Which Non-Latin Scripts Do Free Software Fonts Not Cover?
Here are the responses to my initial postings on various free software font mailing lists, all in one place.
dejavu-fonts@lists.sourceforge.net
No response
fontforge-devel@lists.sourceforge.net
On 16 Oct 2007 15:13:47 -0700, George Williamswrote: > On Tue, 2007-10-16 at 14:47, Dave Crossland wrote: > > Are there non-latin scripts that free software fonts do not cover at > > all at the moment? > > > > Which non-latin script communities are adopting free software and > > would benefit from more fonts? > > > Does FontForge have problems with the production of any scripts > > in particular? Are there any which it supports better than others? > Not that I'm aware of. FF supports all GSUB/GPOS lookups, it doesn't > really pay much attention to the script. > > It might have more trouble filling in some obscure Windoze code page for > the OS/2 table, but you can do that by hand now.
inkscape-devel@lists.sourceforge.net
On 16/10/2007, bulia byakwrote: > On 10/16/07, Dave Crossland wrote: > > Does Inkscape have problems with setting any scripts in particular? > > Yes. We have bug reports about japanese and korean (at least) > rendering. Unfortunately we don't have experts on these languages so > the reports are unfixed and even unconfirmed. Any kind of help will be > greatly appreciated; search our bug tracker for reports. Even if you > just test these reports and comment on whether the problem is real > with the latest version of Inkscape, it will be great help.
fonts@gnome.org
Hi Nicolas! On 17/10/2007, Nicolas Mailhotwrote: > Le mardi 16 octobre 2007 à 23:59 +0100, Dave Crossland a écrit : > > Hi All, > > > > I've just started the Masters degree in Typeface Design at the > > University of Reading (for which Gentium was a submitted project) and > > we're expected to do a non-latin script complement to our font, and > > I'd thought I'd ask around if anyone here has any opinions about the > > following: > > > > Are there non-latin scripts that free software fonts do not cover at > > all at the moment? > > DejaVu tracks user requests in its bugzilla and in the sfd fonts > themselves so I don't think finding glyphs to work on would be hard. Thanks for this tip, I'll do this and look into DejaVu itself for ideas :-) > Also DejaVu has a friendly & responsive team that can coach new type > designers. Unless you have the time and energy to bear the burden of a > major font project alone, I wouldn't advise contributing to a font > project that is not already organised to accept contributions from > external designers. I've just started the MA Typeface Design programme at the University of Reading, so I'll be getting a lot of coaching from proprietary designers, and will be committing a year of time and energy into a major font project alone, since its a difficult question of how to assess student work that is public-collaborative. Maybe some clever svn and fontdiff tools could sort that out, but I'm not a programmer (yet..) > > Which non-latin script communities are adopting free software and > > would benefit from more fonts? > > A common request is japanese in DejaVu. But that's a major piece of work > and I wouldn't start it without having native japanese contacts to QA > the result. (Red Hat also cited lack of FLOSS font support for indic as > one reason they bought liberation, and indic is another major piece of > work) I am heading more towards Indic, I think: Perhaps > > Or is everyone very happy with Vera? > > Distributions are abandoning Vera for DejaVu since Vera coverage haven't > evolved for years. But DejaVu is not perfect either — there's a lot of > glyphs to add, and a lot of glyphs to perfect. Sorry, I meant to say Vera-derived fonts, inclusive of DejaVu. > You won't find a lot of work to do in the latin range though — what's > missing is complex scripts. :-) On 21/10/2007, Nicolas Mailhot wrote: > Le dimanche 21 octobre 2007 à 12:04 +0100, Dave Crossland a écrit : > > > I've just started the MA Typeface Design programme at the University > > of Reading, so I'll be getting a lot of coaching from proprietary > > designers, > > However they won't help you with FLOSS font tools and if what you > release is only modifiable using $$$$$ proprietary tools the result > won't be really a FLOSS font I am personally only using free software, so there's zero chance of that! :-)
inkscape-user@lists.sourceforge.net
