User:Kent G. Budge/Response to College Teaching critique

The Oxford English Dictionary tells us that the original meaning of encylopedia (ἐγκύκλιος παιδεία) is ‘encyclical education’, meaning "the circle of arts and sciences considered by the Greeks as essential to a liberal education." The first definition that follows this is "The circle of learning; a general course of instruction." If a credible source judges the bulk of Wikipedia unsuitable for opportunistic learning by college undergraduates, then we have a serious problem.

What is Wikipedia? I've long had a short blurb on this at my user page:

• It is an encyclopedia, from which a reader unfamiliar with a topic can get a decent introduction.
• It is a collection of information nodes (via Wikilinks) that allow a really curious reader to traverse a whole set of related articles in a subject area he finds fascinating.
• It is the human-curated index to the wealth of human knowledge available on the Internet, complementing non-curated robotic search engines like Google or DuckDuckGo.

The third is a notion I find interesting, but I freely admit it may be at odds with many other editors' notions of what Wikipedia is. But I think the first two are fundamental and all but indisputable. We must also understand what Wikipedia is not: It is not a handbook. By that I mean that it is not a reference for experts in a field.

By education, I am an astronomer (Ph.D., Caltech, 1990), but most of my career has been in computational physics. I choose to do almost all my editing in earth sciences instead. The process of editing is an excellent way for me to learn more about a field I've come to find fascinating, and I am in no danger of inappropriately promoting my own research (I have done next to none in geology). I would like to believe that, as an outsider to the field, I may have a better feel for what needs to be explained to an intelligent general reader, but I've studied it enough that that may no longer be true.

I've edited Wikipedia for a long time, but took a long break and only resumed editing about four years ago. It was only then that I began focusing on geology. In the criticisms that follow, I offer a hearty mea culpa; I am guilty of pretty much every offense I'm going to outline here.

What I take away from the College Teaching study is that our STEM articles are too often inaccurate, and even when accurate, are too often badly written. They don't get the facts straight and they don't tell a good story.

Accuracy

I have been systematically working through the most-visited articles in geology. A large fraction of second-tier articles are astonishingly badly referenced. That's a serious blow for accuracy right there. I am increasingly inclined to summarily delete any statement in a STEM article that lacks a source and for which I can't find a source reasonably quickly from either my home library or Google Scholar, rather than put a {{cn}} template on it. I have not yet had the chutzpah to delete the entirety of a badly referenced article, per WP:Nuke it from orbit, but perhaps that should happen more often than it does. Putting it another way: I am becoming ruthless, but am I becoming ruthless enough?

Most of what I find in STEM articles is mostly right. When I dig out the references to check a statement, it is rarely flat out wrong; more often, it is slightly off. The problem is that "mostly right" isn't nearly good enough. There's an old analogy: Do you know what you get when you add a teaspoon of wine to a barrel of sewage? A barrel of sewage. Do you know that you get when you add a teaspoon of sewage to a barrel of wine? A barrel of sewage. We have to be ruthless about poorly sourced material.

We have a problem with accuracy in spite of nominally strong requirements on sourcing; in spite of an emphasis on accuracy in Good Article and Featured Article reviews (which, in any case, are a tiny fraction of our article base); and in spite of peer review. We have an even more serious problem with due weight. Here I feel at a particular disadvantage as an outsider editing articles in a STEM field. I can verify that a statement is supported by multiple independent reliable sources. That does not always translate to the statement being an accurate statement of the majority view on the subject, or being an important statement about the subject in question.

Readability

Accuracy is a problem. For our STEM articles, bad writing is a more serious problem.

Every now and then I come across an article that is astonishingly well-written but oddly lacking in inline citations. The language is elegant; the prose dances and sings. And the information is over a century out of date, because the article is a copypaste of the public domain 1911 Encyclopedia Britannica. Wikipedia advertises itself as an encyclopedia anyone can edit; but hardly anyone can write the way the authors of the 1911 Encyclopedia Britannica wrote. More's the pity.

And every now and then I find an article, or a section of an article, that may not rise to the level of the 1911 Encyclopedia Britannica, but is at least good solid prose. The lack of inline citations is a red flag: It often turns out to be either a copypaste of a U.S. Government public domain source, or it's a flagrant copyright violation. Any time I find well-written prose lacking citations, I feed it to Earwig's copyvio detector to check.

So what are the reasons why Wikipedia STEM article prose is (at best) pedestrian or (at worst) inpenetrable? Some thoughts, in no particular order, and with the usual mea culpa:

• Writing a good STEM article is inherently really hard. Any good writer will tell you that writing a good long piece is actually easier than writing a good short piece. I've been doing technical writing of one kind or another for forty years now, and based on feedback, I'm not bad at it. But I struggle to master the skill of writing a good Wikipedia STEM article lead. The article main body that follows is, in many ways, easier; but it's still really hard.

• We don't know our audience. Who is Wikipedia written for? I think part of the difficulty is that different articles are written for different audiences. The examples I give on my user page are Volcano and Alkaline magma series. The Volcano article has gotten 65,362 hits in the last thirty days, making it one of the most visited geology articles. It's a very general topic. The lead of this article ought to be comprehensible, interesting, and informative to an intelligent high school student. The body of the article should likewise be comprehensible, interesting, and informative to an intelligent undergraduate. And an intelligent high school student should be able to skim the lead paragraphs of each section and pick up the main points. In the language of the College Teaching study, it should be suitable for opportunistic learning.

On the other hand, the Alkaline magma series article has gotten 437 hits in the last thirty days. It is not a heavily visited article. The term is not something someone would stumble across. We cannot and should not expect its lead to bear the weight of explanation necessary to make it comprehensible, interesting, and informative to an average high school student. Its target audience is an undergraduate in a STEM field.

• We don't respect our audience. Too many articles are written by self-styled experts to impress other experts (Mea maxima culpa). We don't extend our audience the courtesy of taking the time and verbage to unfold the story in an interesting and comprehensible way. As I mentioned earlier, I have a Ph.D. in astronomy and minored in mathematics as an undergraduate; but I find even the lead of a lot of mathematics articles impenetrable. (Some mathematics articles have improved markedly in the last few years -- I'll give credit where it's due.)

• We forget that all good writing tells a story. This is true even of an encyclopedia article in a STEM field. In my experience, it is true even of a scientific paper in a specialist journal. A STEM article that is a spew of facts, without adequate exposition, is of no interest to anyone but an expert looking for a particular fact -- and probably not even to them; Wikipedia is not a handbook.

• We allow precision to get in the way of good exposition. Consider the lead sentence of Volcano as it exists this morning, December 29, 2021: "A volcano is a rupture in the crust of a planetary-mass object, such as Earth, that allows hot lava, volcanic ash, and gases to escape from a magma chamber below the surface." That's accurate and very precise. It's also quite a mouthful for a 16-year-old who wants the opportunity to learn about volcanoes. And, no, I don't immediately know how to improve it; it's going to require a lot of thought. (And then my change will probably be reverted by another editor for lacking precision.)

Precision is not the same as accuracy, as any college freshman should have been taught in his introductory STEM classes. But it takes a lot of skill to write a comprehensible sentence that is accurate without being burdened with undue precision.

• We don't write hierarchically. We understand (or should understand) that a good lead summarizes the main points of an article. But the first paragraph of a section of an article should likewise be a summary of the main points that will be made in that section. Exposition then follows. In the old advice for speakers: Tell them what you going to tell them; tell them; then tell them what you've told them. In writing, where the reader can easily go back to the lead paragraph, we may omit the final summary; but we must have the lead summary.

How do we improve?

<exercise left to the student>