Mathematicians Use The Darndest Syntax


For Goodness Sakes Just Put It In Prolog Or Lisp Notation Like You're Going To Do Anyway Since Nobody Uses Subscripts And Superscripts And Stuff In Programming And You're Literally Designing A Programming Language


Don't Be A Subscript Cargo Cultist

(this was 1991 so I guess nobody had desktop publishing then anyway but still, ouch)

Is it all the nested parens that scares the maths people? really? that? Well then okay there's a good argument to be made that you can eliminate lots of parens if you allow variables to range over the CDR of an expression so you could put that in Lisp as

(believes 3 a believes 10 b (time 7) like a b)

there, the nasty parens are all gone now

but don't frickin inflict super-sub-scripts on people just because it looks 'mathy', sheesh

@natecull serious answer: maths papers are written to be read by humans. Putting the “static” variable in a subscript keeps the reader’s attention on the important variable.

@tomharris Programming languages are also written to be read by humans, which is why they allow variables to have more than one letter.

@natecull @tomharris Be thankful mathematicians haven't discovered scripts other than Latin and Greek (and that one Hebrew letter).

@natecull you may be on to something, but I always thought that extra complexity was due to the belief that extracting all of that information left a very simple conclusion that seems "elegant".

I find it inelegant and difficult to read.

@ratamacue for me it's more 'good grief, you're inventing a programing language and now you're inventing a second notation for the basic units of your language which your language itself won't be able to parse, BUT, your first language is a UNIVERSAL KNOWLEDGE REPRESENTATION LANGUAGE FOR AI which will of course need to parse its own expressions, which you'll need to reflect back into its syntax, have you even thought five seconds about this'

@ratamacue but this happens ALL THE TIME in programming theory and AI papers, at least in the 1980s - early 90s.

Maybe things have got better today? I sure hope so.

AI papers are a whole additional nightmare to other math papers and stuck in notational purgatory. I see that much _{}^{} and immediately begin to complain that these aren't contra/co-variant, nor are they representing towers of fields, nor...

@natecull Why the blithering fuck does time t need to be subscripted? Make that shit an array index since POSIX machines store the current date/time in seconds.
Sign in to participate in the conversation

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!