It seems like this is merely someone's opinion that Factorio may have some transferable skills, namely a mindset-- along with the limitations-- of automation.
I'm not sure why this would be so frustrating for you. It's not like the author themselves went down your slippery slope to full blown theory. If they had? Sure, that would be a bit of the usually breathless attempts at "thoughtleading", but I see no need for preemptive criticism.
Also congratulations for being really tall. I'm about average height.
No it frustrates me when something so obvious needs to be announced 50 million times as if it's some revelatory discovery.
Here a list of games that are unintentionally turing complete:
Dwarf Fortress
OpenTTD
Terraria
Minecraft
Minesweeper
LittleBigPlanet
Baba is You
Factorio
Cities: Skylines
Opus Magnum
Portal 2
Geometry Dash
You can come up with approximately 10 analogy blog posts for each game... written, of course, from a different blog with each post presenting the analogy as if it's some amazing idea. Then of course post it on hackernews and flood the front page with this stuff.
Well, the world is full of people that have never played Factorio. Some of them are programmers, and when they finally play it they get excited about it and want to tell people about it. Independent rediscovery.
And I'll admit to bring irked sometimes when a recurring theme or repost rears its head on HN pg1 for the dozenth time, so I guess I understand your sentiment. But I console myself with the knowledge that what is old and cliche for me is likely to be eternally new for an ever changing subset of readers.
Also thanks for the list of games-- I had no idea some of them were Turing complete, and will enjoy a bit of dugging internet rabbit holes finding out the details of why & how.
I've played Factorio since 0.13 and am a professional programmer.
It bothers me that people see an equivalence between Factorio and Programming. There's some degrees of programming-like problems, such as spanning trees, graph theory, tree-theory, etc. etc. But Factorio is grossly simplified to comp. sci problems in general.
In particular, the Factorio build tree starts at (Iron/Copper/Stone/Coal/Oil) and ends with (Red/Green/Blue/Purple/Yellow/White science).
Most problems are solved with backpressure. The one exception is oil production (Crude-oil -> Heavy/Light/Gas), which can bottleneck other productions. (Ex: if Heavy oil is deadlocked, light-oil is also blocked). But with an "infinite sink" of science (blue/yellow science uses a lot of Petroleum gas), this problem is solved by simply playing the game more.
------------
Real world programming problems are far more complex. Graphs in fact loop more often, simple trees are rare. Data-organization is far more complex.
In contrast, Factorio has a large number of "valid solutions" because its a simpler problem. Keeping things "simply" a tree with only one multi-output (oil production) really is simple in the great scheme of things.
That said, it is more complex than most games. OpenTTD is more complex when it comes to transportation patterns (path signals, entry, exit, and combo signals, block signals. Tunnels, bridges, etc. etc.). I would recommend Factorio players to try OpenTTD, and see if they can create higher-bandwidth OpenTTD routes than Factorio's simplified train system.
------
I know some Comp. Sci. I'm going to make my n-way balancers off of Benes Networks and CLOS Network designs. But I look at "circuits" and see "vector programming system" and not actually a wire-like / circuit system. The things that are "like programming" aren't very clear IMO. Its somewhat rare for me to actually need to use combinators / circuits in Factorio, because the default actions of all entities "just works" most of the time.
Wube / the original designers are clearly good programmers. The system / game they have made "feels like the fun part" of programming, where you connect mostly-completed systems together in a chain that almost always works. That's because the difficult part of design / architecture has already been done for us by Wube.
Actual programmers designing an API for widespread use come across the same problem. You wanna create an interface that's easy to understand and fun to work with. Wube is clearly good at this. But that makes us at best, the "script kiddies" who simply are laying out Wube's designs, more so than actual "programmers" who express ourselves.
IMO anyway. This isn't a bad thing. The fact that its fun despite all this is a huge degree of success from Wube.
I'm not sure why this would be so frustrating for you. It's not like the author themselves went down your slippery slope to full blown theory. If they had? Sure, that would be a bit of the usually breathless attempts at "thoughtleading", but I see no need for preemptive criticism.
Also congratulations for being really tall. I'm about average height.