OpenType Features - Some Experiments in Using Them
Posted: Thu Mar 08, 2007 8:54 am
On the matter of OpenType and comparing and contrasting using OpenType glyph substitution and Private Use Area codepoint allocations for ligatures and alternate glyphs, some readers might perhaps like to have a look at the following thread.
viewtopic.php?t=1654
> Again, not a lot of point working to add OpenType feature support to FontCreator unless more applications support them.
Just my dreaming and musing, but it would be great if FontCreator and Serif PagePlus from Serif, at the http://www.serif.co.uk webspace, both went OpenType capable simultaneously. Two European products, both reasonable budget priced, used together with good OpenType facilities, would allow an amazing forward burst of creativity.
I use Serif PagePlus quite a lot, often using my own fonts, to produce pdf documents which I put on the web.
For example, the following, linked from the http://www.users.globalnet.co.uk/~ngo/library.htm web page.
http://www.users.globalnet.co.uk/~ngo/cooking.PDF
Also the following, linked from the http://www.users.globalnet.co.uk/~ngo/fonts.htm web page.
http://www.users.globalnet.co.uk/~ngo/housing.PDF
> My work around using the Private Use Area can be done with the current version of FontCreator and will work even in WordPad, though I have to agree that it is not ideal because special glyphs have to be manually selected and will be .notdef glyphs if used with other fonts.
Whilst recognizing that it is not ideal, I wonder if I may suggest that the situation is not perhaps as bad as compared with OpenType as it might be thought.
Firstly, regarding "special glyphs have to be manually selected and will be .notdef glyphs if used with other fonts", the problem can be reduced in severity in many cases when using FontCreator and the Private Use Area, because several fonts can use the same set of Private Use Area mappings. For example, several of my fonts include a ct ligature glyph, mapped to U+E707, which corresponds to Alt 59143 when used with WordPad. Setting some text in one of the fonts needs the Alt 59143 operation in order to get the glyph into WordPad, yet changing font does not need for it to be entered again.
For example, the following display of the word Actually using a ct ligature, which may well show with a black rectangle in this forum.
Aually
I keyed that in WordPad using my font Sonnet to a Renaissance Lady at 72 point. I then highlighted the text and changed the font to my 10000 font. The ct ligature is mapped to U+E707 in both fonts. Actually, in addition, the Sonnet to a Renaissance Lady font has an alternate ct ligature glyph at U+E427 which is Alt 58407. However, the 10000 font returns a .notdef glyph for that codepoint.
Readers who wish to use the U+E707 mapping in their own fonts for a ct ligature are welcome to do so. There is no charge or anything like that at all. There is a list of ligatures, still valid yet needing some additions, in the web pages introduced and indexed at the http://www.users.globalnet.co.uk/~ngo/golden.htm web page and some of them and some others are used in my fonts. In particular my Quest text and Chronicle Text fonts contain many ligatures including some additions.
In fairness, the only font other than my own of which I am aware that has a ct ligature at U+E707 at present is the Code2000 font of James Kass.
http://www.code2000.net/code2000_page.htm
Secondly, regarding "special glyphs have to be manually selected and will be .notdef glyphs if used with other fonts", I am wondering what happens with InDesign if one has, using an OpenType font, used alternate glyphs and one then changes the font to another OpenType font which also has an alternate glyph for the particular character being used. Does InDesign drop back to the basic glyph when the font is changed? Ligatures may be automatic, but what about variants of ligatures? Does anyone reading this thread have access to InDesign and some OpenType fonts with alternate characters who could perhaps try some experiments on this topic and post the results here please?
Some readers might like to have a look at the following thread.
http://blogs.adobe.com/typblography/200 ... priva.html
At present I find that I have mixed feelings about Private Use Area mappings for ligatures and alternate glyphs.
It would be nice to be able to produce OpenType fonts with the ligature glyphs automatic and the alternate glyphs still keeping spell-checking and so on: yet that is not possible for me at present, due to both not having the facilities and not having the knowledge and skills.
Yet Private Use Area glyphs can be used now to produce displays, using reasonable budget software.
I do like the idea that, like a piece of metal type for a ligature being one piece of metal, there is a codepoint for a ligature glyph, as if they could be distributed in a type tray. Idea for a piece of art. A divided tray as in a type tray for handsetting metal type, with U+ expressions in the boxes, so that the e box has many copies of U+0065, the a box has not quite so many copies of U+0061 and the ct ligature box has just a few copies of U+E707 and so on. A six character U+ expression being somewhat like the shape of a piece of metal type, just to the extent of being long and thin.
Yet the unifying part of this is that producing fonts using Private Use Area code points for ligature glyphs and for alternates using FontCreator, is that the fonts are usable for producing displays now and that the artwork is stored in the font ready to be used for producing a fully functioning OpenType font if the opportunity arises in the future.
