NRSI: Computers & Writing Systems
Doulos SIL — Documentation
The Doulos SIL fonts have been moved to a new web site. Please click here for the new site.
Doulos SIL Pages
The content below is old. Please click here for latest fonts and FAQs on the new site.
‘Thanks. I look forward to the liberating effect of being able to type diacritics easily.’
‘Downloading this for a class on linguistics. My TAs recommended it. :)’
Table of Contents for this web page:
The Doulos SIL font is designed to work on systems and with applications that provide support for TrueType fonts and for Unicode character encoding. This includes Microsoft Windows® 9x or greater, as well as recent versions of the Mac OS (version 9.0 and later), and also some implementations of Unix / Linux (TrueType font support on Unix and Linux may depend upon the particular applications in use). On some systems (true, at least, of 32-bit Windows), it can also be used with older applications that use legacy, industry-standard, 8-bit character encodings.
The preceding characterization of system requirements describes the minimum needed to display characters. Realizing the full capabilities of this font involves additional requirements. The first release of this font was designed to work with either of two advanced font technologies, Graphite or OpenType; beginning with release 4.0.12, it also contains AAT tables to offer selection of alternate glyphs on Mac OS X. To take advantage of the advanced typographic capabilities of this font, you must be using applications that provide an adequate level of support for Graphite and OpenType.
Features of the font
The Doulos SIL font contains near-complete coverage of all the characters defined in Unicode 7.0 for Latin and Cyrillic. In total, over 3,600 glyphs are included, providing support for over 2,300 characters as well as a large number of ligated character sequences (e.g., contour tone letters used in phonetic transcription of tonal languages).
In addition, alternately-designed glyphs are also provided for a number of characters for use in particular contexts. The glyphs are accessible in applications that support advanced font technologies, specifically the Graphite or OpenType technologies. These technologies are also utilized to provide automatic positioning of diacritics relative to base characters in arbitrary base+diacritic combinations (including combinations involving multiple diacritics).
Some important issues with respect to Unicode need to be borne in mind. Unicode is a character encoding and not a glyph encoding. Thus you should endeavor to use the character that reflects your character needs rather than finding a glyph that looks right and using its character code. Thus, for example, there is only one code for CAPITAL ENG (U+014A), although there are 4 different glyph shapes for this character in use around the world. Therefore it is necessary to use other means, such as user-selectable font features, to ensure that your document displays the right glyph for the character that you are anticipating. The advanced typographic capabilities mentioned above provide this very capability.
See also “How do I use a feature?”
Advanced typographic capabilities
This font supports various advanced typographic capabilities using the Graphite or OpenType font technologies.
The automatic placement of diacritics is supported for data that may or may not be canonically ordered (as defined by the Unicode Standard). This should normally be the responsibility of application software and text-processing resources (such as input methods), however, and not the user.
These capabilities are available in any application that supports the Graphite technology. They are also available via the OpenType technology, though this requires applications that provide a sufficient level of support for OpenType features. (See System Requirements.)
User-selectable font features
The document below can be downloaded in order to see all the user-selectable font features that are available in the font. The feature names, feature ids, settings and examples are provided. The document was produced with Graphite LibreOffice.
The Graphite features in this font are now handled by the CSS support in Firefox 11+.
Supported character ranges
This font supports over 2,300 characters from the Unicode 7.0 standard as well as a number of Private Use Area (PUA) characters. However, in this version of the font we have finally removed or strongly deprecated many of the PUA characters that are already in Unicode. In total, over 3,600 glyphs are included, supporting stylistic alternates for a number of characters as well as a large number of ligated sequences (e.g., contour tone letters used in phonetic transcription of tonal languages). The following character ranges constitute the list of characters supported by this font:
Supported Character Ranges
Private-use (PUA) characters
There are a number of private-use characters that are supported in this font. These conform to SIL International’s corporate registry for usage of the Unicode private-use areas. Characters which are not yet in Unicode are shown below.
