User: Password:
|
|
Subscribe / Log in / New account

Liberation fonts and the tricky task of internationalization

Please consider subscribing to LWN

Subscriptions are the lifeblood of LWN.net. If you appreciate this content and would like to see more of it, your subscription will help to ensure that LWN continues to thrive. Please visit this page to join up and keep LWN on the net.

By Nathan Willis
June 19, 2012

Fedora is debating dropping the storied Liberation font family from its distribution in favor of a fork. Liberation was one of the highest-profile open fonts, but it has languished since its initial release. Licensing issues were part of the problem, but so was the subtler disconnect of Liberation's origin as a commissioned work from a proprietary company, without an interest in working with the community. But the pressures of internationalization means the community has long sought a replacement; one that it can continue to develop.

Liberation through the ages

Liberation was released in 2007 by Red Hat, who had commissioned the designs from the commercial foundry Ascender Corporation. The initial set consisted of three fonts — Liberation Sans, Liberation Serif, and Liberation Mono — specifically designed to have the same metrics as the proprietary Monotype fonts Arial, Times New Roman, and Courier New, respectively. That meant that every character in Liberation would be the exact same height and width as its counterpart in the proprietary font, and Liberation could serve as a "drop in" replacement without disturbing line breaks or pagination. In 2010, Oracle donated a fourth typeface to Liberation: Liberation Sans Narrow, which was designed to be metric-compatible with Arial Narrow.

The Liberation family was regarded as high-quality, but it covered only the Latin, Greek, and Cyrillic alphabets, which left a lot of writing systems unaddressed. That alone is not a problem; fonts can — and are — extended to new writing systems frequently. But Liberation was licensed under unique terms, which inadvertently prevented such expansion.

Originally, the license was the GPLv2 with the Free Software Foundation's standard font embedding exception (which specifies that embedding the font in a PDF or similar document does not make the document itself a "combined work" triggering the GPL). However, Red Hat subsequently appended additional clauses to the license covering trademark and intellectual property concerns, and included a custom anti-Tivoization provision. After an examination of the extra clauses, Debian decided that they constituted additional restrictions on the GPLv2, which made the license self-contradictory and the fonts impossible to redistribute. The FSF reportedly found the Liberation license not to be a self-contradicting paradox, but said it was incompatible with the GPL. Furthermore, in recent times the GPL-with-font-embedding-exemption approach has fallen out of favor as an open font licensing choice, largely in favor of the SIL Open Font License (OFL). Fedora is aware of this shift, and now recommends the OFL for font projects.

Regardless of the exact details, however, the general consensus was that Liberation's peculiar license was, at best, problematic. More importantly, the practical upshot that few people were interested in contributing new character sets. The fonts have essentially remained unchanged since 2007. Minor fixes and isolated characters have been added, but no entirely new scripts.

Replacement plan

But the metric-compatibility feature of Liberation was its main selling point: it enabled Linux users to share documents with colleagues that had the popular Monotype fonts installed (e.g., all Windows systems), while ensuring compatibility.

On May 17, Parag Nemade emailed the Fedora-devel list to request packaging the Croscore family as a default, to serve as an alternative to Liberation. The Croscore family covers all of the same language blocks as Liberation, plus several new ones (such as Hebrew, Pinyin, and many African alphabets). It consists of three fonts: Arimo, Tinos, and Cousine, which also offer metric compatibility with Monotype's Arial, Times New Roman, and Courier New.

They were commissioned by Google for use in ChromeOS, and not only are they also the work of Steve Matteson, the same designer who created Liberation, but they are in fact a more recent version of the exact same designs. In an amusing bit of irony, however, Ascender Corporation (Matteson's company) was acquired by Monotype in 2010, so the new font family is copyrighted by Monotype, but designed to replace other Monotype fonts.

More to the point, however, Google made Croscore available under the OFL, which makes it simpler for outside contributors to extend the fonts to new character sets. Following the discussion in Nemade's thread, Fedora font packager Pravin Satpute proposed importing the Croscore sources into the Liberation package, replacing the problematically-licensed content rather than starting a separate package.

The Fontconfig package handles automatic font substitution on Linux, so once a change is pushed through with rules that replace (for example) Courier New with Croscore's Cousine instead of Liberation Mono, the only remaining hurdle will be for users to get used to the new names in the "Font" menu. On the other hand, growing the fonts to extend coverage to new writing systems is not trivial. The OFL license makes it easier; it enables developers to import and reshape glyphs from the large assortment of other OFL-licensed fonts.

What makes all this so difficult, anyway?

The Fedora plan calls for the community to continue development on the "Liberation 2.0" series, in the open, where the original Liberation was not. It would probably be a minor story if it were not for the fact that the same stalemate situation has developed for other open font commissions.

