1

MathJax would be handy for formatting things like the formulas in the following abstract:

We analyze the most commonly used method for shuffling cards. The main result is a simple expression for the chance of any arrangement after any number of shuffles. This is used to give sharp bounds on the approach to randomness: $\frac{3}{2} \log_2 n + \theta$ shuffles are necessary and sufficient to mix up $n$ cards. Key ingredients are the analysis of a card trick and the determination of the idempotents of a natural commutative subalgebra in the symmetric group algebra.

Bayer, Dave; Diaconis, Persi. Trailing the Dovetail Shuffle to its Lair. Ann. Appl. Probab. 2 (1992), no. 2, 294--313. doi:10.1214/aoap/1177005705. http://projecteuclid.org/euclid.aoap/1177005705.

| |
  • 2
    MathJax is off by default, but we can enable it if there's a demonstrated need. How frequently do cases like this occur? – hairboat Feb 23 '15 at 18:52
  • @abbyhairboat "MathJax is off by default [...]" Why? Is it detrimental to performance? Does it takes a significant amount of development to enable? Your comment would make a wonderful answer to the question if it was explained. The question wasn't "Please enable it." but "Why is it disabled to begin with?" – Rainbolt Mar 2 '15 at 19:22
  • @Rainbolt, there's been a few questions about probability on this site. Since we cover poker, M:TG deckbuilding theory, and Hi Ho Cherry-O here, probability and math go with us. I'll cheerfully admit that we don't break out the math on the vast majority of our questions, but there's still a few that could use it. – PotatoEngineer Mar 4 '15 at 19:20
  • @PaulMarshall Why was that comment directed at me? – Rainbolt Mar 4 '15 at 19:44
  • Basically, the reason I asked abby about her comment is that you can't only cover the pros. Sure, we could use MathJax. But we need to know the downsides too, so we can weigh the cons and the pros against one another. If we use MathJax on 10% of questions but it decreases performance by 20% on all pages, well that's just not good. – Rainbolt Mar 4 '15 at 19:49
  • 1
    @Rainbolt (4 comments up) yes, it is detrimental to performance. My experience suggests that it adds an extra half second or more to page load times, depending on the speed of the client computer, though I'm sure there are more reliable studies about this. It may also have some impact on the servers. – David Z Mar 5 '15 at 7:53
  • @DavidZ Great. Do you want to post that as an answer? Coming from me, it's speculation. You seem to have experience to back it up. – Rainbolt Mar 5 '15 at 16:23
  • @Rainbolt well, most of what I know comes from the discussions we had with the SE team when deciding whether to enable MathJax on Physics; I wouldn't really call it experience. I don't think I have much of anything useful to add to your answer. – David Z Mar 6 '15 at 0:36
4

There is a trade off with performance. Think about it: if performance was a non-issue, MathJax would be enabled on every site. Commenter David Z suggested (don't hold him to it) that enabling MathJax "adds an extra half a second or more, depending on the speed of the client computer".

Is it worth hurting the performance of every single question in order to make the math prettier on a select few? Not in my opinion.

Also, the abstract you quoted is categorized under The Annals of Applied Probability on a mathematics and statistics website (Project Euclid). Is Board and Card Games the best place to ask a question about that? Perhaps it receive better answers on mathematics.stackexchange.com.

| |
  • Your reasoning is sound, but as you said we shouldn't hold commenter David Z on the statement that it adds at least a half second. Is there any way that we can get more reliable information on how much of a performance hit it really takes? – Thunderforge Mar 6 '15 at 2:50
  • @Thunderforge No, I don't, but common sense tells me that running more Javascript on a page will take longer than running less Javascript on a page. – Rainbolt Mar 9 '15 at 13:08
  • true, but if we can condirm that it adds something negligible (say, 10 ms), then I'd say it's worth turning on. – Thunderforge Mar 9 '15 at 14:33
  • @Thunderforge If the performance hit is negligible, and if there exists a significant number of questions that don't belong on Mathematics SE that require complex mathematical illustrations, then MathJax should be enabled. I think that the burden is on the one requesting MathJax to prove both of those conditionals. – Rainbolt Mar 9 '15 at 14:38

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .