4 ms·
Every once in a while on the HN front page you get some article about a programming analogy or an entrepreneurship analogy. "Here's why the game factorio is li
by deltaonefour 5y ago
Every once in a while on the HN front page you get some article about a programming analogy or an entrepreneurship analogy.
"Here's why the game factorio is like programming or here's why it's like life. yada yada yada"
Yeah, AND this guy takes the next step and turns the analogy into a "mindset."
Great. Revolutionary. First off.. analogy is amazing. Second... turning that analogy into a way of life blows my mind. Is your mind blown too? This guy is a genius.
The next step, (and I've seen plenty of this on HN), is to turn that analogy into a full blown "theory" of some sort... complete with diagrams.
- dentemple 5y agoI mean, out of most of the analogy examples posted to HN, Factorio is actually a good one. The observations that the writer makes are all immediately apparent when you fire up the game, and there isn't anything here that I'd call a "deep cut" or a strained metaphor. Factorio can literally be renamed "Iterative Development: The Game" and players wouldn't find anything confusing about any of it.
- deltaonefour 5y ago>I mean, out of most of the analogy examples posted to HN, Factorio is actually a good one I think all the analogies I've seen are good. The problem is they're obvious. They don't really bring anything new to the table. Great you identified an analogy. Big deal?
- dentemple 5y agoAnd yet, there's always at least one HN reader who makes a comment in every popular thread that the original post is irrelevant in some way. Still, it never seems to stop people from doing it despite not "bringing anything new to the table" with their comment. So why do it?
- deltaonefour 5y agoWhat I'm bringing to the table, is helping everyone notice that these posts bring nothing new to the table. That's the problem with these analogy posts. People like them but they don't offer anything of substance. Think about it. You identified a connection between two things, yet nothing new was introduced. You get this artificial euphoria of seeing a connection but that's it, it's just an illusion.
- ineedasername 5y agoIt 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.
- deltaonefour 5y agoI'm frustrated because these types of opinions are everywhere. Especially factorio: https://www.google.com/search?q=factorio+is+like+programming https://www.google.com/search?q=factorio+is+like+programming There are tons and tons of these.
- ineedasername 5y agoI program, and I've played Factorio. It feels kind of like programming, so it does not frustrate me that such an opinion can be found everywhere. Is it your opinion that Factorio does not have elements of gameplay that require (or benefit from) an approach to what is used in programming?
- dragontamer 5y agoI'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.