Hacker News Reader

Comment by Tainnor | Hacker News Reader

As someone who has worked on a computer algebra system and knows what a pain it is to extract semantic meaning from e.g. LaTeX, I absolutely agree that a better format is needed to encode mathematics semantically. I just don't understand why they had to use a representation that is borderline unreadable and showcases XML's worst excesses.

View on HN

More than that: this is (just) semantic-web applied to the topic of maths. I personally hate semantic-web stuff and don't understand how it can possibly still be a research subject (with university departments dedicated to that). They talk about languages and semantics but keep ignoring the actual recent (i'm talking second-half 20th c.) research in programming language, type systems and formal maths. Some web focused project to "write formal maths with semantics", how more ridiculous can we get? Someone needs to get them to look at theorem provers.

Sorry for the rant, i'm just not gonna write a blog post here.

I wish you did though ;)

I am not a mathematician. But there is also a JSON representation [0] besides XML. As an exchange format between pieces of software, isn't it usable?

[0] https://www.openmath.org/standard/om20-2019-07-01/omstd20.ht...

No, that is just as bad. It's not about them using XML, it's about incorporating the worst aspects of XML culture by making everything overly abstract and unreadable. This problem persists in the JSON representation.

I agree with the other comment that it would have been better to look at e.g. theorem provers or maybe even at something like Mathematica.

Reply on news.ycombinator.com