[fontforge-users] What encoding should I choose?

[Ultra]

http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455.html
<http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455.html>
will give some context to my question.
There sure are a lot of them. I'm not sure what to tell you about my needs,
since I don't know the strengths of weaknesses of the various encodings. I
want to release this to a wide public, so I guess that means TrueType's
compatibility. But there is more than 1 type of TrueType. And what's the
"bitmap encoding"? Of course, there are also a ton of other options.



--
View this message in context: http://fontforge.10959.n7.nabble.com/What-encoding-should-I-choose-tp15491.html
Sent from the User mailing list archive at Nabble.com.

[s***@att.net]

An "encoding" determines the set of characters that are afforded
by a font, and the ordering in which they appear in the font so
the software can get the chars it wants without having to do
lookups. The International Standards Organisation (ISO) defines
many standard encodings, the most widely used for representing
English being ISO1859-1. There are also non-ISO encodings that
came into being before ISO standardisation and are still widely
used for certain purposes.

Equally important is the file format you choose, because it
determines how many encodings you can choose from.

The most comprehensive is OTF (.otf and also, strangely, .ttf)
because it allows more than 65 000 chars in one font , but only
the lower chars are accessable to all software without some extra
effort (e.g., using HTML notation such as &#12345; to put the
char living in cell 12345 out onto the screen/printed page).
Software that lets you work in Unicode will typically have tools
to (somewhat) mask the way higher-order cells such as 12345, out
of single-byte reach, are addressed.

T1, PS Type 1 (Postscript-based, .pfb, .pfm, .afm) is the
original, and based on the postscript language's bezier curves,
which are highly accurate. They are limited to 255 chars.

True Type (.ttf) came out after T1 and before OTF. It offers the
same set of chars as T1 does, but uses quadratic curves, which
are more limited than beziers. The limitations are only usually
visible in an editor. FF masks those limitations by allowing you
to work with bezier curves and then translating them
automagically when you create a .ttf font. Like T1, TTF fonts
have space for only 255 chars.

WOFF (.woff) is, I think, the newest type, meant to be
downloadable with the web pages that use it so that end-user
systems can see the pages as the designer meant them to be seen.
It's a simpler file format than .ttf et al, and (I think) only
stores the cells with a glyph defined in them, and uses a lookup
table to translate the nominal cell number to the actual one.

SVG (.svg) is still mainly Unix-based, though SVG (Scalable
Vector Graphics) is a general representation language as well as
a file format, and is like postscript in that one can actually
write programs in it by hand, though most people use a wysiwyg
editor that will generate the svg code for them.

I don't know what "bitmap encoding" might be, but it sounds like
a term for a bitmapped font (.fon), which like the others is a
file format. There might be only one encoding that's ever used
(I've never checked). Bitmapped chars are represented by actual
bitmaps / pixel maps. Pixel maps don't really re-size except in
a very clunky way because they do it by doubling or discarding
pixels depending on whether you're making them bigger or smaller.
Rendering software turns vector fonts (T1, TT, OTF, WOFF, SVG),
into bitmaps/pixel maps on the screen/page on the fly, and
because they're really little programs, they can do nifty stuff
like decide where individual pixels should go to produce the best
visual effect.

I hope that helps.

Post by Ultra
http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455.html
<http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455.html>
will give some context to my question.
There sure are a lot of them. I'm not sure what to tell you about my needs,
since I don't know the strengths of weaknesses of the various encodings. I
want to release this to a wide public, so I guess that means TrueType's
compatibility. But there is more than 1 type of TrueType. And what's the
"bitmap encoding"? Of course, there are also a ton of other options.

[Jose Da Silva]

Hi ***@att.net,
ISO1859-1 is the most common, and probably/perhaps the best choice.
If you look at the world around you, many countries are taking English as a
second language, and even if there are a couple of paces determined to
enforce their main language, the rest of the world moves-on leaving them
behind. In terms of the technology fields, if you want to get stuff done,
english is the preferred language in terms of computing, programming,
likewise electronics, and various other fields (for example, international
flight pilots need to know english, otherwise they run into difficulty at many
airports).
If you're looking towards the future, you're probably best using ISO1859-1.

In terms of file formats, all those file formats lose some information in one
place or another. If your intention is to keep as much info as possible,
why not maintain the sfd format? It is also a registered format
/* Officially registered with IANA on 14 May 2008 */
mime type: application/vnd.font-fontforge-sfd
https://www.google.ca/search?q=sfd+file

Post by s***@att.net
An "encoding" determines the set of characters that are afforded
by a font, and the ordering in which they appear in the font so
the software can get the chars it wants without having to do
lookups. The International Standards Organisation (ISO) defines
many standard encodings, the most widely used for representing
English being ISO1859-1. There are also non-ISO encodings that
came into being before ISO standardisation and are still widely
used for certain purposes.
Equally important is the file format you choose, because it
determines how many encodings you can choose from.
The most comprehensive is OTF (.otf and also, strangely, .ttf)
because it allows more than 65 000 chars in one font , but only
the lower chars are accessable to all software without some extra
effort (e.g., using HTML notation such as &#12345; to put the
char living in cell 12345 out onto the screen/printed page).
Software that lets you work in Unicode will typically have tools
to (somewhat) mask the way higher-order cells such as 12345, out
of single-byte reach, are addressed.
T1, PS Type 1 (Postscript-based, .pfb, .pfm, .afm) is the
original, and based on the postscript language's bezier curves,
which are highly accurate. They are limited to 255 chars.
True Type (.ttf) came out after T1 and before OTF. It offers the
same set of chars as T1 does, but uses quadratic curves, which
are more limited than beziers. The limitations are only usually
visible in an editor. FF masks those limitations by allowing you
to work with bezier curves and then translating them
automagically when you create a .ttf font. Like T1, TTF fonts
have space for only 255 chars.
WOFF (.woff) is, I think, the newest type, meant to be
downloadable with the web pages that use it so that end-user
systems can see the pages as the designer meant them to be seen.
It's a simpler file format than .ttf et al, and (I think) only
stores the cells with a glyph defined in them, and uses a lookup
table to translate the nominal cell number to the actual one.
SVG (.svg) is still mainly Unix-based, though SVG (Scalable
Vector Graphics) is a general representation language as well as
a file format, and is like postscript in that one can actually
write programs in it by hand, though most people use a wysiwyg
editor that will generate the svg code for them.
I don't know what "bitmap encoding" might be, but it sounds like
a term for a bitmapped font (.fon), which like the others is a
file format. There might be only one encoding that's ever used
(I've never checked). Bitmapped chars are represented by actual
bitmaps / pixel maps. Pixel maps don't really re-size except in
a very clunky way because they do it by doubling or discarding
pixels depending on whether you're making them bigger or smaller.
Rendering software turns vector fonts (T1, TT, OTF, WOFF, SVG),
into bitmaps/pixel maps on the screen/page on the fly, and
because they're really little programs, they can do nifty stuff
like decide where individual pixels should go to produce the best
visual effect.
I hope that helps.

Post by Ultra
http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455.
html
<http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td1545
5.html> will give some context to my question.
There sure are a lot of them. I'm not sure what to tell you about my
needs, since I don't know the strengths of weaknesses of the various
encodings. I want to release this to a wide public, so I guess that
means TrueType's compatibility. But there is more than 1 type of
TrueType. And what's the "bitmap encoding"? Of course, there are also
a ton of other options.

------------------------------------------------------------------------
------ Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

[Martin B. Brilliant]

Don’t we really mean ISO 8859-1? Or am I missing something?

