MathML existed then but support for it was scarce, in browsers it was poor to non-existent. For Internet Explorer there was the MathPlayer plugin. MathJax is a

MathML existed then but support for it was scarce, in browsers it was poor to non-existent. For Internet Explorer there was the MathPlayer plugin. MathJax is a JavaScript library for rendering MathML that's very good. However much of the use of MathML is focused on the presentation of the math, how it's rendered, rather than its semantics so that still makes it more challenging to learn. One of the better ways to access math is to learn human-readable TeX; it's just the parts of the TeX language needed to represent equations and scientific notation. TeX is already heavily used in STEM fields for laying out papers and such so there's material already in that form and there are many people who can write it. Even more people are familiar with its conventions like using the carat ^ to mark the beginning of an exponent, e.g. E=mc^2. Platforms like Wikipedia's MediaWiki can take human-readable TeX in the editor, render it to an image file in the published page but keep the TeX as the alt text of the image. MathJax can also render it. From Japan, there's InftyReader [1], OCR software that specializes in converting scans of mathematical equations, and ChattyInfty, a speech interface for InftyEditor, a mathematical document editor. Unfortunately, both are expensive. For statistics, the R language has existed for a while but I think it has become a lot more popular in the past five years or so. Having a command line interface, it's much more accessible than other statistics software like SPSS. [1] http://www.inftyproject.org http://www.inftyproject.org [2] https://www.r-project.org https://www.r-project.org