(not clickbait)

Many of my longer blog posts take a very, very long time to write, as in both the amount of time spent working on it and also the amount of time between when I start the first draft and when I hit publish. One thing that delays me significantly is doing research and procrastinating on doing research.

Since this is actively hindering my progress on a couple drafts I have going, I thought, “What can I write about that requires no research?” I figure I can write about my own writing.

I generally try to fact check and source my claims. Any post that does not link to sources should be taken with a grain of salt; I’m just sharing my perspective in those cases. When a post does have a list of references at the end, that means I did attempt to source every important factual claim. That’s not a guarantee of accuracy, though.

Why include references?

Factual claims, especially big or controversial ones, shouldn’t be accepted without evidence. Because providing no evidence at all is such a clear problem, that often becomes the focus.

“Citation needed.”

People make unevidenced claims all the time, so you kind of get used to this. It’s tempting, I think, to fall into thinking that because not providing sources is the problem, that means providing sources is the solution. A long list of citations can look impressive, but on its own it doesn’t say much.

Providing references to information sources is the bare minimum. It’s up to the reader, if they care, to determine whether each source is reliable and says what the author is claiming it says. This can often be very difficult.

For example, a science communicator on YouTube might reference several scientific papers in a YouTube video, but the average viewer has neither the access nor the expertise to read these papers and assess whether the video is accurate. In such situations, it benefits the public for creators to fact check one another. This is similar to the principle underlying peer review.

My area of formal expertise is math and math education. Suppose I read a math paper and say in a blog post that theorem 2 is saying such-and-such, with a link back to the paper. Most of my readers can’t understand post-graduate level math, and few would be interested in fact-checking me. Most people will just take my word for it.

A personally embarrassing story

Living in a college town, there are actually many people in my community who have far, far more expertise in math than I do (like professional mathematicians). I once posted a link to a blog post on NextDoor (a geographically local social media network), and to my surprise and dismay an astrophysicist showed up to fact check me.

I felt extremely embarrassed, and I’ve thought a lot about my error since then. The details of the math I got wrong aren’t important. One conclusion I came to is that I’m certain the professor didn’t read the blog post, and was only responding to the blurb I wrote for NextDoor.

While I did make a mistake in the post itself, it was inconsequential to the point I was making in the post, and could be corrected by just adding a few words. On the other hand, what I wrote for NextDoor was a quick and sloppy attempt at a “hook” (clickbait?) which overstated the error I was making. The fact check was in direct response to something I wrote there.

I also came to the related conclusion that the professor didn’t know what the point of the blog post was or what I was saying at all. I imagine that they were scrolling NextDoor, noticed the obvious error in my blurb, left a comment correcting the error, and scrolled on.

So, while technically helpful, their response lacked any context from the post and I had to interpret what it meant for the post itself. The blurb magnified the error because I was trying to sensationalize, which made what the blurb was saying outright false. That was careless on my part. However, in classic clickbait fashion, what the blurb was saying was not what the post was saying. The problem with the post itself was that I omitted one important piece of information, as well as a slightly misleading statement in the introduction (for similar reasons that the blurb was incorrect).

With the inclusion of the relevant information, the other statement was clarified so as to no longer be misleading.

As an aside, I quickly gave up on NextDoor because of the extreme toxicity. You would think that interacting with people in the same town might cause people to act nicer online, but apparently not. This doesn’t have anything to do with the interaction I discussed above, but rather things like people being racist or transphobic and insulting one another.

What references aren’t

Even when people use sources and are not trying to be misleading, the information might be wrong. It’s very easy to misunderstand what a source is saying. I’m highly subject to this as someone who frequently writes outside my area of expertise. I am often sharing my understanding of something I only learned myself by doing research for that post, and that is highly error-prone.

And, it’s been demonstrated that I make errors within my area of expertise as well. Or maybe I made it all up.

More generally, sources themselves can be wrong or only partially correct. The overall reliability of a source is never a guarantee of its accuracy. Anything you read online should be accepted only provisionally (if at all).

Most information is hardly reliable at all. Something I’ve thought about recently is that criticism of LLMs as providing unreliable information massively oversells how reliable human information sources are.

Something I believe would benefit society is if everyone felt more comfortable with not knowing. There are many pieces of information I’ve heard that are inconsequential to my life; trivia, you might say. It’s not important to me to have a high degree of certainty as to whether each one is true or false. Certainty for certainty’s sake is dogmatic folly.

So inspect the sources, try your best to evaluate the claims, but also maybe just don’t worry about it too much. It’s okay to talk about things that simply might be true. It’s not necessary to commit ourselves to the truth or falsity of every proposition under the sun.

---

Photo by me