Why don't we site Wikipedia

Wikipedia, weather you will admit it or not, it is probably the first place you go when you want to learn something about a new topic you are unfamiliar with. It is hands down the most useful site on the Internet for people looking for knowledge. And as the 6^th most popular website in existence, it is clearly everyone’s go to site for learning. So why is there still a taboo about siting Wikipedia as a source of information?

In short, I think the reason is that people believe that, given Wikipedia’s nature, some of the information on the site may be wrong. Since the site is technically not peer-reviewed, mistakes may carelessly slip through the cracks and be portrayed as truth. This clearly is possible, but I think, as usual, human perception is holding us back. Wikipedia, from my own anecdotal standpoint, is significantly more trustworthy then peer-reviewed journals.

I read a fair number of journal articles (about 350 this year so far) and I run into a fair number of published errors. Most of the errors you run into are simple mislabeled graphs or incorrect unit labels (here are two examples of reviewed published works with glaring errors, note the author on example two). The reason you run into these errors more often is not necessarily because they are the most common, but because they are the easiest to spot.

On the far more sinister side, many papers include incorrect values that can only be uncovered when you perform your own testing. These values are either erroneously recorded or measured, but appear accurate until compared against tested values. Sometimes you can spot these types of values by tracing the sources back to their origin. Here, Kenisarin gives us a great database of high temperature PCM properties. For the industrious CSP enginer, this paper looks like a gold mine. But when you dig in you see that more than 10% of the values reported in the paper are taken from 80s era Soviet patents. Lets put the fact that patents are not reviewed for technical accuracy aside, who here thinks that the 80s was a great time to be doing thermal testing in the Soviet Union?

Now, for comparison, lets look at the Wikipedia pages. First we can go to the thermal energy storage page. Drilling down, we could learn about PCM thermal energy storage. Finally, we could look at overview pages on certain salts or data pages to get specific enthalpy of fusion and specific heat values.

The most remarkable thing about Wikipedia is not the breadth of information, but the extent to which it has been publicly discussed. Each of those pages listed above has a talk page where the merits of each page are discussed:

The perception is that Wikipedia is a place where one-off edits get written in stone and that peer reviewed articles emerge from a thorough process based around limiting errors. The reality is just the opposite: Wikipedia achieves a level of peer review higher than most journals while most journals have such a laissez-faire review process that almost no errors are caught.