Grants:PEG/Kelvinsong/Catastrophe copyleft font

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search

Kelvinsong/Catastrophe copyleft font
Fund the creation of a quality, comprehensive, and free typeface that meets the requirements to display Wikipedia articles in
targetEN, FR, DE, SV, FI, ES, NO, IT, PT, PL, NL, CA, DA, ET, RO, SIMPLE, SK, SL, TR
strategic priorityImproving quality, Increasing reach
start dateImmediately on approval
start year2016
end date6–9 months after approval
end year2016–2017
budget (USD)20,555–31,890 USD (see explanation)
grant typeIndividual
non-profit statusNo


A sample of the experimental Noctilucenta font released last May.

About one year ago, I posted a proposal on the Commons village pump that outlined the necessity of developing a copyleft font to support the operation of Wikipedia and the other text-based wikiprojects, as well as my efforts to develop a proof-of-concept of such a typeface. This proposal met very positive reception from users, and two WMF employees, on the Commons VP. (For reasons I don't know, those WMF employees never followed up, nor did they respond to my emails or talk page messages.)

The first goal of mine was to create a functioning example of a body-copy serif font that could be used to read Wikipedia articles in. This font came to be known as project Noctilucenta, which was developed with community guidance and received very positive feedback from testers on the Commons [1] [2]. Even though I am no longer developing Noctilucenta, some users still patch it into their common.css to this day. A build of Noctilucenta was successfully released about two months after the initial proposal, and you can view its demo page here. (github repo.)

While testers reported a massive improvement in the legibility of Wikipedia articles using Noctilucenta (see the above community discussions), Noctilucenta was never more than an experimental prototype. As I found while developing it, the production of a quality serif font requires much more time and resources than I had expected or was able to put into Noctilucenta, and so I was never able to produce anything more than a proof-of-concept product solely on volunteer time.

However, the Noctilucenta project did demonstrate that there is a huge potential for improvement in article legibility, that a better article font would be well received by the community, and that given proper resources and backing it would be very possible to accomplish such an improvement in wiki typography.


A working demo of the Catastrophe typeface. Please see the other pages of PDF to view more specimens.
A partial demo of Catastrophe’s italic. (I have not drawn the capitals yet.) Please see the other pages of the PDF to view more specimens.
  • The creation of a truly copyleft, quality text font adequate and appropriate for setting text in Wikipedia articles in.
  • This font is currently codenamed project Catastrophe, and I already have a working demo of the regular style, as well as a partial demo of the italic style, shown to the right. Also see a magazine demo I set, which I cannot upload on Commons due to copyright on the embedded image.
  • Catastrophe will be a family consisting of four to six standard styles (Light, Light Italic, Book, Book Italic, Bold, Bold Italic). The two light styles will depend on the extent that the WMF is willing to invest in this project.
  • Each style will support all the characters necessary to display text from the 19 target wikipedias. If additional funding is approved (see Budget), the Regular weight will also receive IPA characters which are common in many wikipedia articles, and supported by no currently existing libre text fonts (that are appropriate for article body copy).
  • Catastrophe will meet the standards of proprietary font quality, and provide an improvement in the legibility of wikipedia articles.
  • Catastrophe will be free to use by all, benefiting authors, creators, and the FOSS community at large.



This grant will fund the development and testing of the Catastrophe typeface by allowing me to work part-time on its construction. (Note that I am also a contracting type designer for Google Fonts, the Ambarisha typeface was a font I was commissioned by Google to design and create.)

Testers, overseers, and users in general will be able to follow the progress of the typeface on github where it will be developed in the open, with public contribution history. This will be in contrast to many “source-open fonts” today, and exemplify at least the transparency aspect of the “wiki model”.

Catastrophe will support the following character sets, sufficient to support a minimum of nineteen languages at the industry standard of quality:

  1. the Basic ASCII
  2. the Latin Supplement
  3. Selected glyphs in the Latin A : (Ā ā Ă ă Ą ą Ć ć Č č Ď ď Đ đ Ē ē Ė ė Ę ę Ě ě Ğ ğ Ģ ģ Ī ī Į į İ ı Ķ ķ Ĺ ĺ Ļ ļ Ľ ľ Ł ł Ń ń Ņ ņ Ň ň Ō ō Ő ő Œ œ Ŕ ŕ Ŗ ŗ Ř ř Ś ś Ş ş Š š Ţ ţ Ť ť Ū ū Ů ů Ű ű Ų ų Ÿ Ź ź Ż ż Ž ž ſ ƒ Ș ș Ț ț)
  4. Selected glyphs in the General Punctuation : (‐ ‑ ‒ – — ‘ ’ ‚ “ ” „ † ‡ • … ‰ ‹ ›)
  5. Selected glyphs in the Mathematical Operators : (∂ ∆ ∏ ∑ − ∕ ∙ √ ∞ ∫ ≈ ≠ ≤ ≥ ◊)
  6. Old style figures
  7. Small capitals
  8. Subscript and superscript digits (0–9)
  9. the IPA Extensions (if included in grant)

Catastrophe will also be built with support for Mathematical bold and italic, which will allow it to be used with MathJax or MathML.


The first two sections of the w:Saturn article set in the Catastrophe demo font.

