MathJax uses MathML internally for representation, so it's not necessarily an either/or question. The bit that MathJax does is interpreting and rendering LaTeX formatted equations (which are...
MathJax uses MathML internally for representation, so it's not necessarily an either/or question. The bit that MathJax does is interpreting and rendering LaTeX formatted equations (which are ubiquitous in many academic fields) into consistent output for the web.
I've used MathJax a good amount but only have the barest understanding of MathML, so I certainly couldn't comment on whether or not scientists should be using MathML directly rather than translating from LaTeX - but even if it were the most perfect and elegant markup ever devised, inertia would count for a whole lot and keep many systems on LaTeX for years to come.
Although it's been a couple of years since I last used it, I'm very glad to see that MathJax is still getting serious development. It was one of the more significant sources of technical headaches in the project we were working on, but it did a damn good job of a very complex task. Having it do so without fighting against the rest of the codebase it's working alongside would be a big win!
A bit offtopic, but since I had never heard of MathJax before this post and your comment, and so was curious if it would help with implementing LaTeX on Tildes, I did some diving and discovered an...
A bit offtopic, but since I had never heard of MathJax before this post and your comment, and so was curious if it would help with implementing LaTeX on Tildes, I did some diving and discovered an interesting article on the performance differences between KaTeX vs MathJax:
What are the advantages of MathJax vs. MathML? I've only used the latter, so I'm not very familiar with the former.
MathJax uses MathML internally for representation, so it's not necessarily an either/or question. The bit that MathJax does is interpreting and rendering LaTeX formatted equations (which are ubiquitous in many academic fields) into consistent output for the web.
I've used MathJax a good amount but only have the barest understanding of MathML, so I certainly couldn't comment on whether or not scientists should be using MathML directly rather than translating from LaTeX - but even if it were the most perfect and elegant markup ever devised, inertia would count for a whole lot and keep many systems on LaTeX for years to come.
Although it's been a couple of years since I last used it, I'm very glad to see that MathJax is still getting serious development. It was one of the more significant sources of technical headaches in the project we were working on, but it did a damn good job of a very complex task. Having it do so without fighting against the rest of the codebase it's working alongside would be a big win!
A bit offtopic, but since I had never heard of MathJax before this post and your comment, and so was curious if it would help with implementing LaTeX on Tildes, I did some diving and discovered an interesting article on the performance differences between KaTeX vs MathJax:
https://www.intmath.com/cg5/katex-mathjax-comparison.php
Turns out KaTeX is still significantly faster than even MathJax 3.