William Overington
8 March 2007
viewtopic.php?t=1654
> Again, not a lot of point working to add OpenType feature support to FontCreator unless more applications support them.
Just my dreaming and musing, but it would be great if FontCreator and Serif PagePlus from Serif, at the http://www.serif.co.uk webspace, both went OpenType capable simultaneously. Two European products, both reasonable budget priced, used together with good OpenType facilities, would allow an amazing forward burst of creativity.
I use Serif PagePlus quite a lot, often using my own fonts, to produce pdf documents which I put on the web.
For example, the following, linked from the http://www.users.globalnet.co.uk/~ngo/library.htm web page.
http://www.users.globalnet.co.uk/~ngo/cooking.PDF
Also the following, linked from the http://www.users.globalnet.co.uk/~ngo/fonts.htm web page.
http://www.users.globalnet.co.uk/~ngo/housing.PDF
> My work around using the Private Use Area can be done with the current version of FontCreator and will work even in WordPad, though I have to agree that it is not ideal because special glyphs have to be manually selected and will be .notdef glyphs if used with other fonts.
Whilst recognizing that it is not ideal, I wonder if I may suggest that the situation is not perhaps as bad as compared with OpenType as it might be thought.
Firstly, regarding "special glyphs have to be manually selected and will be .notdef glyphs if used with other fonts", the problem can be reduced in severity in many cases when using FontCreator and the Private Use Area, because several fonts can use the same set of Private Use Area mappings. For example, several of my fonts include a ct ligature glyph, mapped to U+E707, which corresponds to Alt 59143 when used with WordPad. Setting some text in one of the fonts needs the Alt 59143 operation in order to get the glyph into WordPad, yet changing font does not need for it to be entered again.
For example, the following display of the word Actually using a ct ligature, which may well show with a black rectangle in this forum.
Aually
I keyed that in WordPad using my font Sonnet to a Renaissance Lady at 72 point. I then highlighted the text and changed the font to my 10000 font. The ct ligature is mapped to U+E707 in both fonts. Actually, in addition, the Sonnet to a Renaissance Lady font has an alternate ct ligature glyph at U+E427 which is Alt 58407. However, the 10000 font returns a .notdef glyph for that codepoint.
Readers who wish to use the U+E707 mapping in their own fonts for a ct ligature are welcome to do so. There is no charge or anything like that at all. There is a list of ligatures, still valid yet needing some additions, in the web pages introduced and indexed at the http://www.users.globalnet.co.uk/~ngo/golden.htm web page and some of them and some others are used in my fonts. In particular my Quest text and Chronicle Text fonts contain many ligatures including some additions.
In fairness, the only font other than my own of which I am aware that has a ct ligature at U+E707 at present is the Code2000 font of James Kass.
http://www.code2000.net/code2000_page.htm
Secondly, regarding "special glyphs have to be manually selected and will be .notdef glyphs if used with other fonts", I am wondering what happens with InDesign if one has, using an OpenType font, used alternate glyphs and one then changes the font to another OpenType font which also has an alternate glyph for the particular character being used. Does InDesign drop back to the basic glyph when the font is changed? Ligatures may be automatic, but what about variants of ligatures? Does anyone reading this thread have access to InDesign and some OpenType fonts with alternate characters who could perhaps try some experiments on this topic and post the results here please?
Some readers might like to have a look at the following thread.
http://blogs.adobe.com/typblography/200 ... priva.html
At present I find that I have mixed feelings about Private Use Area mappings for ligatures and alternate glyphs.
It would be nice to be able to produce OpenType fonts with the ligature glyphs automatic and the alternate glyphs still keeping spell-checking and so on: yet that is not possible for me at present, due to both not having the facilities and not having the knowledge and skills.
Yet Private Use Area glyphs can be used now to produce displays, using reasonable budget software.
I do like the idea that, like a piece of metal type for a ligature being one piece of metal, there is a codepoint for a ligature glyph, as if they could be distributed in a type tray. Idea for a piece of art. A divided tray as in a type tray for handsetting metal type, with U+ expressions in the boxes, so that the e box has many copies of U+0065, the a box has not quite so many copies of U+0061 and the ct ligature box has just a few copies of U+E707 and so on. A six character U+ expression being somewhat like the shape of a piece of metal type, just to the extent of being long and thin.
Yet the unifying part of this is that producing fonts using Private Use Area code points for ligature glyphs and for alternates using FontCreator, is that the fonts are usable for producing displays now and that the artwork is stored in the font ready to be used for producing a fully functioning OpenType font if the opportunity arises in the future.
William Overington
8 March 2007