Recently we have been working on improving the display of math within our article full text by implementing MathJax, an open source JavaScript display engine for mathematics that works in all modern browsers. Because it is JavaScript driven, it does not require the reader to download and install any plugin.

MathJax uses the MathML or TeX stored within the XML of the article and uses modern CSS and web fonts instead of images so that equations scale with the surrounding text at all zoom levels. If MathML or TeX is available MathJax will be used to render the math by default.

[An equation rendered as an image; before and after zooming]

Scaling the page when equations are delivered as graphics causes pixelation and the image becomes difficult to read

Rendering equations using MathJax, however, allows for all math to scale at the same zoom rate as the text

[An equation rendered as using MathJax; before and after zooming].

The reader has the option to turn MathJax off or on using the toggle switch. If turned off, the graphic of the equation will be displayed instead.

 Individual equations can be zoomed by clicking them (you can change the ‘zoom trigger’ to hover or double click; right click any equation to display options).

Other benefits of using MathJax include:

• Copy and paste: lets readers copy equations from articles into Word and LaTeX documents, science blogs, research wikis, calculation software such as Maple, Mathematica and more.
• Accessibility: compatible with screenreaders used by people with vision disabilities, and the Zoom feature allows all readers to see small details such as scripts, primes and hats.

Due to a known scaling problem in Internet Explorer we have decided not to enable MathJax in IE just yet. Most other browsers are supported including Chrome, Firefox and Safari.

Here are some articles which contain MathML and, therefore, rendered using MathJax.

Reducing the worst case running times of a family of RNA and CFG problems, using Valiant’s approach
Shay Zakov, Dekel Tsur and Michal Ziv-Ukelson
Algorithms for Molecular Biology 2011, 6:20 (18 August 2011)

Statistics of spike trains in conductance-based neural networks: Rigorous results
Bruno Cessac
The Journal of Mathematical Neuroscience 2011, 1:8 (25 August 2011)

We’d love to hear what you think!