Post by Jose Da Silva
ISO1859-1 is the most common, and probably/perhaps the best choice.
If you look at the world around you, many countries are taking English as a
second language, and even if there are a couple of paces determined to
enforce their main language, the rest of the world moves-on leaving them
behind. In terms of the technology fields, if you want to get stuff done,
english is the preferred language in terms of computing, programming,
likewise electronics, and various other fields (for example, international
flight pilots need to know english, otherwise they run into difficulty at many
airports).
If you're looking towards the future, you're probably best using ISO1859-1.
In terms of file formats, all those file formats lose some information in one
place or another. If your intention is to keep as much info as possible,
why not maintain the sfd format? It is also a registered format
/* Officially registered with IANA on 14 May 2008 */
mime type: application/vnd.font-fontforge-sfd
https://www.google.ca/search?q=sfd+file

Post by s***@att.net
An "encoding" determines the set of characters that are afforded
by a font, and the ordering in which they appear in the font so
the software can get the chars it wants without having to do
lookups. The International Standards Organisation (ISO) defines
many standard encodings, the most widely used for representing
English being ISO1859-1. There are also non-ISO encodings that
came into being before ISO standardisation and are still widely
used for certain purposes.
Equally important is the file format you choose, because it
determines how many encodings you can choose from.
The most comprehensive is OTF (.otf and also, strangely, .ttf)
because it allows more than 65 000 chars in one font , but only
the lower chars are accessable to all software without some extra
effort (e.g., using HTML notation such as &#12345; to put the
char living in cell 12345 out onto the screen/printed page).
Software that lets you work in Unicode will typically have tools
to (somewhat) mask the way higher-order cells such as 12345, out
of single-byte reach, are addressed.
T1, PS Type 1 (Postscript-based, .pfb, .pfm, .afm) is the
original, and based on the postscript language's bezier curves,
which are highly accurate. They are limited to 255 chars.
True Type (.ttf) came out after T1 and before OTF. It offers the
same set of chars as T1 does, but uses quadratic curves, which
are more limited than beziers. The limitations are only usually
visible in an editor. FF masks those limitations by allowing you
to work with bezier curves and then translating them
automagically when you create a .ttf font. Like T1, TTF fonts
have space for only 255 chars.
WOFF (.woff) is, I think, the newest type, meant to be
downloadable with the web pages that use it so that end-user
systems can see the pages as the designer meant them to be seen.
It's a simpler file format than .ttf et al, and (I think) only
stores the cells with a glyph defined in them, and uses a lookup
table to translate the nominal cell number to the actual one.
SVG (.svg) is still mainly Unix-based, though SVG (Scalable
Vector Graphics) is a general representation language as well as
a file format, and is like postscript in that one can actually
write programs in it by hand, though most people use a wysiwyg
editor that will generate the svg code for them.
I don't know what "bitmap encoding" might be, but it sounds like
a term for a bitmapped font (.fon), which like the others is a
file format. There might be only one encoding that's ever used
(I've never checked). Bitmapped chars are represented by actual
bitmaps / pixel maps. Pixel maps don't really re-size except in
a very clunky way because they do it by doubling or discarding
pixels depending on whether you're making them bigger or smaller.
Rendering software turns vector fonts (T1, TT, OTF, WOFF, SVG),
into bitmaps/pixel maps on the screen/page on the fly, and
because they're really little programs, they can do nifty stuff
like decide where individual pixels should go to produce the best
visual effect.
I hope that helps.

Post by Ultra
http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455.
html
<http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td1545
5.html> will give some context to my question.
There sure are a lot of them. I'm not sure what to tell you about my
needs, since I don't know the strengths of weaknesses of the various
encodings. I want to release this to a wide public, so I guess that
means TrueType's compatibility. But there is more than 1 type of
TrueType. And what's the "bitmap encoding"? Of course, there are also
a ton of other options.

------------------------------------------------------------------------
------ Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

[s***@att.net]

ooops, you're right, I knew it didn't look quite right when I
wrote it, but thought it was just me having a "senior moment".
ISO 8859-1 it is!

Thanks for the catch!

Don’t we really mean ISO 8859-1? Or am I missing something?

