Le Ven 3 août 2012 10:51, Denis Jacquerye a écrit :
> OpenType allows weight values from 1 to 999.
> CSS is the one rounding those to multiples of 100 from 100 to 900.
> Fontconfig maps them to some range from 0 to 210 (I'm not sure there).
> Fontconfig also has multiple aliases for its weight val
Don't be so sure:
http://www.adobe.com/devnet/opentype/afdko/topic_font_wt_win.html
On Fri, Aug 03, 2012 at 10:14:54AM +0100, vern adams wrote:
> It's a bad OS/2 usWeightClass value (nothing to do with CSS). So, the '250' is
> getting rounded up to '300' and therefore clashing with the Light versi
ah it's the font files called 'font.ttf' in each weight that seem to have
fstype 0x0004.
I guess they don't count. My mistake ;p
On 3 Aug 2012, at 13:44, Dave Crossland wrote:
> On 3 August 2012 04:14, vern adams wrote:
>> some of the fonts have fsType of 0x0004 (Documents containing Preview
On 3 August 2012 06:22, vern adams wrote:
> An app like DTL's OpenTypeMaster flags non multiples of 100 as 'invalid
> usWeightClass'
That's a web specification, and this is (like vertical metrics) a
situation where you can't make a good tradeoff, the metadata either
has to be set for the web or f
On 3 August 2012 04:14, vern adams wrote:
> some of the fonts have fsType of 0x0004 (Documents containing Preview &
> Print fonts must be opened "read-only;" no edits can be applied to the
> document), but i assume Adobe means all the Source Sans fonts should be set
> to 0x.
Really?
$ for i
You can use whatever value you like, BUT, Thin=100, ExtraLight = 200, Light=300
etc etc
So if you create a font with a usWeightClass of 250 it's neither ExtraLight or
Light, but some software will need to know which it is and decide :)
Hence, it's a good idea to stick with a multiple of 100.
An a
On Fri, Aug 3, 2012 at 11:14 AM, vern adams wrote:
> It's a bad OS/2 usWeightClass value (nothing to do with CSS).
Where do you get that from?
The specs http://www.microsoft.com/typography/otspec/os2.htm do not
specify such restriction on usWeightClass, even if it describes
specific values. Micros
On Fri, Aug 3, 2012 at 11:14 AM, vern adams wrote:
> It's a bad OS/2 usWeightClass value (nothing to do with CSS). So, the
> '250' is getting rounded up to '300' and therefore clashing with the Light
> version which already has WeightClass of 300.
> I think it's simple human error; the usWeightCla
It's a bad OS/2 usWeightClass value (nothing to do with CSS). So, the '250' is
getting rounded up to '300' and therefore clashing with the Light version which
already has WeightClass of 300.
I think it's simple human error; the usWeightClass in the 'font.ttf' that Adobe
have included with the so
OpenType allows weight values from 1 to 999.
CSS is the one rounding those to multiples of 100 from 100 to 900.
Fontconfig maps them to some range from 0 to 210 (I'm not sure there).
Fontconfig also has multiple aliases for its weight value 40:
extralight or ultralight, and its weight value 210: bl
Okay, I pinged Paul via Twitter about that. Thanks for the hint!
Alexandre
On Fri, Aug 3, 2012 at 12:38 PM, vern adams wrote:
> That's right (i think) the usWeightClass in OS/2 tables must be multiples of
> 100. Source Sans Extra Light has a value of 250. The value should be 200, as
> the Light
That's right (i think) the usWeightClass in OS/2 tables must be multiples of
100. Source Sans Extra Light has a value of 250. The value should be 200, as
the Light version is already 300.
-v
On 3 Aug 2012, at 09:32, Khaled Hosny wrote:
> So I'd say it is a font bug (might be a FontConfig limita
On Fri, Aug 03, 2012 at 10:32:19AM +0200, Khaled Hosny wrote:
> So I'd say it is a font bug (might be a FontConfig limitation as well, I
> have some vague recollection about some discussion somewhere on whether
> OS/2 weight must be multiples of 100 or not).
http://lists.w3.org/Archives/Public/www
On Fri, Aug 03, 2012 at 12:04:47PM +0400, Alexandre Prokoudine wrote:
> On Fri, Aug 3, 2012 at 11:56 AM, Khaled Hosny wrote:
> > On Fri, Aug 03, 2012 at 11:07:50AM +0400, Alexandre Prokoudine wrote:
> >> On Thu, Aug 2, 2012 at 10:49 PM, Dave Crossland wrote:
> >> > Haha damn you that would have b
On Fri, Aug 3, 2012 at 11:56 AM, Khaled Hosny wrote:
> On Fri, Aug 03, 2012 at 11:07:50AM +0400, Alexandre Prokoudine wrote:
>> On Thu, Aug 2, 2012 at 10:49 PM, Dave Crossland wrote:
>> > Haha damn you that would have been my fault if it was messed up! XD
>>
>> BTW, it's the first time I noticed
On Fri, Aug 03, 2012 at 11:07:50AM +0400, Alexandre Prokoudine wrote:
> On Thu, Aug 2, 2012 at 10:49 PM, Dave Crossland wrote:
> > Haha damn you that would have been my fault if it was messed up! XD
>
> BTW, it's the first time I noticed this: is Pango really uncapable of
> seeing Extra Light wei
On Thu, Aug 2, 2012 at 10:49 PM, Dave Crossland wrote:
> Haha damn you that would have been my fault if it was messed up! XD
BTW, it's the first time I noticed this: is Pango really uncapable of
seeing Extra Light weights?
More than that, it renames weights. E.g. in both Inkscape and GIMP, I
get
Haha damn you that would have been my fault if it was messed up! XD
On Thu, Aug 2, 2012 at 10:44 PM, Dave Crossland wrote:
> On 2 August 2012 13:40, Alexandre Prokoudine
> wrote:
>> According to metadata:
>
> What metadata?
Oh, please ignore it. I'm a doofus :)
Alexandre Prokoudine
http://libregraphicsworld.org
On 2 August 2012 13:40, Alexandre Prokoudine
wrote:
> According to metadata:
What metadata?
On Thu, Aug 2, 2012 at 8:30 PM, Dave Crossland wrote:
> http://blogs.adobe.com/typblography/2012/08/source-sans-pro.html
Okay, this is becoming extremely puzzling.
According to metadata:
- typefaces were made by Ascender Corporation
- the license is Apache v2.0
- Open Sans is a trademark of Goog
21 matches
Mail list logo