Applying metaphors from other fields into software development
Posted by jhartikainen@reddit | programming | View on Reddit | 15 comments
Posted by jhartikainen@reddit | programming | View on Reddit | 15 comments
FelisLwipe@reddit
Adding to the overall discussion, I propose you replace the Broken Windows Theory with the Broken Window Parable (which just came to mind reading the article and comments). It's used in Economics to point out how more economic activity is not always good, e.g. if you break a window and hire someone to fix it, GDP may go up, but it would still have been better if you hadn't broken the window in the first place. I'm sure you could tie that to over-emphasis on benchmarks or something. And it gives you another reason to exclude the Broken Windows Theory because two wildly different metaphors with the same name would just be confusing
jhartikainen@reddit (OP)
Never heard of that one before, quite interesting :)
After reading about it on wikipedia, it does sound like metrics or benchmarks could be related here. For example, simply following "tasks resolved" is not necessarily an indicator of value being created, since they could very well just be "windows that you broke yourself"
No-Consequence-1863@reddit
Ok so first Im not sure the author understands Entropy, also like it already is used for generic systems all over the place as a drop in for complexity.
Second, I would suggest using “Broken Windows Theory” as its like a total bull shit theory from the 90s to justify harsh sentences on kids tagging buildings. Its been shown to be entirely ineffective and racistly enforced.
Lastly the other two metaphors were literally the same as “lots of small good decisions make big good results”. Which is like so general you really dont need to “apply it” to the software. It just kinda does.
I don’t understand the premise of this piece. Like I thought we were gonna see some really niche metaphors we were gonna stretch, but these were all so basic applying takes one step.
jhartikainen@reddit (OP)
Thanks for the feedback. I'll have to do another take on this if I find some better metaphors.
Regarding the broken windows theory - I did acknowledge in the post that it has been criticized, and rightly so, as you point out. But does the critique devalue its effectiveness as a metaphor? Or would you say it's not appropriate to use it as a metaphor because of its racist origins?
No-Consequence-1863@reddit
Yes Id say its ineffectiveness directly cuts core to the metaphor. It wasnt true that policing minor infractions to reduce broke windows reduced crime, so trying to apply it elsewhere is a total fallacy.
There may be “broken windows” in a code base but that tells us little. “Broken Windows Theory” mindset is so bad it already has a metaphor in programming and its called “Bikeshedding”
recycled_ideas@reddit
I think the problem here is that we have something that's probably true that's been extended to something that's not true.
The core of broken window theory is probably true. That if things are shitty that people will care less about them. Anecdotally I think we see that all the time, if you scratch your perfect pristine car it's a big deal, if it's already got a thousand scratches you probably don't care much and if the bumper is held on with duck tape it's even moreso.
The mistake is taking that principle and saying, "therefor if we punish people who tag we'll reduce crime" as opposed to "if ensure people aren't living in rat infested hovels and have enough to eat we'll reduce crime".
I don't think it's remotely a stretch to say that the overall state of something affects how people treat it, but some dipshit teenagers tagging the walls was never the underlying problem.
AreWeNotDoinPhrasing@reddit
It’s just like how people treat rentals like shit. They assume everyone else does it. This sorta applies to a whole lot of things.
jhartikainen@reddit (OP)
Interesting, I guess you could look at it from that perspective also. I was thinking of it more literally from the metaphorical perspective of a decaying building, rather than from the perspective of how effective the theory itself was.
Popular-Awareness262@reddit
garden metaphor stuck with me. been pruning my agent skills the same way and skillsgate https://github.com/skillsgate/skillsgate handles the cross-agent sync
qwertydiy@reddit
For some reason as programmers we are unusually bad at naming products and concepts as well as variables (what we are usually famous for)
kaiken1987@reddit
You'd think people would name dozens of things every day would be better at it. Or maybe that's the reason? Can I name my product with Hungarian Notation?
psychometrixo@reddit
szYouCanDoAnything
jhartikainen@reddit (OP)
The paper linked in the post about metaphors and learning discussed "Aha moments" vs analytical thinking. I kinda felt like I don't have many aha moments, and instead have to analyze something for that kind of understanding.
Perhaps there might be some connection between this and naming things, since the paper suggests metaphors created through analysis were less apt than the aha-moment based ones.
SheriffRoscoe@reddit
"I haven't read the book".
Obviously.
jhartikainen@reddit (OP)
If I had, I would probably be an important high powered MBA instead of some guy thinking of metaphors and programming.