Wikipedia has a lot of problems, including some subtle but serious ones that seem difficult to fix without radical changes to how the platform operates.

I’ve watched the growth of Wikipedia since my first edit, which was in 2004. Since then, I’ve accrued 16,044 edits. 10,287 of those were in 2007, when I was very active as an anti-vandalism patroller. Over the years I’ve created 68 pages in total. Wikipedia has always had obvious problems such as vandalism, systemic bias, and link dropping, which are being addressed by a variety of concerted efforts.  Lately I’ve been noticing a lot of more subtle problems with Wikipedia articles, which has caused me to seek out higher quality sources of information. To put it bluntly, Wikipedia articles are just not very well written. They lack logical progression and consistency in their style and level of technical depth. Of course, it’s difficult for the Wikipedia platform to achieve either, since many different authors are constantly adding and subtracting sentences from every article. 

Here are some specific issues I’ve noticed:

Growing lists

One problem I’ve noticed is growing lists. Consider this sentence from the lead of the article on Arkansas:

“Arkansas’s enduring image has earned the state “a special place in the American consciousness”,[11] people such as politician and educational advocate William Fulbright, former President Bill Clinton, former NATO Supreme Allied Commander General Wesley Clark, Walmart magnate Sam Walton,[12] singer-songwriters Johnny Cash and Glen Campbell, the poet C.D. Wright, and physicist William L. McMillan, who was a pioneer in superconductor research; have lived in Arkansans.”

Note how cumbersomely long this sentence is. This sentence is likely to only get longer in the future, because its is easy for anyone to add their favorite Arkansan to the list. Note that the grammar error in the last work of the sentence (Arkansans instead of “Arkansas”) clearly gives away that the sentence was tampered with at least once through its lifetime. 

Something similar can be seen in the lead for the article on Sam Houston:

“Other things named for Sam Houston include: a memorial museum, five U.S. naval vessels named USS Houston (AK-1, CA-30, CL-81, SSBN-609, andSSN-713), a U.S. Army base, a national forest, a historical park, a university, an elementary school in Lebanon, TN (Sam Houston Elementary) and a prominent roadside statue outside of Huntsville.”

In my view, this information (which is all unimportant trivia) does not belong in the lead to the article. The specific names of the naval vessels should be footnote at best, and do we really need to mention that an elementary school named after Sam Houston is called Sam Houston Elementary, and that a roadside statue of Sam Houston is named after Sam Houston?

Trivia dropping

Looking further in the Sam Houston article reveals an example of another problem:

As it filled with Lutheran German immigrants, Houston decided to migrate south with other Scots-Irish, who settled in the backcountry of lands in the Shenandoah Valley of Virginia.[9] A historic plaque in Townland tells the story of the Houston family. It is located in Ballyboley Forest Park near the site of the original John Houston estate. It is dedicated to “One whose roots lay in these hills whose ancestor John Houston emigrated from this area.”[citation needed]

This is an example of a trivia dropping. The problem is that people have many different views on what is important. To whoever placed it in the article, the “historic plaque in Townland” must have been highly salient and significant in the story of Sam Houston. That is fine, but just because something is significant to one person’s understanding of a subject doesn’t mean it is important more generally. It takes a subject matter expert to be able to properly determine what details are important enough for inclusion in an encyclopedia article. For many subjects on Wikipedia, entire books have been written (biographies, histories, monographs). The challenge of writing an encyclopedia articles is to take all that information and extract the bits that are likely to be the most relevant to the average reader, and condense that information into a few pages. More often though, Wikipedia articles appear to be random collections of facts which, while potentially important to some readers are not important for most readers.

Wild fluctuations in technical level

This problem doesn’t seem to be as common, but I’ve noticed it several times over the years, especially in math and physics articles. Often, I’ll be reading an article and then all of a sudden there will be a very technical statement with a lot of unexplained jargon that was probably written by someone with a PhD in the field. As a random example, this statement is found in the lead of the Boole’s inequality article:

“In measure-theoretic terms, Boole’s inequality follows from the fact that a measure (and certainly any probability measure) is σ-sub-additive.”

Less commonly, one sometimes finds extremely obvious statements. For instance, the second sentence in the article on humidity is “Water vapor is the gaseous state of water and is invisible.”, for which a reference to a random website is given (clearly link dropping).

Do people really care?

Despite these problems, Wikipedia is still very useful and will remain so. It seems like the primary way that people use Wikipedia as a quick way to look up facts – things like the population of Shanghai, whether a tomato is a fruit or a vegetable, or how old Hillary Clinton is. Wikipedia was never meant for deep dives into a subject, for that books will always be the best. Wikipedia isn’t the best option for a serious review of a subject, because (in my opinion) the quality of information is not nearly as good as what can be found review articles published in peer reviewed journals (if you have access) or, for particular case of philosophy, the Stanford Encyclopedia of Philosophy.  There are also other projects like ScholarPedia  that offer higher quality articles on many subjects. Still, for many subjects Wikipedia remains the best option. For many people though, the presence of so many intra-wiki links makes Wikipedia not very conducive for long term reading, leading to what XKCD calls “The Problem With Wikipedia”.  [Personally, I have found that I can not do long form reading within in a browser at all. Fortunately, I recently discovered that it easy to send webpages to the Kindle using the “Send to Kindle” extension.]