On Tuesday 15 October 2002 09:29 am, Jody Harris wrote:

On Tue, 2002-10-15 at 06:04, Steven T. Hatton wrote:

...

Providing the artisticly talented people with the tools to handle the technology involved should make the transition from font follower to font leader a reality.

I'm not sure where MS got all of it's fonts, but I know that *great* fonts are not cheap not easily produced.

Well, when you have a trillion dollars, you can get just about anything you want.

Have you priced any Adobe Postscript fonts lately?

Yes. I was looking at them today. I have a lot of good fonts from Word Perfect. I might be able to get them to work for my purposes if I understood font management a bit better. I also have whatever is on M$ Office and XP. I'm sure I have some other fonts lying around here somewhere. One big problem I'm having is getting the 'correct' set of fonts in place. I believe I finally cracked the code on why Mathematica is so hooked on Helvetica. It seems to be the most popular font for Mac. Looking over the WRI files, I get the feeling they develop for Mac and then prot to other systems.

TrueType fonts take a lot of shortcuts and still come out as good fonts, but even those can't be cheap to produce.

Years ago, I got a copy of MacroMedia's Fontographer (Windows/Mac). I thought it would be cool and fun to make my own fonts. After just a little reading, my head started to swim! Fontographer was so popular, that I don't think it's been updated in four years.

Fonts *shouldn't* be all *that* hard to create. The way I see it, there is 13-dimensional configuration space where each dimension represents one of the parameters in the font string. For example: -b&h-lucida-medium-r-normal-sans-8-80-75-75-p-45-iso8859-9 The last field should really be UTF. We need to make that switch. It's just too insane to continue with this heterogeneous collection. (OTOH, I know of some problems with UTF, which may be harder to correct than one might think. ) So, assuming we take UTF as a constant for the encoding, We now have 12 variables to deal with from the font identifier. Then we have the character access. That is the range of characters along the UTF number line. What a calligrapher(?) would typically work on would be letters of a particular alphabet, and a specific font family. Since thoes are fixed during the development, the person creating the fonts doesn't need to think about that. The variables which remain are those dealing with weight, variant, size, serif, or sans serif, etc. The size variation doesn't requier a complete reproduction for every different size, just for the special cases where the fonts get very small, and some creative illusions need to be implemented. So the ideal tool would present the artist with a template into which he or she enumerates the characters to be represented. There would then be a row for each possible combination of variables other than size. All the artist should have to do, is select a particular cell in this grid, and create the image representing that particular character in the given variant, weight, etc. An easily help system with definitions of the variable which define the location in the grid should be available. The vision I have is a grid with individual cells. Each cell hold one glyph. Each glyph, when clicked, will pop up in a palette similar to KIconEdit.

Now, all that being said, I don't think creating a great font is all that much more difficult than creating a powerful, free, stable, multi-user, multitasking operating system, and the free software community has not only managed to do that, but done it better than most of the commercial Unix companies were able to do with millions of dollars and a 30 year lead.

To be fair, the comercial Unix companies have contributed a great deal to Linux. There's been a lot of money put into linux R & D.

Given a good, Free tool that really works, I can see lots of bad fonts being placed out under an "Open Font Lincense" (or whatever), but I can also see that many great fonts would come out of it. The hard part is to write the really, really good font creation package first. It will have to allow the "fontographer" to be totally involved in the font work without having to fight the software.

I'm very tired right now, and know that my description above is less than complete, but it's a first stab. What do you think about this approach?

Yeah, you're right, it could be done!

It _shall_ be done! Did you look at the links in my document? http://baldur.globalsymmetry.com/proprietary/com/wri/ch05.html In particular: http://crl.nmsu.edu/~mleisher/xmbdfed.html http://pfaedit.sourceforge.net/

Jody Harris

STH -- Hatton's Law: There is only One inviolable Law

On Tuesday 15 October 2002 01:27 pm, Jody Harris wrote:

On Tue, 2002-10-15 at 10:44, Steven T. Hatton wrote:

There's much more to designing a good font than that. I'm no fan of MS, but they do do some things right (occassionally, and I'm sure by accident). Here's a good article on fonts:

http://www.microsoft.com/typography/developers/fdsspec/default.htm

I guess I'm not sure if you are saying there is more to designing fonts specifically for computers, or in general. I know there's a lot to the proper form of the fonts, and some fonts scale, and others have outlines. I'll have to admit I don't know a lot about these things. Are you suggesting that the mechanics of creating the actulal glyph cannot be separated from the underlying technology, or that there is a lot to know about the shapes, geometry, leading, ascent, advance, descent, point size, and all that stuff?

...

In particular: http://crl.nmsu.edu/~mleisher/xmbdfed.html http://pfaedit.sourceforge.net/

I hadn't looked into those documents, but I'll spend some quality time with them now.

What I'm finding very frustrating is that I can't get a good answere as to what fonts I should have installed in order to use Mathematica, and the associated documentation which ships with it. I'd be willing to pay a few bucks for the correct set of fonts. I may even own a license of these. I've the feeling Micorsoft are doing some kind of substitution for the Helvetica named in the Mathematica documentation, which isn't happening in the Linux distribution. I'm pretty sure you can use Microsoft fonts on a Linux box. I know you can use some of them, I'm doing it. I copied them directly off the NTFS partition. Part of the problem is that the name Helvetica is a trade mark, and therefore cannot legally be used by vendors without paying the licensing fee. That may be why I'm having such a hard time getting direct and useful answers from people.

Jody Harris

STH -- Hatton's Law: There is only One inviolable Law

On Thursday 17 October 2002 12:12 pm, Mike Fabian wrote:

"Steven T. Hatton" writes:

...

I was searching the web for a font creation tool, and came across this:

http://lists.suse.com/archive/m17n/2002-Jul/0016.html

For some reason, that didn't seem to make it into the SuSE Linux 8.1 distribution.

Of course it is in 8.1, it is only missing from the CD sets because of lack of space. Of course it will be in the upcoming FTP version.

I'm just waiting for the day when we get 7 DVDs. :-)

A font editor for outline fonts is a very specialized tool which only very few people need.

Here is the version of PfaEdit which will be in SuSE Linux 8.1 FTP:

ftp://ftp.suse.com/pub/people/mfabian/8.1-i586/PfaEdit-020724-35.i586.rpm ftp://ftp.suse.com/pub/people/mfabian/8.1-src/PfaEdit-020724-35.src.rpm

PfaEdit-020910 is already release and I'll update the SuSE package soon.

I don't know how much time I'll be able to spend on this myself, but I'll do my best to advertise its availability. I'm not sure how easy it will be to learn to use. I may be way off about this, but it seems to me that the actual mechanics of creating the fonts could be separated from the technical details in such a way that a non-technical person could be given a tool for creating glyphs, and some instructions which would be very similar to instructions for creating physical fonts. I'm no expert on the subject, but my understanding is that much of the underlying geometry for creating and displaying computer fonts is based on, and mathematically identical to, traditional typsetting. This is not to trivialized the art of typsetting and font creation. I'm just trying to envision a tool that an intelligent schoolgirl with an artistic ability could produce a decent set of fonts without having to worry about a bunch of computer geek stuff. I suspect a good deal of the effectiveness of such a tool would have to do with the quality of the associated documentation. STH -- Hatton's Law: There is only One inviolable Law