Target readership[edit]

The creation of a dedicated article font will directly impact every reader on the nineteen supported language wikis. Each reader will view significantly more legible articles, articles which will also be much more consistent across platforms, and less dependent on unreliable system fonts. Catastrophe will serve as a foundation for future expansion, which will extend its benefits to users of Greek, Cyrillic-based, or Vietnamese wikis (or more). A quality body-copy text font will also facilitate print publications, raising the quality of our Wikibooks, for example.

Fit with strategy[edit]

  • Increasing quality
Project Catastrophe will increase the quality of our content for all the aforementioned reasons. Articles will be significantly improved in on-screen legibility and consistency, and facilitate a higher-quality conversion of our content into print form. Having a serif body-copy font will also be crucial going into the future with the increasing usage of high-resolution retina displays, which favor serif type over sans-serif type.
  • Increasing reach
A secondary effect of project Catastrophe will be an increase in access to Wikipedia among users who lack the necessary fonts on their computers to view certain special characters in Wikipedia articles. Templates such as w:Template:Contains special characters will become less common, particularly if the WMF approves the IPA extensions for Catastrophe.

Measures of success[edit]

This project will be considered successful if users respond positively to the finished typeface.

Resources and risks[edit]


  • I am an experienced vector artist and designer, with professional experience as a contracting type designer with Google Fonts. My vector art, since I started contributing in 2012, has received 31 Featured Picture distinctions on Commons.
  • I am working with highly experienced professional type designers such as Eben Sorkin (creator of the Merriweather font), who is assisting me with testing and quality assurance.
  • I received a TPS grant in 2015 to attend the Libre Graphics Meeting in Toronto and give a talk about the state of SVG illustrations on Commons. (Report)


This is a fairly low-risk project—in fact, there already exists a working prototype of the design which you can review above.


I have found the following formula to be useful for determining the cost of commissioning a font family (all values in USD):

Every font starts with an amorphous (conceptualization/design/testing/misc/etc) base cost of $1,000. Kerning costs about $500. Then the components of the font can be assigned work units, based on how long it takes to produce that portion of the font.
A–Z + a–z + 1–10 62 units
ASCII symbols 16 units
Latin supplement 10.5 units
Latin A 4 units
General Punctuation 5 units
Small capitals 26 units
Old style numerals 10 units
proportional/monospace numerals 2 units
Super/subscripts 7.5 units
Total 143 units
IPA Extensions 60 units
Total (with IPA) 203 units
Italics take the same amount of work, but bolds and lights take about one-half the number of work units of their regular-weight base fonts.
Style Without IPA With IPA
Regular 143 units 203 units
Italic 143 units 203 units
Bold 71.5 units 101.5 units
Bold Italic 71.5 units 101.5 units
Light 71.5 units 101.5 units
Light Italic 71.5 units 101.5 units
Then the total cost is found by adding up all the units, multiplying by the unit rate (my designer’s rate, for example, is between $90–100 per unit), and adding that to the $1,500 fixed cost.
Total cost of project

Were this a commercial project, a basic four-point typeface with no IPA will cost $42,110 at $90 per unit. However, for the benefit of the Wikimedia project and the FOSS community at large, I believe I can still produce this font with about half that amount: $20,555.

Regular 143 units
Italic 143 units
Bold 71.5 units
Bold Italic 71.5 units
unit rate × $90 / unit
fixed costs + $1,500
Total Value = $42,110
× 0.5
Total = $20,555

A six-point font with IPA support in the regular and the italic will cost $31,890. (There is rarely any need for bold or bold italic IPA glyphs so I do not recommend that the WMF fund those at the moment.) A reason to fund a six-point font is that having six masters instead of four allows us to use interpolation to generate multiple intermediate weights between the three drawn weights. The vast majority of proprietary fonts offer multiple weights.

Regular 203 units
Italic 203 units
Bold 71.5 units
Bold Italic 71.5 units
Light 71.5 units
Light Italic 71.5 units
unit rate × $90 / unit
fixed costs + $1,500
Total Value = $63,780
× 0.5
Total = $31,890
Total amount requested from the Project and Event Grants program

I've calculated the following matrix to determine the amount I would need to develop these possible feature combinations for Catastrophe:

Four styles 20,555 USD 25,455 USD
Six styles 26,490 USD 31,890 USD

Please note the market value for such a font is often up to 80,000 to 100,000 USD.

Additional sources of revenue that may fund part of this project, and amounts funded

This project was initially slated to be funded by Google Fonts, however Google dropped its support due to a shift in focus on their part towards expanding existing sans serif and decorative fonts in their own library. I also reached out to Canonical/Ubuntu and the Document Foundation but neither was interested or able, respectively, to fund this project at the moment.

Non-financial requirements[edit]

  • Lobbying assistance in deploying the completed fonts on the target wikis.


Community notification[edit]

You are responsible for notifying relevant communities of your proposal, so that they can help you! Depending on your project, notification may be most appropriate on a village pump, talk page, mailing list. Please paste a link below to where the relevant communities have been notified of this proposal, and to any other relevant community discussions. Need notification tips?



Do you think this project should be selected for a Project and Event Grant? Please add your name and rationale for endorsing this project in the list below. Other feedback, questions or concerns from community members are also highly valued, but please post them on the talk page of this proposal.