On 17/10/2007, Maarten van der Veldewrote: > > Hey, > > This is one of my many "sorry I can't realy help, BUT..." answers: > > I don't know if I understand you right, but aren't the a LOT of localised > LINUX (-LiveCD) Distro's out there? Even for excusez-le-mot, "ridiculous" > small language-communities? I'd say: have a look around among all the > Ubuntu releases and pick a small language, and see if they have all the > nice fonts WE have. I remember reading an article about a small, > Boudhist, east-asian state (200.000 citizens, or so?) where the gouverment > changed to Linux because Microsoft couldn't provide them with a localised > version of ther OS. They have their own alphatet. Anybody else > remembers this? They had a Tux in orange munk's robe as mascotte. > > Are you willing to share some nice fonts with us ^ ^ ? > > Maarten On 21/10/2007, Dave Crossland wrote: > On 21/10/2007, Maarten van der Velde wrote: > > Okay, this might get a little off-topic, but I was wondering if it is > > possible to design characters for a language one is not literate in? > > Certainly! Doing a complex script is one of the big parts of the > Masters degree I'm doing (see http://www.typedesign.rdg.ac.uk) and > apparently non-literate designers can do better work, because they > have to be more careful and study the details of the letterforms more > than a literate person. > > > Thinking of asian (f.e. Chinese) calligraphy, I remember that there was a > > lot ado about the balance and the harmony (etc.) of a character. > > That's true for all scripts! :-) > > > Don't get me wrong: this is no critique, just sheer curiosity from a > > font-fetishist... :D > > :-)
silgraphite-devel@lists.sourceforge.net
On 17/10/2007, Sharon_Correll@sil.orgwrote: > > Dave Crossland wrote on 10/16/2007 04:58:26 PM: > > > Hi All! > > > > I've just started the Masters degree in Typeface Design at the > > University of Reading (for which Gentium was a submitted project) and > > we're expected to do a non-latin script complement to our font, and > > I'm wondering: > > > > > > > > Are there any scripts which Graphite does not support well or at all? > > I don't know of any scripts that Graphite *couldn't* support, except for > something like Sign Writing, and we have ideas of an extension we could add > to make that possible. The next most complex would be Nastaliq-style Arabic; > we did design Graphite to be able to handle this, and knowledgeable sources > say it should be possible (didn't say easy :-). > > Now there are quite a number of scripts that don't have any Graphite-enabled > fonts in existence (Nastaliq Arabic, most of the Indic scripts, several more > from SE Asia).
scribus@nashi.altmuehlnet.de
On 17/10/2007, Gregory Pittmanwrote: > Philipp Klaus Krause wrote: > > Dave Crossland schrieb: > > > > > >> Are there complex scripts that free software fonts do not cover at > >> all at the moment? > >> > > > > -Mongolian (no longer used in Mongolia since the revolution). > > -Limbu. > > -Balinese. > > -Ancient greek musical notation (not really a script, but there's 75 > > unicode charcaters for it). > > -Byzantine musical notation. > > -Arabic lacks ligatures. > > -Sylheti Nagri (riginal sylheti script before they switched to bengali > > script). > > -Phagspa (can be used to write chinese, mongolian and tibetian, fell out > > of use with the rise of the ming dynasty). > > -Tagbanwa. > > -Buhid. > > > > I haven't checked very carefully though, so maybe there is some font for > > some of these in some obscure place. > > Philipp > > In general, I think we could say most South Asian, Southeast Asian > fonts. Better fonts would certainly be a great help to Scribus > developers for incorporating them into Scribus. > > Greg On 17/10/2007, Robert Memering wrote: > Am Mittwoch, 17. Oktober 2007 14:13 schrieb Dave Crossland: > > > > Which complex script communities are adopting free software and > > would benefit from more fonts? > > Hi Dave, > > here's another suggestion: A Greek OpenType font for special > editions, with ligatures like those in Matthew Carter's > Wilson Greek, plus everything needed for a critical apparatus. > > Of course there is Philokolia, > (http://www.ctan.org/tex-archive/fonts/philokalia) > but its letter shapes are quite peculiar. > > Regards, > Robert On 21/10/2007, Robert Memering wrote: > > Dave, > > should you choose to work on a greek font, I would > happily offer my help (e.g. with minor things you might > not want to do as part of the project, like providing > LaTeX support for these fonts). > > Regards, > Robert
gtk-i18n-list@gnome.org (Pango)
