4 ms·
> Programs do come with man pages! And tutorials, interactive tours, contextual help, and more. Emacs comes with 3 books, and a tutorial. (GNU) libc has a book
by Jensson 5y ago
> Programs do come with man pages! And tutorials, interactive tours, contextual help, and more. Emacs comes with 3 books, and a tutorial. (GNU) libc has a book to it. Firefox has a whole portal (MDN) as its documentation. Visual Studio comes with MSDN and a huge amount of explanatory material. And when it comes down to code, you have auto-completion, go to definition, search for callers; you can hover over a symbol and you get a popup with documentation and types. I just can't see how you can even put math papers and programs on the same scale in terms of accessibility!
You are comparing big teams and products to a single guy writing a paper intended for a niche audience and to be read maybe a few hundred times if he is lucky. People makes mistakes and sometimes forget to document everything, they try to document everything though as can be seen in their papers where most things are documented well, but sometimes they miss things and unlike code you don't have compiler warnings telling you about it. And given how few people read those papers it isn't worth investing in a team to go through and update all of those papers to properly add definitions for everything they missed.
The equivalent to those programs in math would be high school textbooks, and they are extremely well documented and easy to read in most cases.
- klibertp 5y ago> it isn't worth investing in a team to go through and update all of those papers to properly add definitions for everything they missed. Thank you. There's nothing else left to discuss.
- Jensson 5y agoThanks for understanding, math is a small field without money for things like this, there is no way anyone should expect those niche papers to be as well documented as big programming projects used by millions. If you still think that is a problem then start some open source organization to fix that. Nobody has done that yet though since so few people care about math papers, but since you feel so strongly about this you could do it, someone has to be the one to start it.
- klibertp 5y agoNo, I mean, well, it's very understandable when you describe it that way. Actually, I think your post here changed my perception of the problem the most out of all discussions I had on the subject. It made me think about people who are behind the papers. I somehow missed it. Thank you. (And, sorry for being a jerk in this thread. I said too much in a few places, exactly because I didn't think of innocent mathematicians who might read it. I'm still convinced that there is a lot that math can borrow from CS and SE, but I'm definitely going to argue this differently.)
- Jensson 5y agoI wrote one math paper before I went into programming. It is a lot of work, like code reviewing but much much longer. It isn't fun. A big reason I got into programming is because that process is so much work. Of course I, the professor who reviewed it and the professors who looked at it afterwards understood it, but I can't guarantee that someone who hasn't read a lot about research level topology or combinatorics will easily understand much at all. However I doubt that anyone who didn't do those things will ever read it since it is an uninteresting niche topic. I'd be surprised if even 10 people read it fully.
- klibertp 5y agoYeah, I didn't think about it at all - I didn't realize that what I'm saying is basically demanding people to work for free (and on things that won't be useful to anyone in 99% of cases), and that's on top of already huge effort that is writing the paper in the first place. Honestly, I was behaving like people who open tickets in an open source project just to demand that someone implements a particular feature, just for them, and right now. I dislike such behavior, and realizing that I'm doing the same hit me hard :)
- Jensson 5y agoAnyway, lots of mathematicians works really hard to make everything as understandable as possible. Learning programming isn't the same thing as learning math. In programming you learn an instruction set, and then you use those instructions to compose programs, therefore learning that instruction set is relatively easy. Mathematics isn't like that at all, instead you just continue to learn ever more complex instruction sets, you don't compose much at all in mathematics because the new instructions typically can't be expressed using the old ones. Therefore learning the instruction set in math is equivalent to learning math, it isn't something you just do once and then is done with it, learning the instruction set is close to the entirety of learning math. For example, when learning programming you have to learn about arithmetics, pointers, functions and structs. And that is it, now you can program everything as long as you have an API to program against since everything else builds on top of those. That is equivalent to learning algebra in math. You can do a lot of things using algebra, but every new course introduces a lot of new concepts that can't be expressed in elementary algebra. Programming however becomes the art of expressing everything using these very simple instructions, while math is the art of creating instructions that can express things simply.