If you read my post, you'll see that I'm not advocating for that. I'm suggesting the bare minimum: "Start a blog. Post an interesting technical article to it once or twice a year—something you’ve learned, or a bug you’ve fixed, or a problem you’ve solved. After a few years stop bothering entirely, but leave the blog online somewhere."
The point of my piece is that you don't need to dedicate huge amounts of effort to doing this kind of work. Do a little bit, stick it online somewhere and it will still give you a huge benefit compared to other candidates who don't have any public evidence of their work.