An English teacher that I had around ~20 years ago told the class that there were studies that considered the effect of a colon in the title alone and they suggested that having a colon in your title resulted in a better grade for your paper.
As a result, I generally take a second or two to write a title that works as phrases separated by a colon. I'll assume I've reaped some ever-so-mild benefit in my career. ;)
At least papers don't have subtitles (I think). When I have to record the title of a book, if there's a colon or a dash in it then I find it hard to know whether it's being used to separate the "title" from the "subtitle". Alternatively, if I know there's a subtitle then I don't know what to do when entering the data into a form that doesn't accept a subtitle. See https://en.wikipedia.org/wiki/Subtitle_(titling) for more information on this point of pedantry.
I don't think those 'considered harmful papers' are actually computer science papers though? Things like 'Go To Statement Considered Harmful' were magazine articles rather than papers.
It seemed too frivolous to include in the blog post (I'm the author), but I fit an exponential to the last 10 years for Twitter consumption, which predicts 100% colon titles as early as 2050: https://twitter.com/mm_jj_nn/status/1055555513658810368
A lot of the most recent titles featuring colons aren't the type "Proposition: Question?". Most of the papers I read with colons in the title are similar to "Project Name: Project Description"