Much the same sequence of events befell the Bitstream Vera font family, which was designed by Bitstream (another commercial foundry which has since been acquired by Monotype) for the GNOME Foundation, and released in 2003. It, too, was under a license unique to the project, and has not seen any significant updates since its original release. Google has commissioned two fonts for distribution with Android: the familiar Droid family and the newer Roboto; both licensed under the Apache License (as is most of Android itself). Both offer wide language coverage in at least one of the faces (the sans serif), but have not otherwise seen significant expansion.

About the only open font commissioned from a commercial foundry that has grown to include more languages and alphabets is the Ubuntu font family designed by the Dalton Maag foundry. Although the details are of course private, Dalton Maag has an ongoing arrangement with Canonical to add more character sets over time. But the project does use a public issue tracker and accepts input and feedback from the user community, which none of the other commercial font commissions do.

Those differences are revealing. Commissioned open font projects such as Liberation and Bitstream Vera invariably attract significant attention — as do large "donations" of other types to open source. But when they are delivered in a self-contained bundle and not developed further, they have far less impact. It is easy for those of us who natively read European languages to forget just how many writing systems are not covered by basic Latin, Greek, and Cyrillic. Meanwhile, there are purely community-driven font projects that do cover far more of the globe's writing systems, such as Linux Libertine or DejaVu (the latter of which extends Bitstream Vera, side-stepping the peculiar Bitstream license by releasing its changes into the public domain).

The perception among the public is that the commercial fonts are of higher quality than the community-built fonts; a charge likely to rankle anyone who works on free software professionally. But by choosing non-standard licenses and not establishing the font as a software package that can be studied and patched, the early commercial commissions made that charge difficult to disprove. The problem is exacerbated when the foundry is uninterested in continuing to participate (as Bitstream was, when it said it was only interested in extending Vera if it were paid to do so).

But stagnation is detrimental to a font package just as it is to any software. Not only can every font be extended to cover more languages, but there are updated technologies like OpenType features and the Web Open Font Format (WOFF) to consider. Adding new character sets to a font is clearly a challenging task, demanding familiarity with multiple alphabets, and often requiring patches to be integrated by hand in a tool like FontForge. Hopefully the re-started Liberation 2.0 effort can draw on the lessons learned by Dalton Maag and DejaVu, and grow a sustainable project around the family. The original Liberation fonts filled a vital gap on Linux desktops and watching them languish has been disconcerting. Liberation now has the opportunity to re-import an entire codebase under a better license than the one that has hampered it for five years; few projects get a chance to start over at that same level — this is one that at least deserves to take a second shot.


(Log in to post comments)

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 19:15 UTC (Tue) by juliank (subscriber, #45896) [Link]

Please note that the Ubuntu font is considered non-free by Debian, as it imposes requirements on how derived works have to be named.

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 19:25 UTC (Tue) by directhex (subscriber, #58519) [Link]

Huh? DFSG clause 4:

The license may restrict source-code from being distributed in modified form _only_ if the license allows the distribution of patch files with the source code for the purpose of modifying the program at build time. The license must explicitly permit distribution of software built from modified source code. The license may require derived works to carry a different name or version number from the original software. (This is a compromise. The Debian group encourages all authors not to restrict any files, source or binary, from being modified.)

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 19:32 UTC (Tue) by juliank (subscriber, #45896) [Link]

Right. A different name. But the Ubuntu font license requires it to carry a specific derivation of the Ubuntu font name in some cases, not just a different name. The requirement was thus to strict and thus non-free.

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 19:34 UTC (Tue) by juliank (subscriber, #45896) [Link]

The "you must keep Ubuntu" in the name part can be considered equivalent to front cover and back cover texts in the GFDL.

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 19:57 UTC (Tue) by Jonno (subscriber, #49613) [Link]

Yes, and Debian don't consider GFDL-licensed works with invariant sections to be free either.

(GFDL-licensed works that don't contain any invariant sections are considered free by Debian, but not copyleft, as it is permissible to add an invariant section to it and thus make it non-free)

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 15:07 UTC (Wed) by fb (subscriber, #53265) [Link]

Really a pity that the license terms are such that they can hinder the adoption of a great font collection. Ubuntu Monospaced fonts replaced Inconsolata as my font of choice for programming.

Liberation fonts and the tricky task of internationalization

Posted Jun 21, 2012 20:41 UTC (Thu) by przemoc (subscriber, #67594) [Link]

Have you tried terminus? I'm using it (for ages) in debian as console font and as fixed-width font in X11. Indispensable.

Liberation fonts and the tricky task of internationalization

Posted Jun 22, 2012 11:19 UTC (Fri) by jengelh (subscriber, #33263) [Link]

>Have you tried terminus? I'm using it (for ages) in debian as console font and as fixed-width font in X11. Indispensable.

The problem is Terminus is a pure bitmap font. http://www.fixedsysexcelsior.com/ is the *real* thing ;-)

Liberation fonts and the tricky task of internationalization

Posted Jun 26, 2012 14:31 UTC (Tue) by Tet (subscriber, #5433) [Link]

The problem is Terminus is a pure bitmap font

That's a good thing. For printing, you want a vector font. But for terminal use, I still use bitmap fonts, and will continue to do so for the foreseeable future.

Liberation fonts and the tricky task of internationalization

Posted Jun 26, 2012 14:43 UTC (Tue) by jengelh (subscriber, #33263) [Link]

Pure bitmaps (like .fnt, .pcf) do not offer any advantage currently.
- There is no support to have them automatically scaled up. In contrast, you can run Fixedsys Excelsior TTF at 32px (i.e. twice its designed size) just by telling xterm/fontconfig to render it as such.
- Besides vectorized glyphs, truetype fonts can also have embedded bitmaps. This is very dominant in the CJK glyph range where the (antialiased) scaledown is often too blurry. But Arial/Times also have bitmaps for Latin scripts.
IOW, I'd love to have Terminus as TTF, just to be able to run it somehow at sizes larger than 32px.

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 20:18 UTC (Tue) by rfontana (subscriber, #52677) [Link]

This article contains an inaccuracy. It says:

> Originally, the license was the GPLv2 with the Free Software
> Foundation's standard font embedding exception (which specifies that
> embedding the font in a PDF or similar document does not make the
> document itself a "combined work" triggering the GPL). However, Red
> Hat subsequently appended additional clauses to the license covering
> trademark and intellectual property concerns, and included a custom
> anti-Tivoization provision.

This is not correct. The license was GPLv2 with the font-embedding
exception and the additional clauses from the beginning. This was a
result of Red Hat's agreement with Ascender.

Richard Fontana - Red Hat, Inc.

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 20:29 UTC (Tue) by rahulsundaram (subscriber, #21946) [Link]

Ah. I was wondering where the author sourced that information since I remembered it differently. Thanks for clarifying.

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 21:11 UTC (Tue) by rfontana (subscriber, #52677) [Link]

Also:

> The fonts have essentially remained unchanged since 2007. Minor
> fixes and isolated characters have been added, but no entirely new
> scripts.

This ignores the contribution of Liberation Sans Narrow by Oracle,
does it not?

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 21:17 UTC (Tue) by mstefani (guest, #31644) [Link]

It is there in the 2nd paragraph:
"In 2010, Oracle donated a fourth typeface to Liberation: Liberation Sans Narrow, which was designed to be metric-compatible with Arial Narrow."

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 22:06 UTC (Tue) by rfontana (subscriber, #52677) [Link]

Ah yes. My bad.

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 21:42 UTC (Tue) by n8willis (subscriber, #43041) [Link]

Sans Narrow would be a new face, not a new script. Interestingly enough, it is true that there doesn't seem to be a narrow equivalent from Croscore, so the question remains whether it would be less time-consuming to get an OFL-licensed version of Sans Narrow donated then add the new scripts, or take the regular-width from Croscore and tighten up its belt.

Nate

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 5:19 UTC (Wed) by nim-nim (subscriber, #34454) [Link]

Which shows another problem of custom non-standard licenses: good luck finding anyone at Oracle willing to relicense Narrow now they got rid of OpenOffice.

So Narrow is likely to be a casualty of the rebasing (it's missing Google-side)

And BTW while the OFL is a nice BSD-ish license, it's not making consensus now because everyone wants a BSD-ish licence for fonts, but because the FSF never bothered to write a good font copyleft license resulting in painful experiments like the font exception or the liberation license.

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 16:46 UTC (Wed) by rfontana (subscriber, #52677) [Link]

> Which shows another problem of custom non-standard licenses: good
> luck finding anyone at Oracle willing to relicense Narrow now they
> got rid of OpenOffice.

I'm the last person to defend the Liberation Fonts license, but the
problem you refer to has nothing too much to do with whether a license
is standard or not. It would exist even if Liberation Fonts had been
under vanilla GPLv2 + GNU font-embedding exception.

It does have something to do with whether the font license is copyleft
or not. As to that, though, you say:

> the OFL is a nice BSD-ish license, it's not making consensus now
> because everyone wants a BSD-ish licence for fonts, but because the
> FSF never bothered to write a good font copyleft license resulting
> in painful experiments like the font exception or the liberation
> license.

I don't understand the basis for describing the OFL as "BSD-ish". It's
actually a copyleft font license. See clause 5.

