I think there is a huge difference between what the author mentions as "shallow" side projects (where you are just trying to mess around with some new tech), and side projects where the aim is to develop a legit product. I imagine the latter, when seen all the way through, is much more attractive to recruiters. Although, I don't have any solid anecdotal experience with recruiting to validate that claim.

I just think it's important to not classify all projects outside of your job as "shallow" side projects, and then completely discard them career wise like the article seems to be doing.