On 17/10/2007, Jens Herdenwrote: > > > > Which non-latin script communities are adopting free software and > > would benefit from more fonts? > > if you are looking for a real challenge you may visit www.khmeros.info > In Cambodia they are adopting free software step by step and there are not > that many Unicode fonts yet. You can find some fonts for Khmer on this > website and I am sure they would be happy about more. However Khmer script is > a little complicated and you need to work with Opentype tables. I guess that > you would get help from the fontdesigner who did the current fonts. > > I am CC'ing the project leader, please feel free to contact him directly for > further questions, because I am not in Cambodia anymore. > > Good luck > > Jens On 17/10/2007, mpsuzuki@hiroshima-u.ac.jp wrote: > Hi, > > On Wed, 17 Oct 2007 00:02:18 +0100 > "Dave Crossland" wrote: > >Are there non-latin scripts that free software fonts do not cover at > >all at the moment? > > Excuse me, you are asking about "there are proprietary fonts > for the script, but no free fonts for the script"? Or, > you are asking about "a script without any existing fonts"? > If former, please clarify the license you recognize as free. > For example, "Open Font License" is free in your context? > > >Are some scripts supported by Pango better than others? > > It's difficult question, I think. There are 2 kinds of > mechanisms to support a script by Pango: one is based > on OpenType technology, and another is based on legacy > encoding (I mean, non-Unicode) in the fonts designed > before OpenType (you can find some bits in Thai modules > of Pango). > > The code for 2nd mechanism is not growing anymore. > In fact, some of these are already removed (e.g. there > had ever been Indic support by legacy encoding). > > Growing is the 1st mechanism. But, the 1st mechanism > is based on OpenType technology which the specification > of script support is defined by Microsoft. Please check: > http://www.microsoft.com/typography/SpecificationsOverview.mspx > You can find a list of scripts which Microsoft supports > officially, in Script-specific Development. If a script > is not listed here but requires special complex text layout, > the script is not supported by "official" OpenType specification. > > A typical example might be N'Ko script. Pango already > includes a module to layout N'Ko script by OpenType font. > But Microsoft does not provide any specification for OpenType > table design to support N'Ko script. So, there is a > possibility that Microsoft can define incompatible "official" > N'Ko support in future and "official" N'Ko fonts doesn't > work well with current Pango implementation (and current > N'Ko OpenType fonts doesn't work well in future Microsoft > products). ICU does not support N'Ko script yet by such > reason, please check: > http://bugs.icu-project.org/trac/ticket/5865 > How do you think of? > > >Are there any scripts which Pango does not support well or at all? > > Some slightly-minor scripts that requires complex > text layout features are not supported yet, I guess. > For example, Limbu is an Indic script but current > Indic module does not support Limbu (basic module > is assigned to the script). There is a free-charged > font including Limbu is existing (damase_v.2.ttf), > but there's no OpenType support to tune the position > of multiple vowel signs, so it does not display the > ligature shown in the end of Table 10-5 in Unicode > 5.0 p. 362, I guess. > > But, I'm afraid, if you choose such script as your target, > your precious time will be spent for contacting native > users or linguistic experts, than for typographic works. > Do you want to concentrate yourself to typography? > > # In Japan, there were several academic projects to develop > # fonts and text rendering system for various Asian scripts > # (arabic & indic variants) before Unicode popularization, > # but the speed of projects had a bottle neck in the > # difficulty of communication between Japanese students in > # font production and native users of the scripts, I heard. > > Regards, > mpsuzuki On 17/10/2007, Behdad Esfahbod wrote: > On Wed, 2007-10-17 at 18:44 +0900, mpsuzuki@hiroshima-u.ac.jp wrote: > > > > A typical example might be N'Ko script. Pango already > > includes a module to layout N'Ko script by OpenType font. > > But Microsoft does not provide any specification for OpenType > > table design to support N'Ko script. So, there is a > > possibility that Microsoft can define incompatible "official" > > N'Ko support in future and "official" N'Ko fonts doesn't > > work well with current Pango implementation (and current > > N'Ko OpenType fonts doesn't work well in future Microsoft > > products). ICU does not support N'Ko script yet by such > > reason, please check: > > http://bugs.icu-project.org/trac/ticket/5865 > > How do you think of? > > IMO it's very obvious that if/when Microsoft supports N'Ko at some > point, it will be using the Arabic model, and that's what Pango does. > In fact, N'Ko was added to Pango simply by extending the Arabic joining > tables to include N'Ko characters and no change to the shaper code at > all.