Post by Jose Da Silva
ISO1859-1 is the most common, and probably/perhaps the best choice.
If you look at the world around you, many countries are taking English as a
second language, and even if there are a couple of paces determined to
enforce their main language, the rest of the world moves-on leaving them
behind. In terms of the technology fields, if you want to get stuff done,
english is the preferred language in terms of computing, programming,
likewise electronics, and various other fields (for example, international
flight pilots need to know english, otherwise they run into difficulty at many
airports).
If you're looking towards the future, you're probably best using ISO1859-1.
In terms of file formats, all those file formats lose some information in one
place or another. If your intention is to keep as much info as possible,
why not maintain the sfd format? It is also a registered format
/* Officially registered with IANA on 14 May 2008 */
mime type: application/vnd.font-fontforge-sfd
https://www.google.ca/search?q=sfd+file

Post by s***@att.net
An "encoding" determines the set of characters that are afforded
by a font, and the ordering in which they appear in the font so
the software can get the chars it wants without having to do
lookups. The International Standards Organisation (ISO) defines
many standard encodings, the most widely used for representing
English being ISO1859-1. There are also non-ISO encodings that
came into being before ISO standardisation and are still widely
used for certain purposes.
Equally important is the file format you choose, because it
determines how many encodings you can choose from.
The most comprehensive is OTF (.otf and also, strangely, .ttf)
because it allows more than 65 000 chars in one font , but only
the lower chars are accessable to all software without some extra
effort (e.g., using HTML notation such as &#12345; to put the
char living in cell 12345 out onto the screen/printed page).
Software that lets you work in Unicode will typically have tools
to (somewhat) mask the way higher-order cells such as 12345, out
of single-byte reach, are addressed.
T1, PS Type 1 (Postscript-based, .pfb, .pfm, .afm) is the
original, and based on the postscript language's bezier curves,
which are highly accurate. They are limited to 255 chars.
True Type (.ttf) came out after T1 and before OTF. It offers the
same set of chars as T1 does, but uses quadratic curves, which
are more limited than beziers. The limitations are only usually
visible in an editor. FF masks those limitations by allowing you
to work with bezier curves and then translating them
automagically when you create a .ttf font. Like T1, TTF fonts
have space for only 255 chars.
WOFF (.woff) is, I think, the newest type, meant to be
downloadable with the web pages that use it so that end-user
systems can see the pages as the designer meant them to be seen.
It's a simpler file format than .ttf et al, and (I think) only
stores the cells with a glyph defined in them, and uses a lookup
table to translate the nominal cell number to the actual one.
SVG (.svg) is still mainly Unix-based, though SVG (Scalable
Vector Graphics) is a general representation language as well as
a file format, and is like postscript in that one can actually
write programs in it by hand, though most people use a wysiwyg
editor that will generate the svg code for them.
I don't know what "bitmap encoding" might be, but it sounds like
a term for a bitmapped font (.fon), which like the others is a
file format. There might be only one encoding that's ever used
(I've never checked). Bitmapped chars are represented by actual
bitmaps / pixel maps. Pixel maps don't really re-size except in
a very clunky way because they do it by doubling or discarding
pixels depending on whether you're making them bigger or smaller.
Rendering software turns vector fonts (T1, TT, OTF, WOFF, SVG),
into bitmaps/pixel maps on the screen/page on the fly, and
because they're really little programs, they can do nifty stuff
like decide where individual pixels should go to produce the best
visual effect.
I hope that helps.

Post by Ultra
http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455.
html
<http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td1545
5.html> will give some context to my question.
There sure are a lot of them. I'm not sure what to tell you about my
needs, since I don't know the strengths of weaknesses of the various
encodings. I want to release this to a wide public, so I guess that
means TrueType's compatibility. But there is more than 1 type of
TrueType. And what's the "bitmap encoding"? Of course, there are also
a ton of other options.

------------------------------------------------------------------------
------ Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

[s***@att.net]

Post by Jose Da Silva
In terms of file formats, all those file formats lose some information in one
place or another. If your intention is to keep as much info as possible,
why not maintain the sfd format? It is also a registered format
/* Officially registered with IANA on 14 May 2008 */
mime type: application/vnd.font-fontforge-sfd
https://www.google.ca/search?q=sfd+file

I wasn't aware that the intermediate .sfd format can be used for
output. Are you sure about that? IANA mime registration is
good for being able to transfer files without getting them
stepped on as they go, and for keeping the wrong interpreters
from trying to interpret them, but I've never heard of any other
uses for it.

