Happy to discover that the most recent build of Noto Serif Kannada can now handle Tulu language https://github.com/n8willis/opentype-shaping-documents/issues/163
Bobby de Vos
- 11 Followers
- 24 Following
- 57 Posts
@CelloMomOnCars I think you are correct about having resistive heat for very cold situations. I appreciate the additional details of how your system can still work in rather cold weather, but with lower efficiency I had wondered about that trade-off myself.
@counterpunch Even with shaping a real font in the editor (which sounds very nice, I need to try this) there are still differences between different OpenType shaping engines. Granted, some of those differences are small, some can be overcome with restructuring lookups. Other differences that affect real world applications are script itemizers.
@CelloMomOnCars In the article does -22 mean -22 C or -22 F? Assuming -22 F which would be -30 C (which is cold, granted) but people live in places where it gets colder than that.
Adobe InDesign seems to zero the width of a Mn character, even though it is not used in a lookup. This happens with the World Ready Composer, but not HarfBuzz (in or outside of InDesign). Why keep using the World Ready Composer? It seems to handle hyphenation differently than in HarfBuzz is enabled.
I have used the same LANG and LC_TIME example for years.
SIL uses ttfautohint for our staic Latin fonts, and it really helps with displaying small text on Windows. With a variable build of our Latin fonts, it seems to me that the display of small text is better in Windows, even without hinting the variable font.
@ohbendy I am not quite sure what is happening in this example, but in general, I would have thought that the PDF would have needed ActualText specified. Otherwise, if you have U+1000 (ka), U+1031 (e), the e needs to be displayed to the left of the ka, so the glyph stream would be uni1031, uni1000. How is the PDF reader, without ActualText, going to be able to reverse the re-ordering? Likewise with U+1000, U+1039 (virama), U+1000 the glyph stream is going to be uni1000, uni1000.sub. (no virama)
@ohbendy I was not aware that Myanmar script would be routed to the USE, I thought it had a dedicated shaper (mym2). Is there a mym3 tag like with Indic scripts (at least in some shaping engines)? Padauk has the character U+1700 TAGALOG LETTER A which Martin Hosken said is needed to trigger Myanmar shaping on older versions of Windows. Hopefully U+1700 is no longer needed.
@ohbendy With some other scripts I have been asked to help represent a sound in a language that the major language for a script does not have. So a diacritic, or a new character, or a shaping engine change might be needed. Especially with a new character or bugs in software, an understanding of Unicode properties is helpful to get a solution that will work well for users of that language. All that seems to be a bit different that the visual design issues you mention.