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 19:24 UTC (Wed) by nim-nim (subscriber, #34454) [Link]

> I'm the last person to defend the Liberation Fonts license, but the
> problem you refer to has nothing too much to do with whether a license
> is standard or not.

Of course it has. Standard licenses have well-known relicensing effects and you have a pool of other projects with compatible licensing to cross-pollinate with without involving lawyers. Projects with non standard licenses often end up like liberation : no other projects to draw on, few contributors willing to touch it (who wants his contributions shackled by terms that prevent future re-use), no other project you can contribute to without license poisoning (either because terms are incompatible with other licenses or because they would burden other projects with terms they don't want)

Even dejavu despite its own success remains alone and hampered by Vera's one-of-a-kind (if liberal) license. The few non-standard clauses make it impossible to integrate Vera glyphs in other FLOSS fonts without compromising their own OFL or GPL with font exception license (and good latin blocks are in high demand by creators of fonts for more exotic scripts)

If Liberation had used a safe standard license no one would need to ask Oracle today to relicense Liberation Narrow to extend its life.

> I don't understand the basis for describing the OFL as "BSD-ish". It's
> actually a copyleft font license. See clause 5.

It's a very weak copyleft, sources are not defined (so it's ok to reconstruct a font project from the produced binary font file, and to hoard the actual work files that the font editor uses), the naming and no-advertising clauses are definitely BSD-like

Which is not to say the OFL is a bad license, just that it's a lot weaker than the GPL is for software, and many would have been more comfortable with a stronger license if it was available for fonts (GPL with font exception does not really count as it's hard to understand in a font context by someone with no free software background, and causes no end of administrative problems if any link in the release process forgets to reaffirm this exception)

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 19:31 UTC (Wed) by nim-nim (subscriber, #34454) [Link]

And to complete: any project with non-standard licensing will be converted to a standard license sooner than later if possible just to benefit from the legal certainty of audited terms will well-known effects and because no project member likes having to explain how his license works to every passerby (the exception being if the project member is a lawyer — lawyers love licenses and having to discuss them. Non lawyers would like to forget the subject exists).

Any project which could not follow this process suffers from everything I explained before.

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 21:16 UTC (Wed) by rfontana (subscriber, #52677) [Link]

> (the exception being if the project member is a lawyer — lawyers
> love licenses and having to discuss them. Non lawyers would like to
> forget the subject exists).

I would just like to point out that not all lawyers are bad. :-)

Liberation fonts and the tricky task of internationalization

Posted Jun 21, 2012 5:42 UTC (Thu) by nim-nim (subscriber, #34454) [Link]

I didn't write they were (I hope :))

(offtopic) ELAW

Posted Jun 28, 2012 10:26 UTC (Thu) by gvy (guest, #11981) [Link]

Let me help you.

They mostly are unfortunately: the very approach of substituting natural human conscience with synthetic piles of buggy legal code tends to twist the very perception of basic concepts like truth and false.

And yes, we can still make such statements without fearing to catch a nice lawsuit -- freedom is still here; from Russia (actually Ukraine) with love :)

(offtopic) ELAW

Posted Jul 10, 2012 1:16 UTC (Tue) by dvdeug (subscriber, #10998) [Link]

Natural human conscience is wonderful, until you start to get into questions about whether we should or shouldn't kill witches, whether it's okay to copy a new movie (what if it's 50 years old? How about 100?), when we said we were making this software free, did we really mean people could sell it (okay, but that didn't mean Microsoft? Okay, but it's only fair that we have access to their changes, isn't it? We didn't make this software so people could make baby mulchers with it!)

Making agreed upon contracts, whether for everyone or only for a small group, about what may or may not be done is the only way to make sure everyone is on the same page.

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 21:14 UTC (Wed) by rfontana (subscriber, #52677) [Link]

> Standard licenses have well-known relicensing effects and you have a
> pool of other projects with compatible licensing to cross-pollinate
> with without involving lawyers.

Agreed (though is that point so applicable to font projects?).

> Projects with non standard licenses often end up like liberation :
> no other projects to draw on, few contributors willing to touch it
> (who wants his contributions shackled by terms that prevent future
> re-use), no other project you can contribute to without license
> poisoning (either because terms are incompatible with other licenses
> or because they would burden other projects with terms they don't
> want)
>
> If Liberation had used a safe standard license no one would
> need to ask Oracle today to relicense Liberation Narrow to extend
> its life.

There's no "end of life" for Liberation Narrow. It can continue on
under the Liberation Fonts license, as it has done for two years, in
the worst case. (I am actually hopeful that there could be some way
for Oracle to effectively relicense Liberation Narrow under SIL OFL.)