[Jose Da Silva]

Hi scratch65535,
If you are trying to hold onto as much information as possible, I would
choose to hold the toolbox instead of some individual tools. Granted, sfd
is not a commonly known file format, but it is "the" toolbox :-)

What you can do, is put the sfd file (the Toolbox), plus a text based
README.TXT help file, and maybe an output resultant .otf .utf or other
(exported fonts), into a zip file.

You indicated ISO-8859-1, and the more I thought about it, you may probably
want to follow the unicode format since it includes the entire range of
characters, and it's got several supporters (see unicode.org).
Unicode 10 will be coming out shortly (June/July maybe), and it is far
larger than 65535 codes.

From what I've seen in FontForge, there is some backend stuff that uses the

unicode range, and I think there are improvements in the Unicode
subdirectory that can be added to take advantage here (making them more
like sub-lists instead of independent lists).

Getting back to reading about 8859-1, there are a few versions too.
The latest version is referred to as ISO-8859-1 (two dashes), and it is
sometimes confused with windows-1252, as a matter of fact, htm has to
interchange the terms 8859-1 and windows-1252 mislabelling. Probably worth
reading:
https://en.wikipedia.org/wiki/ISO/IEC_8859-1

Another format is this, but...:
https://en.wikipedia.org/wiki/ISO/IEC_8859-15
"There were attempts to make ISO-8859-15 the default character set for 8-
bit communication, but it was never able to supplant the popular
ISO-8859-1."

Maybe not the best answer, but these are somethings to think about,
Joe

Post by Jose Da Silva
In terms of file formats, all those file formats lose some information
in one place or another. If your intention is to keep as much info as
possible, why not maintain the sfd format? It is also a registered
format /* Officially registered with IANA on 14 May 2008 */
mime type: application/vnd.font-fontforge-sfd
https://www.google.ca/search?q=sfd+file

I wasn't aware that the intermediate .sfd format can be used for
output. Are you sure about that? IANA mime registration is
good for being able to transfer files without getting them
stepped on as they go, and for keeping the wrong interpreters
from trying to interpret them, but I've never heard of any other
uses for it.
------------------------------------------------------------------------
------ Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

[Ultra]

Well, that was comprehensive! Did you spend all 2 weeks working on it? :)
Actually, I need to do it in Unicode. Right now, only the BMP, but I'm
hoping that one day some things in the private use area will be incorporated
into the SMP. But I can cross that bridge when I get to it.
I'm pretty sure that even with all the references, those 31 original
characters do not blow a 255 budget. But I'm concerned about that remark to
make about lower case chars. Most of the chars are in the upper case region.
What more do I need to do?
I don't know how much I care about precision, but what about compatibility?
Will they all work on all platforms or are some more universal than others?

n

Post by s***@att.net
An "encoding" determines the set of characters that are afforded
by a font, and the ordering in which they appear in the font so
the software can get the chars it wants without having to do
lookups. The International Standards Organisation (ISO) defines
many standard encodings, the most widely used for representing
English being ISO1859-1. There are also non-ISO encodings that
came into being before ISO standardisation and are still widely
used for certain purposes.
Equally important is the file format you choose, because it
determines how many encodings you can choose from.
The most comprehensive is OTF (.otf and also, strangely, .ttf)
because it allows more than 65 000 chars in one font , but only
the lower chars are accessable to all software without some extra
effort (e.g., using HTML notation such as &#12345; to put the
char living in cell 12345 out onto the screen/printed page).
Software that lets you work in Unicode will typically have tools
to (somewhat) mask the way higher-order cells such as 12345, out
of single-byte reach, are addressed.
T1, PS Type 1 (Postscript-based, .pfb, .pfm, .afm) is the
original, and based on the postscript language's bezier curves,
which are highly accurate. They are limited to 255 chars.
True Type (.ttf) came out after T1 and before OTF. It offers the
same set of chars as T1 does, but uses quadratic curves, which
are more limited than beziers. The limitations are only usually
visible in an editor. FF masks those limitations by allowing you
to work with bezier curves and then translating them
automagically when you create a .ttf font. Like T1, TTF fonts
have space for only 255 chars.
WOFF (.woff) is, I think, the newest type, meant to be
downloadable with the web pages that use it so that end-user
systems can see the pages as the designer meant them to be seen.
It's a simpler file format than .ttf et al, and (I think) only
stores the cells with a glyph defined in them, and uses a lookup
table to translate the nominal cell number to the actual one.
SVG (.svg) is still mainly Unix-based, though SVG (Scalable
Vector Graphics) is a general representation language as well as
a file format, and is like postscript in that one can actually
write programs in it by hand, though most people use a wysiwyg
editor that will generate the svg code for them.
I don't know what "bitmap encoding" might be, but it sounds like
a term for a bitmapped font (.fon), which like the others is a
file format. There might be only one encoding that's ever used
(I've never checked). Bitmapped chars are represented by actual
bitmaps / pixel maps. Pixel maps don't really re-size except in
a very clunky way because they do it by doubling or discarding
pixels depending on whether you're making them bigger or smaller.
Rendering software turns vector fonts (T1, TT, OTF, WOFF, SVG),
into bitmaps/pixel maps on the screen/page on the fly, and
because they're really little programs, they can do nifty stuff
like decide where individual pixels should go to produce the best
visual effect.
I hope that helps.