U+F130 FONT BASELINE AND SIDE-BEARING MARKER LEFT
U+F131 FONT BASELINE AND SIDE-BEARING MARKER RIGHT
U+F132 FONT VERTICAL METRICS MARKER LEFT
U+F133 FONT VERTICAL METRICS MARKER RIGHT
U+F1A1 MODIFIER LETTER SMALL AE
U+F1A3 MODIFIER LETTER SMALL REVERSED E
U+F1A4 MODIFIER LETTER SMALL CLOSED REVERSED OPEN E
U+F1AB MODIFIER LETTER SMALL O WITH STROKE
U+F1AE MODIFIER LETTER SMALL CAPITAL OE
U+F1B4 MODIFIER LETTER SMALL CAPITAL Y
U+F1B5 MODIFIER LETTER SMALL RAMS HORN
U+F1BC MODIFIER LETTER SMALL H WITH STROKE
U+F1CD MODIFIER LETTER SMALL CAPITAL U BAR
U+F1CE MODIFIER LETTER SMALL TURNED Y
U+F1F1 MODIFIER LETTER PITCH ONE
U+F1F2 MODIFIER LETTER PITCH TWO
U+F1F3 MODIFIER LETTER PITCH THREE
U+F1F4 MODIFIER LETTER PITCH FOUR
U+F1F5 MODIFIER LETTER PITCH FIVE
U+F1F6 MODIFIER LETTER PITCH SIX
U+F1F7 MODIFIER LETTER PITCH SEVEN
U+F1F8 MODIFIER LETTER PITCH EIGHT
U+F1F9 MODIFIER LETTER PITCH NINE
U+F20D LATIN CAPITAL LETTER D WITH STROKE AND HOOK
U+F234 LATIN CAPITAL LETTER Z WITH PALATAL HOOK
U+F235 LATIN SMALL LETTER EZH WITH PALATAL HOOK
U+F247 LATIN SMALL LETTER AT
U+F248 LATIN CAPITAL LETTER AT
U+F258 LATIN LETTER SMALL CAPITAL I OVER SMALL SCHWA
U+F259 LATIN LETTER SMALL UPSILON OVER SMALL SCHWA
U+F267 LATIN SMALL LETTER TURNED Y WITH BELT
U+F268 LATIN LETTER SMALL CAPITAL L WITH BELT
U+F269 LATIN SMALL LETTER TURNED R WITH LONG LEG AND HOOK
U+F26D LATIN SMALL LETTER B WITH LEFT HOOK
U+F326 CYRILLIC CAPITAL LETTER GHE WITH STROKE AND DESCENDER
U+F327 CYRILLIC SMALL LETTER GHE WITH STROKE AND DESCENDER
Deprecated SIL Corporate PUA characters
WOFF (Web Open Font Format) is a font 'wrapping' technology that is poised to become the standard for web fonts for the long-term future and has broad support, even in Internet Explorer 9. It compresses the font data, so is more efficient (and faster) than referring to hosted TrueType or OpenType fonts. It also is likely to have better long-term support. As of October 2010, WOFF fonts are supported in Firefox 3.6+, Chrome 6, Internet Explorer 9 preview versions and development versions of Safari.
All of our Roman fonts now have standard WOFF versions of our fonts that you can download and host on your server. See the download page for the DoulosSIL-5.000-web.zip version of the font.
Further information on using and testing are available here:
In order to use this font with existing data that was created for use with fonts developed using the Encore Fonts system, or with custom-encoded fonts created by other means, it is necessary to re-type or convert data to produce data that is encoded in conformance with the Unicode Standard. TECkit is one program that can be used for character encoding conversion. TECkit allows users to write their own custom conversion mappings.
The TECkit package is available for download from SIL’s TECkit Web site.
Some TECkit mapping files have already been created for some of the more widely used SIL legacy fonts.
Many of SIL's Private Use Area codepoints are now in Unicode. All processes (input methods, mappings) that create Unicode data should be revised to generate the proper Unicode values instead of PUA codes.
If you have data that contains these PUA codes, it should be updated by replacing each PUA character with its official Unicode counterpart. This will facilitate data interchange and the use of standard fonts and software.
The ability to obtain full benefits of this font is also dependent upon having means for Unicode character input. This package does not include keyboard input methods. Most current operating systems provide keyboard input methods for a number of different languages that have writing systems based on the Roman or Cyrillic scripts. Various means may be available for different operating-system platforms to create additional input methods for other languages. Some of these are listed here: Some tools and resources for character input.