I now see the point you're making, which assumes that even GPL +
font-embedding-exception is a problematically nonstandard license with
bad effects on potential contributor community development. I suppose
I can agree with that. We relicensed the Lohit fonts, a somewhat
parallel situation to Liberation Fonts but not quite so bad, from
GPLv2 + font-embedding-exception to SIL OFL, and one of the
motivations for doing so from the project maintainers' perspective was
to expand at least the user community for the fonts. Our friends at
Google helped us out in this effort.

The correct conclusion may indeed be that no amount of legitimate and
well-intended addition of permissions atop GPL will yield a good font
license.

Liberation fonts and the tricky task of internationalization

Posted Jun 21, 2012 5:50 UTC (Thu) by nim-nim (subscriber, #34454) [Link]

>> Standard licenses have well-known relicensing effects and you have a
>> pool of other projects with compatible licensing to cross-pollinate
>> with without involving lawyers.

> Agreed (though is that point so applicable to font projects?).

It is very applicable — most free and open font authors will focus on their own native script, so producing fonts that work in a globalized context requires sharing between works what started as separate projects.

Ghostscript fonts, Vera, etc have been reused multiple times (and show up as licensing problems in new fonts regularly since their licensing is not as clean as new from-scratch projects)

Liberation fonts and the tricky task of internationalization

Posted Jun 21, 2012 5:53 UTC (Thu) by nim-nim (subscriber, #34454) [Link]

> The correct conclusion may indeed be that no amount of legitimate and
> well-intended addition of permissions atop GPL will yield a good font
> license.

The GPL intent could be declined in a new font-oriented license, but just adding new clauses won't work. OFL was successful because it was a license rewrite targeting fonts explicitly, with short and easy-to-understand terms.

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 21:18 UTC (Tue) by n8willis (subscriber, #43041) [Link]

My research said that the original (alpha? beta? unnumbered?) downloads circa May '07 did not have the custom exceptions, only the standard GPL-font-embedding-exception, as it says here: http://www.webcitation.org/query?url=http%3A%2F%2Fwww.press.redhat.com%2F2007%2F05%2F09%2Fliberation-fonts%2F&date=2008-01-17 and that the updated license was added subsequently.

Of course, that doesn't necessarily mean that that was intentional (a la this other licensing bug from Aug 2007: https://bugzilla.redhat.com/show_bug.cgi?id=250753). Unfortunately, the tarballs for the oldest releases here: https://fedorahosted.org/releases/l/i/liberation-fonts/ are timestamped from 2010, so that may be hard to prove conclusively, modulo time travel.

Nate

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 22:04 UTC (Tue) by rfontana (subscriber, #52677) [Link]

> My research said that the original (alpha? beta? unnumbered?)
> downloads circa May '07 did not have the custom exceptions, only the
> standard GPL-font-embedding-exception, as it says here:
> http://www.webcitation.org/query?url=http%3A%2F%2Fwww.pre...
> and that the updated license was added subsequently.

Ah, an understandable confusion. Mark Webbink described the whole
shebang as "GPLv2 + exception", which is not how I (who was not at Red
Hat at that time) would have described it. Indeed, if you look at the
text of the Liberation Fonts license you will see that both the
font-embedding exception and the anti-Tivoization provision are
described as "exceptions"
https://fedoraproject.org/wiki/Licensing:LiberationFontLi...

Yet this is not a standard use of "exception" in the GPL
context. Customarily, or in accordance with FSF practice at least, a
GPL "exception" is a grant of additional permission. So the
font-embedding exception is an exception, but the so-called
anti-Tivoization provision is not a true exception, but is, rather, an
imposition of an additional restriction. It is similar to, though
broader and less specific than, the User Products provision in GPLv3.

It is, in fact, for that reason that I urged Fedora to call the
Liberation Fonts license the Liberation Fonts license (rather than
describe it as "GPLv2 + exception" which in Fedora packaging metadata
parlance means GPLv2 plus additional permission).

Correction

Posted Jun 19, 2012 21:33 UTC (Tue) by paravoid (subscriber, #32869) [Link]

debian-legal serves as a discussion mailing list, not as a place were decisions are being made. The decision on whether something is free enough to be distributed in Debian main (vs. non-free, or no distribution at all) is taken by the ftp-master team.

I don't know about the specifics on this case, I do know however that fonts-liberation (previously known as ttf-liberation) is part of Debian for quite some time and is even in the current stable release (squeeze):
http://packages.debian.org/search?keywords=liberation

Liberation fonts and the tricky task of internationalization

Posted Jun 19, 2012 22:16 UTC (Tue) by rfontana (subscriber, #52677) [Link]

Sorry, I have one more criticism of this article. It says in its opening sentence:

> Fedora is debating dropping the storied Liberation font family from its
> distribution in favor of a fork.