ofl-discuss@openlists.sil.org
On 18/10/2007, Ed Tragerwrote: > Hi, Dave, > > How about the Lanna script of Northern Thailand and surrounding areas? > There are, to the best of my knowledge, no free Unicode fonts yet > available for this recenlty Unicode-encoded script. The script is > quite beautiful and has a number of interesting features which I think > will keep someone of your intelligence and caliber entertained and > interested in such a project :-) > > This script was just approved for inclusion in Unicode in September, > 2006. Martin Hosken from SIL and Michael Everson collaborated on the > proposal. > > Ngwe Tun (Myanmar) had had some reservations about a few details in > the encoding, which resembles the Khmer encoding model closely, so I > spent a couple of weeks taking a careful look at it. Ngwe's interest > and concern stemmed from the fact that the Lanna script is very > closely related to the old Mon script, so it is a lot like old > Burmese. The script was traditionally used for writing Buddhist > scripture in northern Thailand. > > Nowadays one also sees Lanna on shop signs in Chiangmai -- I believe > that is part of a modern cultural revivalism phenomenon. In addition, > there are some minority groups for whom Lanna is the only script they > use for writing. Martin can give you details. > > In any case, there are no freely available Unicode Lanna fonts. There > are some Lanna fonts that assume the Thai encoding, but that solution > was just a "kludge" used prior to Unicode encoding. The existing > fonts are worth taking a look at. You will also get to have a lot of > fun looking at hand-calligraphed old Buddhist scriptures and such. > > I have references for a number of resources which you can look at. As > soon as I find them all, I will send them to you in another email. > > Best Wishes -- Ed Trager > > P.S: Here are a few web resources to get you started: > > http://www.omniglot.com/writing/lanna.htm > http://se-ed.net/tua-mueng/ > http://www.seasite.niu.edu/thai/lanna/default.htm > > > > On 10/16/07, Dave Crossland wrote: > > Hi All, > > > > I've just started the Masters degree in Typeface Design at the > > University of Reading (for which Gentium was a submitted project) and > > we're expected to do a non-latin script complement to our font, and > > I'd thought I'd ask around if anyone here has any opinions about the > > following: > > > > Are there non-latin scripts that free software fonts do not cover at > > all at the moment? > > > > Which non-latin script communities are adopting free software and > > would benefit from more fonts? > > > > :-) > > > > -- > > Regards, > > Dave > > _______________________________________________ > > OFL-discuss mailing list : open discussion around the SIL Open Font License > > OFL-discuss@openlists.sil.org > > http://openlists.sil.org/mailman/listinfo/ofl-discuss > > Opinions do not necessarily reflect the views of SIL International > > > _______________________________________________ > OFL-discuss mailing list : open discussion around the SIL Open Font License > OFL-discuss@openlists.sil.org > http://openlists.sil.org/mailman/listinfo/ofl-discuss > Opinions do not necessarily reflect the views of SIL International >
fontforge-users@lists.sourceforge.net
On 16/10/2007, Barry Schwartzwrote: > Dave Crossland skribis: > > I've just started the Masters degree in Typeface Design at the > > University of Reading (for which Gentium was a submitted project) > > So was that wonderful typeface Maiola, but it, unlike Gentium, > separated me from many of my $$$$$$$$$$$$$. :) > > > and > > we're expected to do a non-latin script complement to our font, and > > I'd thought I'd ask around if anyone here has any opinions about the > > following: > > > > Are there non-latin scripts that free software fonts do not cover at > > all at the moment? > > > > Which non-latin script communities are adopting free software and > > would benefit from more fonts? > > Maybe you should study Alan Wood's Unicode font list: > http://www.alanwood.net/unicode/fonts.html On 17/10/2007, Gustavo Ferreira wrote: > hehe, the same message echoing in many lists (with small variations, > i noticed that). > > but you are asking to programmers and type-designers – not to graphic > designers, the main users of fonts! > > try asking on the scribus list – i bet you will get more replies > there... > > - - - > > regarding your question, my oppinion is: > > forget about minority scripts – if you want to promote free & open > fonts, think in terms of a large user base. > > option 1: focus on greek + cyrillic*. european union, dude! this is > where the problem with multilingual type is most relevant (in terms > of $cale). in the eu many products have info in many languages... i > remember how most samples i've seen while in germany/holland sucked – > you would have all latin-based languages set in the company's > corporate font, and greek and cyrillic in some helvetica/futura... > very ugly. > > * don't forget bulgarian localized forms! they are very proud of that. > > option 2: arabic! after latin probably the second most used alphabet > in the world. the question of modern arabic types is hot > (typographically and politically), there has been a lot of activity > around it. have you seen the typographic matchmaking project? > pascal's website? decotype's tasmeem system? > > http://www.khtt.net/artefact-266-en.html > http://29letters.wordpress.com/ > > good work, > - gustavo.
localization@lists.laptop.org
On 10/25/07, Xavier Alvarezwrote: > Hi! > > We seem to be getting the L10n effort under way (in a new server), > and the subject pretty much sums up the situation: > > We need > - translators (obviously), > - coordinators (that can actually manage each language) and > - volunteers (the universal glue?) > > All languages are welcome, but it should be noted that there's a > need for those languages used in the 'green countries', which > are: Amharic, Arabic, English, Spanish, French, Hausa, Hindi, > Igbo, Nepali, Portuguese, Romanian, Russian, Kinyarwanda, Thai, > Urdu, & Yoruba.
The Which Non-Latin Scripts Do Free Software Fonts Not Cover? by David Crossland, except the quotations and unless otherwise expressly stated, is licensed under a Creative Commons Attribution-Share Alike 3.0 Unported License.
November 6, 2007 |
Comments
Leave a Reply