Post by Ultra
http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455.html
will give some context to my question.
There sure are a lot of them. I'm not sure what to tell you about my needs,
since I don't know the strengths of weaknesses of the various encodings. I
want to release this to a wide public, so I guess that means TrueType's
compatibility. But there is more than 1 type of TrueType. And what's the
"bitmap encoding"? Of course, there are also a ton of other options.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

--
View this message in context: http://fontforge.10959.n7.nabble.com/What-encoding-should-I-choose-tp15491p15519.html
Sent from the User mailing list archive at Nabble.com.

[Jose Da Silva]

FYI guys,
nabble is a secondary copy of the sourceforge list.
It copies lists, but what's shown there won't come back into the main
sourceforge list unless someone takes the time to copy-n-paste it back into
the sourceforge list.
...example this discussion below is incomplete here and leaves us readers
puzzled.... it was probably better left back there in nabble as it's
incomplete here.

...or maybe to re-quote a vegas term...
what happens in nabble stays in nabble

...if you want it seen in the main user list, it's here:
https://lists.sourceforge.net/lists/listinfo/fontforge-users

Post by Ultra
Well, that was comprehensive! Did you spend all 2 weeks working on it?
:) Actually, I need to do it in Unicode. Right now, only the BMP, but
I'm hoping that one day some things in the private use area will be
incorporated into the SMP. But I can cross that bridge when I get to
it.
I'm pretty sure that even with all the references, those 31 original
characters do not blow a 255 budget. But I'm concerned about that remark
to make about lower case chars. Most of the chars are in the upper case
region. What more do I need to do?
I don't know how much I care about precision, but what about
compatibility? Will they all work on all platforms or are some more
universal than others?
n