But Croscore is not a "fork" of the existing Liberation Fonts.

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 9:27 UTC (Wed) by rvfh (subscriber, #31018) [Link]

Maybe the author meant 'drop it or fork it' about Liberation? That's how I understand it, especially as Croscore comes later in the article.

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 10:14 UTC (Wed) by randomguy3 (subscriber, #71063) [Link]

I think it was meant in the sense of

> not only are they also the work of Steve Matteson, the
> same designer who created Liberation, but they are in
> fact a more recent version of the exact same designs

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 16:41 UTC (Wed) by rfontana (subscriber, #52677) [Link]

Yes, I see Liberation and Croscore as they exist today as branches of the same Steve Matteson tree.

Liberation fonts and the tricky task of internationalization

Posted Jun 28, 2012 10:32 UTC (Thu) by gvy (guest, #11981) [Link]

s/a fork/another fork/ might be the proper fix.

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 8:09 UTC (Wed) by michaeljt (subscriber, #39183) [Link]

What is the story with Microsoft's new fonts (Calibri and friends)? They now seem to be a major hurdle for exchanging documents with Windows users if preserving formatting is important. Does anyone know if there is work going on to duplicate them in a FLOSS-compatible way?

Liberation fonts and the tricky task of internationalization

Posted Jun 20, 2012 13:55 UTC (Wed) by n8willis (subscriber, #43041) [Link]

That's a good question; I don't know. But I'm certainly going to have to look into it now....

Nate

Liberation fonts and the tricky task of internationalization

Posted Jun 22, 2012 23:07 UTC (Fri) by vonbrand (guest, #4458) [Link]

And what about the STIX fonts? They are under SIL, and have important backers (with LaTeX support promised for July).

Liberation fonts and the tricky task of internationalization

Posted Jun 23, 2012 12:51 UTC (Sat) by michaeljt (subscriber, #39183) [Link]

Are the STIX fonts a drop-in substitute for MS C* fonts for non-Mathematical things? My understanding (possibly incorrect, as I don't actually deal much with MS stuff) is that the most common problem is their recent use by default in Office documents, which makes the formatting come out badly when they are opened on non-MS platforms.

Linux Libertine

Posted Jun 20, 2012 12:14 UTC (Wed) by kramal (guest, #84631) [Link]

The article says, regarding community-developped fonts like Linux Libertine and DejaVu, that «The perception among the public is that the commercial fonts are of higher quality than the community-built fonts».

Does that perception really exist for Linux Libertine? It's one of the most beautiful general-use fonts that I know (including for-pay fonts). I've typed in whole books in it and loved it. Its italic characters are particularly nice and subtle, and I've heard people express admiration (without knowing what font it was, or that it was free).

Liberation Serif is also looks quite decent, but I'd say it's not quite so elegant,with those big pointed serifs and the pointy hook of the lowercase 'r'.

Linux Libertine

Posted Jun 20, 2012 12:28 UTC (Wed) by mpr22 (subscriber, #60784) [Link]

Libertine looks quite nice, yes. It's a shame that Biolinum shares with Arial the flaw of I-l underdistinction.

Linux Libertine

Posted Jun 22, 2012 22:34 UTC (Fri) by rahvin (subscriber, #16953) [Link]

Any san serif font is going to have the problem because it's the serif's that make the distinction. Frankly that's the whole point of using a serif font.

Linux Libertine

Posted Jun 20, 2012 13:54 UTC (Wed) by n8willis (subscriber, #43041) [Link]

I'm definitely not arguing it's a well-informed position. More of the "how good can it be if it's free?" mentality, which is still held by some people against open source software in general.

There are open fonts that are very highly-regarded by typographers (the SIL fonts, for example), but they have to fight against some prejudice from people who conflate "free" which "cheapo."

Nate

Linux Libertine

Posted Jun 20, 2012 15:19 UTC (Wed) by hmh (subscriber, #3838) [Link]

Usually the font face itself is fine, but the devil is in the quality of the hinting, and how well the small sizes in low DPIs will look. That, and the "fancier" (and by no means superfluous) ligatures, etc.

Professional commercial fonts do take high pains to get these right as a rule. The "free" ones you find scattered all over the web only do very rarely, AFAIK (which *does* mean the font deserves the low quality moniker, it will look awful at low DPIs and small sizes). IMHO, that stereotype is also attributed to the open fonts, be it deserved or not.

