I wrote the very long expository comment in that article (when it came out). I really think that many industrial processes would make surprisingly good (or bad) video games – it's a novel training method at the very least, and I think Sim Refinery was way ahead of its time. There's a fine line between "fun" and "simulation" for many of these things.

You're quite right that Ars Technica appears to just package articles from HN and Reddit at times though...