Post by s***@att.net
An "encoding" determines the set of characters that are afforded
by a font, and the ordering in which they appear in the font so
the software can get the chars it wants without having to do
lookups. The International Standards Organisation (ISO) defines
many standard encodings, the most widely used for representing
English being ISO1859-1. There are also non-ISO encodings that
came into being before ISO standardisation and are still widely
used for certain purposes.
Equally important is the file format you choose, because it
determines how many encodings you can choose from.
The most comprehensive is OTF (.otf and also, strangely, .ttf)
because it allows more than 65 000 chars in one font , but only
the lower chars are accessable to all software without some extra
effort (e.g., using HTML notation such as &#12345; to put the
char living in cell 12345 out onto the screen/printed page).
Software that lets you work in Unicode will typically have tools
to (somewhat) mask the way higher-order cells such as 12345, out
of single-byte reach, are addressed.
T1, PS Type 1 (Postscript-based, .pfb, .pfm, .afm) is the
original, and based on the postscript language's bezier curves,
which are highly accurate. They are limited to 255 chars.
True Type (.ttf) came out after T1 and before OTF. It offers the
same set of chars as T1 does, but uses quadratic curves, which
are more limited than beziers. The limitations are only usually
visible in an editor. FF masks those limitations by allowing you
to work with bezier curves and then translating them
automagically when you create a .ttf font. Like T1, TTF fonts
have space for only 255 chars.
WOFF (.woff) is, I think, the newest type, meant to be
downloadable with the web pages that use it so that end-user
systems can see the pages as the designer meant them to be seen.
It's a simpler file format than .ttf et al, and (I think) only
stores the cells with a glyph defined in them, and uses a lookup
table to translate the nominal cell number to the actual one.
SVG (.svg) is still mainly Unix-based, though SVG (Scalable
Vector Graphics) is a general representation language as well as
a file format, and is like postscript in that one can actually
write programs in it by hand, though most people use a wysiwyg
editor that will generate the svg code for them.
I don't know what "bitmap encoding" might be, but it sounds like
a term for a bitmapped font (.fon), which like the others is a
file format. There might be only one encoding that's ever used
(I've never checked). Bitmapped chars are represented by actual
bitmaps / pixel maps. Pixel maps don't really re-size except in
a very clunky way because they do it by doubling or discarding
pixels depending on whether you're making them bigger or smaller.
Rendering software turns vector fonts (T1, TT, OTF, WOFF, SVG),
into bitmaps/pixel maps on the screen/page on the fly, and
because they're really little programs, they can do nifty stuff
like decide where individual pixels should go to produce the best
visual effect.
I hope that helps.

Post by Ultra
http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455
.html
&lt;http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td
There sure are a lot of them. I'm not sure what to tell you about my

needs,

Post by s***@att.net

Post by Ultra
since I don't know the strengths of weaknesses of the various
encodings. I want to release this to a wide public, so I guess that
means TrueType's compatibility. But there is more than 1 type of
TrueType. And what's the "bitmap encoding"? Of course, there are also
a ton of other options.

[Martin B. Brilliant]

The last time I used nabble to send something, I got a warning that it would not go to the list unless I was registered on the list. I am, and it did. Nabble is also delayed. Jose Da Silva’s message has not yet appeared on nabble, and Ultra’s message is datelined almost a day later than it is below.

Apparently anything that could be posted to the list directly can be posted to the list through nabble. Perhaps the missing links were posted by people who are not registered on the list.

There are, however, more messages on this topic in the list archive than there are below. Maybe there’s a problem in the way people reply. The archive has everything (except delayed messages from nabble).

Post by Jose Da Silva
FYI guys,
nabble is a secondary copy of the sourceforge list.
It copies lists, but what's shown there won't come back into the main
sourceforge list unless someone takes the time to copy-n-paste it back into
the sourceforge list.
...example this discussion below is incomplete here and leaves us readers
puzzled.... it was probably better left back there in nabble as it's
incomplete here.
...or maybe to re-quote a vegas term...
what happens in nabble stays in nabble
https://lists.sourceforge.net/lists/listinfo/fontforge-users

Post by Ultra
Well, that was comprehensive! Did you spend all 2 weeks working on it?
:) Actually, I need to do it in Unicode. Right now, only the BMP, but
I'm hoping that one day some things in the private use area will be
incorporated into the SMP. But I can cross that bridge when I get to
it.
I'm pretty sure that even with all the references, those 31 original
characters do not blow a 255 budget. But I'm concerned about that remark
to make about lower case chars. Most of the chars are in the upper case
region. What more do I need to do?
I don't know how much I care about precision, but what about
compatibility? Will they all work on all platforms or are some more
universal than others?
n