Linux Libertine

Posted Jun 21, 2012 11:20 UTC (Thu) by alankila (guest, #47141) [Link]

Tangentially, I'm working on a little patch to pixman that enables font rendering results that look like the upper samples (default rendering is on the bottom):

http://bel.fi/~alankila/pixman/fontscol.png
http://bel.fi/~alankila/pixman/fontsinv.png
http://bel.fi/~alankila/pixman/fonts.png

There is no hinting of glyph shapes here, just gamma-corrected compositing from sRGB surface to sRGB surface with freetype's light lcd filter to ensure grayness of the composition results. You should be able to see that the default code as a rule produces too dark rendering results. (And the default lcd filter chosen by cairo generates other kind of color fringing at the smallest sizes, but that is separate topic.)

Merely getting the gamma-corrected compositing right should improve results considerably, especially for small sizes where the error made is the greatest, as can be seen on the darkening and color fringing that results.

Linux Libertine

Posted Jun 21, 2012 18:37 UTC (Thu) by daniel (guest, #3181) [Link]

Wow, beautiful, I can read your 5 point font.

Also tangentially, here is a little project I worked on earlier this year, a slightly unusual take on 3D font rendering:

http://phunq.net/files/shiny81.png

That is Liberation Serif, which I found very pleasant to work with by the way, both for this subdivision modelling and classic low resolution 2D rendering. Sad to hear its not to be updated. The name is great, it would be a shame if that got locked up and died.

Linux Libertine

Posted Jun 21, 2012 0:18 UTC (Thu) by gdt (subscriber, #6284) [Link]

Most of everything is rubbish, it's just more apparent with fonts. Consider that there must be thousands of web servers written by students each year as they learn their craft, but no one would seriously use one in production. They're certainly not all collected together on websites with names like Free Web Servers.

Google Web Fonts were a real step forward, as they trawled through the dross and said "these fonts are good". Moreover since they are "web" fonts the quality of the font on a screen was important.

Linux users don't download software from the web, so why are they looking for fonts on the web? I'd suggest the reason is the complete lack of attention to fonts in distribution's package managers. They don't even do the simplest thing, such as displaying a sample of the font, let alone allowing searching for fonts by their properties.

Linux Libertine

Posted Jun 28, 2012 10:45 UTC (Thu) by gvy (guest, #11981) [Link]

We do to some extent, and you might have missed Fontmatrix.

Linux Libertine

Posted Jun 21, 2012 20:10 UTC (Thu) by daniel (guest, #3181) [Link]

Liberation Sans 20 pt, hinted, from an unusual perspective:

http://phunq.net/files/font24.png

(Apologizes in advance for the slight jaggies in the near field, which need a higher point size for the nearer glyphs, beyond the scope for this demo.)

This seems like a high quality font to me, including the hinting. In orthogonal, pixel for pixel rendering the effect of hinting is more dramatic of course but I think it improves the quality of even the perspective case. I noticed one odd and most probably undesirable behaviour: the hinting algorithm may move the end point but not the interior control point of a quadratic Bezier hull, resulting in an S-shaped artifact in the smooth outline. This can't be good for rasterizing. I doubt it is an error in my outline generator, but of course it could be, and I am not 100% sure whether the issue comes from the font hints or the hinting engine, but it looks like the latter to me.

Linux Libertine

Posted Jun 20, 2012 17:33 UTC (Wed) by walex (subscriber, #69836) [Link]

«commercial fonts are of higher quality than the community-built fonts».

Does that perception really exist for Linux Libertine? It's one of the most beautiful general-use fonts that I know (including for-pay fonts).

Here almost certainly "quality" refers to "on-screen quality", that is how well the font renders at much low DPI than printing.

This used to require extensive hinting, especially for TrueType fonts, and hinting TTF used to require special expensive tools, and a large amount of tedious work, and thus as a rule only commercial fonts, like the Microsoft Web Fonts, were well-hinted.

Various Linux suppliers have been in recent years paying foundries to improve the hinting of the fonts they sponsor, the free FontForge tool now makes hinting easier, and as a result some freeware fonts are not somewhat well hinted for low DPI, notably DejaVu, in my impression.

Linux Libertine

Posted Jun 21, 2012 9:23 UTC (Thu) by nim-nim (subscriber, #34454) [Link]

The big ugly secret is that Microsoft web fonts were *not* well-hinted.

To get good web font rendering in windows microsoft had to workaround those web font bugs in the font rendering stack of windows of the time

That's why :
1. microsoft web fonts look ugly anywhere else (it's not the other font stacks which are bad it's the font themselves which have buggy hints)
2. other fonts (with clean hints) look bad on windows when they trigger web-font specific workarounds they don't need

Linux Libertine

Posted Jun 21, 2012 6:54 UTC (Thu) by ncm (subscriber, #165) [Link]

To me the big mystery is why anyone still bothers with Liberation, or Deja Vu, or any of the numerous other ugly faces. Libertine is so clearly superior to them all -- even Droid -- there can be no contest where merit is considered.

