News for hackers. Stuff that matters.
Re:Mathematicians should stop naming things after each other by kxyvr on Saturday September 05, 2020 @03:04PM (#24386053)
As a counterpoint, I've a Ph.D in math and I largely agree with the article. Frankly, it's difficult enough to keep track of the names in my own field much less interact with people in other fields who would greatly benefit from these results if they were comprehensible.

While I agree that precision is necessary, is it really that hard to say something like a complete inner product space vs a Hilbert space? Should we have an argument about whether we should call the triangle inequality the Cauchy-Schwarz or the Kantorovich inequality? How many years did it take to recognize Karush for providing algebraic optimality conditions long before Kuhn and Tucker?

Anyway, outside of general griping about names, very specifically, I have to regularly interact with technical people in other fields and memorizing names is one additional, often superfluous, barrier to their understanding. Even if it's important for someone credit the (supposed) originator outside of a citation, giving an intuitive name for the context would go a long way in improving understanding. For example, the triangle inequality from Cauchy and Schwarz or the algebraic optimality conditions from Karush, Kuhn, and Tucker.

By the way, I'll also mention that the majority of other mathematicians that I know don't feel this way and prefer using names after people. I understand; I just disagree.

Re:Mathematicians should stop naming things after each other by Davidzheng on Saturday September 05, 2020 @04:20PM (#24386569)
If you have to write complete inner product space a hundred times in an article that's pretty annoying tbh
Re:Mathematicians should stop naming things after each other by kergonath on Saturday September 05, 2020 @05:12PM (#24386878)
In a Physics article, we'd just call it CIPS after the first instance. It does get a bit unwieldy if there are too many acronyms or initialisms in a sentence.
Re:Mathematicians should stop naming things after each other by xigency on Saturday September 05, 2020 @05:28PM (#24386975)
Terrible excuse. Things are read much more than they are written. That’s like using short variable names rather than descriptive ones as they are easier to type. With current IDEs that isn’t even the case and I’m sure word processors could use the same autocomplete. Every writing shortcut adds cognitive load for the reader to remember.
Re:Mathematicians should stop naming things after each other by ben0x539 on Saturday September 05, 2020 @05:49PM (#24387119)
Short variable names can be easier to read, too. Having people decode meaning from the name, rather than keeping track of the meaning separately and associating it with a symbolic name, doesn't always feel like a good use of cognitive resources.

imo, "Hilbert space" is a lot easier to read, you really only have to recognize the guy's name, it's a fairly easily recognizable word. "Complete inner product space" has me look more closely at more words, and it's all extremely boring-looking words so they don't even stand out in a block of text.

Re:Mathematicians should stop naming things after each other by CopOnTheRun on Saturday September 05, 2020 @06:34PM (#24387387)
I mean long class names in Java in particular are pretty much a meme [1] because of how ridiculous they are. There is definitely a balance to be struck between succinctness and expressiveness.

[1] https://news.ycombinator.com/item?id=3215736

Re:Mathematicians should stop naming things after each other by srtjstjsj on Saturday September 05, 2020 @07:01PM (#24387550)
If only math used completely intuitive names for concepts everywhere, like class, set, group, ring, field, ideal, domain (not the same as "domain"!), fiber, sheaf, manifold, oh wait.