient to protect the privacy and civil liberties of Americans... I think we've fully addressed anybody's concerns. I think the way is now clear to stand it up and go warm on it."

To translate: "You should trust us, because we're trustworthy people who would never do anything wrong (please ignore all we've done wrong over the past few years). So, now that that's settled, let's get this baby rolling..."

36 Comments | Leave a Comment..

In Google We Trust

from the if-Google-says-so... dept

You can certainly question whether or not this study has a large enough or varied enough sample size (22 students from the same university), but the results were that the students studied appear to have something of a blind-trust in the authority of Google. That is, if Google deems a certain site to be more relevant based on a higher search engine ranking, the students often accepted that view -- even if sites that showed up lower in the results were actually much more relevant. This really isn't that surprising (or disturbing). After all, Google is often quite a good judge of relevance, and so there's value in trusting it to be right most of the time in its rankings. It helps people function faster, rather than having to verify everything carefully -- and in most cases, that's probably okay. As long as people realize the situations when they actually do need to verify things more carefully, it doesn't seem all that worrisome that kids use Google's rankings as a reasonable short cut in judging relevance.

26 Comments | Leave a Comment..