Summary: | DejaVu serif italic æ | ||
---|---|---|---|
Product: | DejaVu | Reporter: | Mattias Ellert <mattias.ellert> |
Component: | Serif | Assignee: | Deja Vu bugs <dejavu-bugs> |
Status: | RESOLVED FIXED | QA Contact: | |
Severity: | normal | ||
Priority: | high | ||
Version: | unspecified | ||
Hardware: | x86 (IA32) | ||
OS: | Linux (All) | ||
Whiteboard: | |||
i915 platform: | i915 features: | ||
Attachments: |
comparison between a and æ
ae and oe in Times New Roman Italic DejaVu Standard ae Design sample of proposed design at tiny size |
Description
Mattias Ellert
2006-09-10 08:14:48 UTC
Created attachment 6898 [details]
comparison between a and æ
This was done on purpose. Some other fonts do that too. It's a way to avoid confusion between æ U+00E6 and œ U+0153 at most sizes. (In reply to comment #2) > This was done on purpose. Some other fonts do that too. It's a way to avoid > confusion between æ U+00E6 and œ U+0153 at most sizes. Since noone is going to use a propotional italic font as a terminal font this argument is not really valid. I have seen this kind of "intentional design flaw" on bitmapped fixed width fonts at low resolutions. In some of these cases a truetype version of the same font does not have this behaviour. As it is now the font is not really useful, because every time a reader sees the italic æ the first reaction is to be annoyed that it doesn't look like it belongs to the font. This means that the message written with the font will not be communicated to the reader without this distraction. Please reconsider. Created attachment 8057 [details] ae and oe in Times New Roman Italic >> As it is now the font is not really useful, because every time a reader sees the italic æ the first reaction is to be annoyed that it doesn't look like it belongs to the font. << Until you pointed it out, I never even noticed. It is indeed very hard to distinguish these glyphs in some fonts, such as Times New Roman (attached). Created attachment 8058 [details]
DejaVu Standard ae Design
I think that DejaVu takes the right approach here. If one used the standard
design for this font, the ae and oe would be almost impossible to tell apart.
Regarding comment #4: The example you show displays a good design. Both the æ and the œ in the Times New Roman font looks fine next to the a and the o and they are clearly distinguishable. Making a similar thing for DejaVu would make the DejaVu font usable. Regarding comment #5: The sample glyph you show in the centre of this attachment would be a much better design. This æ does not look out of place when placed next to the a, and the distinction between this æ and the œ is even greater than the corresponding difference in the Times New Roman font shown in comment #4 (which is already clear). Whether or not you find the current æ to be annoying or not could be a matter of how frequently you use this letter. If you only write English or some other language where this letter is rare, you might not be annoyed by the bad design, but if you frequently read Scandinavian languages where this letter is common and is used to well designed fonts, the current glyph is really annoying. Created attachment 9465 [details]
sample of proposed design at tiny size
This is what the proposed design would look like at a tiny size.
It is true that the current ae doesn't fit well with a, but it is too easy to confuse it with oe.
Would it be fine if the proposed ae design was used with Danish, Norwegian and Icelandic? They don't use oe, so the confusion is not dangerous.
The only problem with language specific variants is that it's not very well supported by most font rendering systems. If we have a design the isn't easily confused with oe, then it can be the default. The proposed design looks much better than the current glyph. I really think it should be the default for all languages. I can't see why I should be penalised with an inferior glyph design just because I happen to write in e.g. English. OK, I made the switch in r1899. The italic ae is now used by default and the storey ae is accessible throught advanced OT features. Thanks for reporting and commenting on the bug. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.