Post by s***@att.net
An "encoding" determines the set of characters that are afforded
by a font, and the ordering in which they appear in the font so
the software can get the chars it wants without having to do
lookups. The International Standards Organisation (ISO) defines
many standard encodings, the most widely used for representing
English being ISO1859-1. There are also non-ISO encodings that
came into being before ISO standardisation and are still widely
used for certain purposes.
Equally important is the file format you choose, because it
determines how many encodings you can choose from.
The most comprehensive is OTF (.otf and also, strangely, .ttf)
because it allows more than 65 000 chars in one font , but only
the lower chars are accessable to all software without some extra
effort (e.g., using HTML notation such as &#12345; to put the
char living in cell 12345 out onto the screen/printed page).
Software that lets you work in Unicode will typically have tools
to (somewhat) mask the way higher-order cells such as 12345, out
of single-byte reach, are addressed.
T1, PS Type 1 (Postscript-based, .pfb, .pfm, .afm) is the
original, and based on the postscript language's bezier curves,
which are highly accurate. They are limited to 255 chars.
True Type (.ttf) came out after T1 and before OTF. It offers the
same set of chars as T1 does, but uses quadratic curves, which
are more limited than beziers. The limitations are only usually
visible in an editor. FF masks those limitations by allowing you
to work with bezier curves and then translating them
automagically when you create a .ttf font. Like T1, TTF fonts
have space for only 255 chars.
WOFF (.woff) is, I think, the newest type, meant to be
downloadable with the web pages that use it so that end-user
systems can see the pages as the designer meant them to be seen.
It's a simpler file format than .ttf et al, and (I think) only
stores the cells with a glyph defined in them, and uses a lookup
table to translate the nominal cell number to the actual one.
SVG (.svg) is still mainly Unix-based, though SVG (Scalable
Vector Graphics) is a general representation language as well as
a file format, and is like postscript in that one can actually
write programs in it by hand, though most people use a wysiwyg
editor that will generate the svg code for them.
I don't know what "bitmap encoding" might be, but it sounds like
a term for a bitmapped font (.fon), which like the others is a
file format. There might be only one encoding that's ever used
(I've never checked). Bitmapped chars are represented by actual
bitmaps / pixel maps. Pixel maps don't really re-size except in
a very clunky way because they do it by doubling or discarding
pixels depending on whether you're making them bigger or smaller.
Rendering software turns vector fonts (T1, TT, OTF, WOFF, SVG),
into bitmaps/pixel maps on the screen/page on the fly, and
because they're really little programs, they can do nifty stuff
like decide where individual pixels should go to produce the best
visual effect.
I hope that helps.

Post by Ultra
http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td15455
.html
&lt;http://fontforge.10959.n7.nabble.com/designing-wrt-the-borders-td
There sure are a lot of them. I'm not sure what to tell you about my

needs,

Post by s***@att.net

Post by Ultra
since I don't know the strengths of weaknesses of the various
encodings. I want to release this to a wide public, so I guess that
means TrueType's compatibility. But there is more than 1 type of
TrueType. And what's the "bitmap encoding"? Of course, there are also
a ton of other options.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
fontforge-users mailing list
https://lists.sourceforge.net/lists/listinfo/fontforge-users
http://fontforge.10959.n7.nabble.com/User-f8781.html

[s***@att.net]

Post by Ultra
Well, that was comprehensive! Did you spend all 2 weeks working on it? :)
Actually, I need to do it in Unicode.

Maybe not. If you're only doing 31 chars, there's no reason to
use unicode unless you also want to be able to use Latin chars
without switching fonts. You can create a vanilla .ttf font
where your glyphs take the places of the usual Latin ones, i.e.
the glyph for the "a" sound in the "a" position, for the "b"
sound in the "b" position, etc. That would make it easy to type
English (or other) words but have them come out in your special
alphabet. That's essentially how the Pigpen dot cypher (q.v.)
works: the symbols map 1:1 to the letters they represent.

Post by Ultra
Right now, only the BMP, but I'm
hoping that one day some things in the private use area will be incorporated
into the SMP. But I can cross that bridge when I get to it.
I'm pretty sure that even with all the references, those 31 original
characters do not blow a 255 budget. But I'm concerned about that remark to
make about lower case chars. Most of the chars are in the upper case region.
What more do I need to do?

If you create a transliterating font, as I suggested above, you
needn't worry about case very much. Put as many of your chars as
will fit in the lower-case cells of the font (because lower case
doesn't require the user to engage the caps lock), and then put
the remainder in whichever upper-case cells seem like the most
easily remembered.

Post by Ultra
I don't know how much I care about precision, but what about compatibility?
Will they all work on all platforms or are some more universal than others?

True Type is currently your best bet for universality. Pretty
much all software understands how to use fonts with that format.