Skip to main content
Meta Stack Exchange

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

Required fields*

61
  • 50
    Admittedly your disclaimer had my finger hover over the downvote button in anticipation. But you’re obviously 100% correct, and this is kind of a deal-breaker. Commented Jan 27, 2021 at 17:56
  • 40
    It's not just MathJax, but all post formatting, see: "What site-specific post formatting settings are available?" - without a live preview composing on most of those sites will be very difficult; it's also a problem that enabling the alpha-test editor on Meta Stack Exchange breaks the other versions of the editors on other sites (already reported). Commented Jan 27, 2021 at 18:08
  • 51
    But this answer is exactly why we have this discussion here - We need to be able to see what we're missing, where our focus needs to turn to ensure parity in experience around the network. This editor, as I say, is months away from ever being shipped even to a single site, let alone made the default - and that assumes that we're able to get it in a good enough shape to meet the needs of the users. Understanding how much effort will need to be put into making it a great experience is why we're asking for your help. :) Commented Jan 27, 2021 at 19:41
  • 33
    While it's typical to start work on a "minimum viable product" and expand the feature set, this answer shows why any rewrite has to have an architecture that's designed from the beginning to handle the most difficult cases. One of the biggest problems with the bit-by-bit agile approach is that it too often eschews that kind of design. Commented Jan 28, 2021 at 19:41
  • 58
    I couldn't agree more with this answer. Latex or MathJax is essentially unworkable in a WYSIWYG setting. Many have tried and failed. For example, Overleaf has a rich text option, but it's not usable. If MathOverflow were to switch over to rich text with no option to revert to markdown, there would be riots. I, and I've no doubt many others, would strongly consider exercising our contractual option to leave the Stack Exchange network. I suspect users at Math Stack Exchange would be similarly upset, and in all likelihood the same could be said about various other technical sites. Commented Jan 29, 2021 at 18:15
  • 24
    It should be a no-brainer, if rich text editing is added, to allow the option to switch back to markdown. For instance, that's what wikipedia has. If you think you can design a workable rich text mathjax editor as an afterthought, when so many others have tried and failed with this as their main goal, then you are afflicted with profound hubris. Commented Jan 29, 2021 at 18:21
  • 35
    I would add, after reading the comments a bit more, than a solution like the MS Equation Editor -- which I'm sure was a huge undertaking to create and is an impressive piece of software -- would still be completely inadequate for use on MathOverflow at any rate. To ask a mathematician to use something like that is like asking them to type with their thumbs. I'm sorry if MO is an edge case, but all we'd need is the option to revert to markdown -- whether it's at the site level, or at the individual question /answer level. Commented Jan 29, 2021 at 18:32
  • 23
    To clarify my last comment -- the MS Equation Editor would be strictly worse than being forced into typing mathjax in rich text on two counts. First, it's less expressive: the range of what you can write the the equation editor is a small subset of what you can write in latex or mathjax. Second, even for the things you can write, it is in every case more cumbersome to to write in the equation editor that it is in mathjax. The only benefit of the equation editor is its decreased learning curve for use, which is a non-factor when the community standard is already Latex / MathJax. Commented Jan 29, 2021 at 18:59
  • 38
    One last comment (hopefully): You would never ask a developer to write their code in Microsoft Word. If it's not clear that there are a dozen good reasons for that, just ask any developer in the world. Asking a mathematician (or, I believe, a person in any of several other technical fields) to write in rich text is laughable for exactly the same reasons. Commented Jan 29, 2021 at 19:03
  • 20
    @DavidMulder In my (limited) experience those fields that can get away with using MS word to write mathematics, do not have really complex math formulas to begin with. For truly complex mathematical formulas (of the kind I routinely use every third lines in my papers for example), writing them in Word would be a nightmare hard to express. Note that I consider even Mathjax lacking (compared to a fully fledged LaTeX), especially in writing commutative diagrams, reverting to something like the equation editor would make writing answers essentially impossible for me. Commented Jan 31, 2021 at 9:13
  • 37
    @HamVocke Apologies -- from your comment, you seem to have missed the point here, and you're certainly not addressing the concerns. Nobody thinks that you will suspend MathJax as a plugin, but that is not what this is about. For the MathJax (+chess,go,music,furigana) composing to actually work, it needs a live preview. If your initial designs don't allow for one, then you're painting yourselves into a corner. It's not about "implementing site-specific plugins later", it's about creating a design framework that won't collapse under the weight of those plugins when you do enable them. Commented Feb 1, 2021 at 11:01
  • 22
    @HamVocke I too am confident you can build something that will fit into most users' workflow -- you have built one already. But it's hard to take the assertion that you're really flexible at face value, when the one thing that you don't seem to have allowed on the table is the one thing which is guaranteed not to break workflows (i.e. the thing we're asking for). Commented Feb 1, 2021 at 11:30
  • 22
    In addition, any suggestion that E.P. has engaged in any form of fearmongering would be completely unfair. E.P. has been meticulous in attention to detail and has bent over backwards to avoid doing any such thing, while at the same time being clear and blunt about the severity of these issues. Commented Feb 1, 2021 at 17:28
  • 47
    @HamVocke: I don’t mean to start a fight, but those aren’t “bold ideas.” They’re very old ideas that have been tried many times and always been eventually discarded in professional settings to return to the simple clean idea of having a separate preview. Commented Feb 1, 2021 at 19:41
  • 20
    @HamVocke That makes MathJax usable in rich-text mode, which is superior to it not being so. It still isn’t a rerendered-as-you-type preview, and it does nothing for the Markdown editor, which is by a vast margin the preferred way of working for almost everyone doing MathJax in the first place. If your proposal is “hey, this is another neat thing we could do,” sure, have at. If your proposal is “hey, if we did this, we wouldn’t need the separate preview, right?” then no, absolutely not. There does not exist any acceptable alternative to the separate preview. You must accept this. Commented Feb 2, 2021 at 15:44