I jimmied my fontconfig to substitute Libertine for everything except mono (where I use Inconsolata -- thanks Raph). I use Biolinum for window titles, and have altered Libertine so that the traditional looped st and ct ligatures are treated as standard. (One of the disappointments of Google Chrome is that it doesn't substitute standard ligatures. Or didn't, when last I checked. It's been a long time since I looked at Chrome or Google+.)

If anybody can suggest how to persuade (e.g.) Firefox to consider traditional ligatures standard without changing the font file, I would be very grateful.

Likewise, if anybody can suggest how to persuade (e.g.) Firefox to do spacing between bold letters correctly, I would be very grateful.

Linux Libertine

Posted Jun 21, 2012 8:51 UTC (Thu) by mpr22 (subscriber, #60784) [Link]

To me the big mystery is why anyone still bothers with Liberation, or Deja Vu, or any of the numerous other ugly faces.
De gustibus non est disputandum.

Liberation fonts: hinting needed

Posted Jun 21, 2012 20:25 UTC (Thu) by Richard_J_Neill (subscriber, #23093) [Link]

Unfortunately, the Liberation fonts aren't well hinted, with antialiasing off. This means that for those of us who value sharp contrast and clarity more than beauty and fidelity-of-typeface, they don't work well.

There are three ways to make fonts work well on a pixel-based display.

[0. Fixed, non-scalabled fonts (the old 75dpi and 100dpi fonts).]

1. Antialias. This adds grey (and sometimes coloured) sub-pixels to avoid jagged outlines. It makes the letters well-shaped, and true to the typeface. C.f. "Cleartype". About 20% of people (including myself) find the effect most unpleasant: it sacrifices contrast, and makes the display look out-of-focus, causing eye-strain. All truetype fonts can be antialiased.

2. Hinting. The shape of the letters is distorted (at small size) so that the lines fit more naturally over the pixel-grid. The letters are sharp and clear. An "e" is no longer a "times new roman 'e'"; more "an optimised e-grid, that is slightly times-new-roman-ish". All pixels are either black or white. With full-hinting and no antialiasing, this works. BUT, you must use the libfreetype which has the "Bytecode interpreter" turned on, and one of the small subset of fonts that are well hinted, Tahoma being a good example.

3. Retina display. Make the pixel density so high that the issue disappears. (alternatively, make the fonts really large, eg 25pt, with antialiasing).

[My personal recommendation is 8pt Tahoma for the GUI, and Terminus for the shell; full hinting, no-antialiasing]

Liberation fonts: hinting needed

Posted Jun 22, 2012 18:31 UTC (Fri) by daniel (guest, #3181) [Link]

"About 20% of people (including myself) find the effect most unpleasant: it sacrifices contrast, and makes the display look out-of-focus, causing eye-strain."

Perhaps that 20% would largely consist of owners of 1024x768 resolution monitors or worse. The "high contrast" you perceive in a black/white font is actually aliasing, an effect that is in general, highly unpleasant. Theory says it goes away completely when the highest spatial frequency of the source image is less than half the pixel grid spacing. This is achieved by low pass filtering, aka anti-aliasing. Reality is, for accurate image display you also need gamma correction, otherwise the intensity curve of a standard RGB monitor is wildly wrong. And the pixel spacing has to be less than what your eye can resolve, which the real test is not whether you can see jaggies or not (jaggies produce an easily perceivable nonlocal intensity variation) but whether you can count the number of lines in a series of one pixel stripes.

If you insist on using black/white text display then you need a really high resolution display to make the jaggies go away, but if you display properly band-limited images you will likely be fine with a display with just modestly improved resolution. Your goal is to get the highest, non-aliasing spatial frequency down below the resolution of your eye. To put this to the test all you have to do is move back from the monitor, increasing the point size of the font as you go until any perceived blurriness disappears. There is always such a distance and it is not as great as you think. Then the ratio between that distance and the distance at which you would actually like to view the screen is the amount by which you need to increase your linear pixel density.

Liberation fonts: hinting needed

Posted Jun 30, 2012 4:25 UTC (Sat) by k8to (subscriber, #15413) [Link]

I own a desktop 1600x1200 display and a laptop 1440x920 display. On both i find font 'anti-aliasing' of the cleartype variety creates blurry letters.

So I don't agree with this assessment.

Moreover the idea that high contrast is the same as aliasing is simply false. It's necessary to have aliasing at low resolutions to have high contrast, but even with "anti-aliasing" techniques, aliasing is still present, it is merely mitigated to some degree. I prefer clarity over prettiness. I'm not sure why one would ever elect differently for reading fonts.


Copyright © 2012, Eklektix, Inc.
This article may be redistributed under the terms of the Creative Commons CC BY-SA 4.0 license
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds