15
$\begingroup$

Recently I've resumed work on a font I'm currently developing, meant to be released as a completely free, open-source, OFL-licensed font designed specifically for use in papers, theses, books, lecture notes, and any kind of mathematics document more generally. It will consist of a family of fonts for text together with a family of fonts of matching style for math mode.

Since there are many typographical choices and specificities connected to mathematical practice, I wanted to ask for suggestions/wanted features from other mathematicians, in particular from those working in areas different from mine, which might have slightly different typographical needs which I'm not aware of.

As a concrete example of this, having support for letters from other alphabets used as math symbols is important in some areas, such as Ш for the Tate–Shafarevich group in arithmetic geometry, or よ for the Yoneda embedding in category theory.

A different kind of suggestion would be LaTeX specific optimisations or features for fonts which would be nice to have in practice, including e.g. a bigger set of delimiters, or built-in optimisation of microtype through a variable width axis (as e.g. discussed here), a feature relating to the text part of the font.

That said, in a vacuum a question like "Suggestions for features a math font should have" would arguably at best be a soft question, and at worst off-topic for MathOverflow. Nevertheless, given the free, open-source, nature of the font and the fact that ultimately it will be available to the broad mathematical community, I believe it would be reasonable to make an exception in this case, asking for input from the broad mathematical community as part of developing the font with the community in mind, and I believe MathOverflow would be a nice place for receiving this kind of input.

So―would it be okay to ask such a question in the main site? There I would define the exact gap the font is meant to fill, as well as give examples of the kind of input I'm looking for as an attempt to focus the question and define it a bit more clearly (although any input is going to be more than welcome)

$\endgroup$
23
  • 2
    $\begingroup$ By the way, here's an image of the current version of the font (which lacks kerning, so please ignore the bad spacing): Link. The image shows a paragraph of a recompiled version of Bhatt–Lurie's The prismatization of $p$-adic formal schemes as an example. $\endgroup$
    – Emily
    Commented Jan 27 at 16:47
  • 3
    $\begingroup$ As I read your description of the question, I am torn between "wow, what a neat question!" and my rules-lawyering complaint "that's not really about research-level mathematics in the usual sense here." Questions that straddle that kind of borderline can do very well, but they can also catch the community on a grumpy day and do very badly. So I think getting a commitment for support here first is a very good idea, so that the question can be re-opened if closed. $\endgroup$
    – LSpice
    Commented Jan 29 at 22:14
  • 2
    $\begingroup$ @LSpice I'm happy to hear you liked it! I feel the same way about the question being right on the boundary of an okay/not okay question for MO. I'd argue the thing that puts it on the "okay" side is the fact that I'm planning to implement the suggestions into an actual font that the community will be able to use for free; if it was just "nice things you'd like in a math font [that no one has committed to make]" I feel it would fall outside the scope of MO. $\endgroup$
    – Emily
    Commented Jan 30 at 0:11
  • 3
    $\begingroup$ The ideal audience for your question would be those in the intersection of MathOverflow and TeX.SE; the average mathematician may not care enough about typesetting matters (nor be knowledgeable enough) to answer questions about the subtlety of font design, and many folks on TeX.SE may not have the math-specific awareness (heck, I didn't know until today that Japanese kanas are used regularly in category theory). I am not sure whether here or TeX.SE is a better venue for your question, in part because I am not sure what you mean by "features". $\endgroup$ Commented Jan 30 at 2:33
  • 1
    $\begingroup$ (The two examples you gave are so completely distinct from one another.) I think for your question to be well-received on either community (remember, the SE platform really is preferring "answerable" questions) it needs to be more concrete. But returning to the examples you gave, things like having math support for non-latin alphabet characters and bigger delimiters do require also ways to access those features, which makes TeX.SE possibly a more natural venue. (See the unicode-math package and the MT pro font.) $\endgroup$ Commented Jan 30 at 2:38
  • 2
    $\begingroup$ Finally, it may be worthwhile to wordsmith your exact question here a bit to see community feedback before asking it on main. $\endgroup$ Commented Jan 30 at 2:40
  • 1
    $\begingroup$ Since Willie Wong mentioned [tex.]se, I will point out that they have a very active chatroom. (The MathOverflow chatroom is visited much less.) $\endgroup$ Commented Jan 30 at 3:37
  • 2
    $\begingroup$ My first thought was "is there really a need for a new LaTeX font?", but upon looking at the example I found it so nice I think I might use it without any need. Great job! $\endgroup$ Commented Jan 30 at 21:58
  • 1
    $\begingroup$ Also you might discuss it with the Typst community typst.app at their discord forum, as there is a budding broad range of open people upgrading Open Source scientific typesetting there. $\endgroup$ Commented Jan 31 at 8:01
  • 1
    $\begingroup$ I think one way to make the question suitable for MO is to phrase it as, what features might research mathematicians want, that might not be obvious to someone who is not a specialist in their field? That said, I'm having some trouble coming up with an answer to that question. Maybe some special binary relations/operations or symbols? $\endgroup$ Commented Jan 31 at 20:42
  • 1
    $\begingroup$ In some literature on the geometry of PDEs the Cyrillic capital letter Э is used to denote "evolutionary vector fields" on a jet bundle. $\endgroup$ Commented Feb 1 at 13:52
  • 2
    $\begingroup$ Now posted on the main site: Suggestions and feature requests for the design of a font for math articles/books $\endgroup$ Commented Feb 11 at 6:33
  • 1
    $\begingroup$ How about cross-posting it to cstheory.stackexchange.com and physics.stackexchange.com ? $\endgroup$ Commented Feb 12 at 11:46
  • 1
    $\begingroup$ @ClaudeChaunier I thought about cross-posting it to https://physics.stackexchange.com a week or two after posting the question on MO, but was afraid people would not like the idea. I'm really happy to know you also thought of this! I think I'll wait a couple weeks then (or at least a week after the question has reached inactivity), and after this open a post on the physics SE meta asking if it would be okay to crosspost it there. $\endgroup$
    – Emily
    Commented Feb 12 at 18:26
  • 1
    $\begingroup$ @ClaudeChaunier I also hadn't thought about cross-posting to https://cstheory.stackexchange.com; that's a really great idea! I think there are probably even more SEs for which cross-posting would be a good idea, like e.g. https://linguistics.stackexchange.com for the IPA part of the font. Thank you so much for the suggestion, Claude! $\endgroup$
    – Emily
    Commented Feb 12 at 18:27

2 Answers 2

10
$\begingroup$

Here's an alternative way of gathering feedback.

  1. Host your project on a standard project repository (like GitHub or something similar).
  2. Gather feedback as feature requests on your project.
  3. Spread word about your project on social media. (I for one will be willing to come out of Twitter dormancy to share it, since like LSpice I find the idea quite neat.)
$\endgroup$
2
  • 2
    $\begingroup$ Thank you so much for the feedback, Willie! I plan to create both a GitHub repository for the font as well a site for it (hosted on GitHub Pages). The only thing I'm afraid of if I were to exclusively choose this is that I fear not many mathematicians are used to opening issues on GitHub, as well as that I would probably not reach as wide an audience than if I additionally asked also on MO. I was hoping to hear mathematicians who are both knowledgeable about LaTeX, fonts, GitHub, etc. and those that are not so much, as I feel there will probably be great suggestions from both groups $\endgroup$
    – Emily
    Commented Feb 2 at 0:07
  • $\begingroup$ (P.S. I'm also super happy to hear you liked the idea! =) $\endgroup$
    – Emily
    Commented Feb 2 at 0:12
3
$\begingroup$

Here's a draft for the question on the main site, following Willie's idea.


Introduction. I'm currently working on a font family built for use in body type, named "Darwin" (after one of my dogs!), in particular in mathematics articles, books, etc. It is meant to fill the following requirements, of which no currently available fonts for LaTeX fill:

  1. It must be completely free and open-source.
  2. It must have exceptional symbol coverage among many different languages.
  3. It must come with a number of optical sizes, which are additional font families with size-specific optimizations―think an extra font for footnotes, a font for usual paragraph text, and an extra font for book covers.
  4. It must come with a matching math font family to uniformise the style across the document. This includes all the usual math symbols as well as font families for \mathbb, \mathsf, etc.

I believe Linus Romer's Elemaints typeface will fill the above gap well too once it's ready. Darwin will then work as an additional option with a significantly different style, similarly to Computer Modern vs. Palatino or Libertine.

Currently the regular style of the font looks like this (please ignore the bad spacing): PDF samples in English, Russian, and Greek are available here. These were compiled from sources of papers directly downloaded from the arXiv―I bear no relation to their authors.


Question. Throughout my own research as well as discussions with friends and colleagues, I've often found situations in math which require special symbols or features from fonts/LaTeX packages, as well as distinct preferences when it comes to certain choices of fonts. For example, these include:

  1. The use of symbols that are specific to certain areas and not covered usually, like the hiragana よ for the Yoneda embedding in category theory, the Cyrillic letter Ш for the Tate--Shafarevich group in arithmetic geometry, the Cyrillic letter Л for the Lobachevsky function in hyperbolic geometry, the Cyrillic letter Э for evolutionary vector fields on a jet bundle in some articles (e.g. page 14 of arXiv:0410185 on the geometry of PDEs (as pointed out by Igor Khavkine), etc.
  2. Better glyph support for \mathcal, \mathbb, \mathfrak, etc., ranging from \mathbb for the lowercase alphabet, to \mathbb for Greek, to more exotic things like \mathcal or \mathfrak for Greek. Appropriate support for these should have the style of the new symbols match the rest of the document, something which current workarounds like \usepackage[bbgreekl]{mathbbol} fail to fulfill.
  3. A preference of "empty fill" over double-strucked blackboard bold or vice versa (i.e. ams vs txof \mathbb).
  4. A bigger set of delimiters, as discussed here, for situations where one would e.g. need parentheses bigger than \big(\big) but smaller than \Big(\Big).
  5. Built-in optimisation of microtype through a variable width axis, allowing for an even better spacing optimisation than microtype already wonderfully does.

I plan to address all these points in the development of Darwin, adding all Cyrillic glyphs for mathematics, designing matching families for \mathbb Greek and alike, having an option available to choose between one of the two styles of blackboard bold above, as well as implementing the LaTeX specific features outlined in Items 4 and 5 above.

However, I'm sure the above list is anything but exhaustive, and there ought to be countless more such features which would benefit the community. So, in the spirit of the items above, I'd like to ask: are there any font features you want to have as part of your work, but weren't available or required workarounds at the time you thought of them?

Here features should be understood as including things in the spirit of the above examples, but not limited to them.

I also want to point out that suggestions without technical details (like implementations in LaTeX) are very welcome as well, e.g. "this symbol sketched below would work great for doing [x] in area [y], but I'm not aware of an implementation of it, or find the current ones flawed because of [reason]"

$\endgroup$
9
  • 4
    $\begingroup$ I like the way the question is phrased here! (Especially since it brought to mind a few things that I would have liked to have, things like additional variable size delimiters like doubled parentheses ((...)) and doubled brackets [[...]] and doubled angle brackets << ... >>.) $\endgroup$ Commented Feb 2 at 2:31
  • 1
    $\begingroup$ @WillieWong I'm very happy to know that! =) (I also share your pain! I remember having to put together a super messy macro some time ago for ((...)) and [[...]] to use with (what should have been) simple things like $\mathbb{F}_{p}((t))$, Laurent series fields, etc.; these sucked so much!) $\endgroup$
    – Emily
    Commented Feb 2 at 8:13
  • 2
    $\begingroup$ What's the format of the font that you are developing? Is it going to be an OpenType font, similar to NewCM (ctan.org/pkg/newcomputermodern?lang=en), or a more conventional format, as arxiv doesn't support lualatex or xelatex as of now? $\endgroup$ Commented Feb 3 at 10:13
  • 2
    $\begingroup$ The reason I mentioned NewCM is that it seems to me the project is similar to what you are developing, with a very large coverage of glyphs. $\endgroup$ Commented Feb 3 at 10:39
  • 1
    $\begingroup$ @ApoorvPotnis I'm planning to implement both the text font and the math font as a LaTeX package, so that it could be easily used with just \usepackage{darwin} and would work also in pdflatex for the arXiv. The text part of the font is also going to be released as OTF and TTF fonts, and I might eventually also make an OpenType version of the math font, after everything is done. $\endgroup$
    – Emily
    Commented Feb 3 at 19:25
  • 2
    $\begingroup$ @ApoorvPotnis NewCM looks really nice! I think once it has more optical sizes it would also fill the same "font gap" as Elemaints and Darwin are hoping to fill. $\endgroup$
    – Emily
    Commented Feb 3 at 19:25
  • 1
    $\begingroup$ At the cost of increased workload, can you also consider creating a bold math font? Normal math combined with bold text in section and chapter headers doesn't look great, imo. $\endgroup$ Commented Feb 3 at 20:02
  • 1
    $\begingroup$ @ApoorvPotnis Yep, thanks for the suggestion! I'm also planning to create optical sizes for math, so there will be extra fonts optimised specifically for footnotes, headers, and very large sizes, like book covers, with adjustments like these. $\endgroup$
    – Emily
    Commented Feb 3 at 21:06
  • $\begingroup$ @WillieWong I think one should also include {{...}}among these. $\endgroup$
    – YCor
    Commented Feb 23 at 17:08

You must log in to answer this question.

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