New Royalty-Free Fonts for Scientific Writing/Publishing

stotterj writes: "Writing anything up in science almost always means changing fonts a lot to use all the characters necessary for formulas and units (times, symbol, arial). This is annoying. People at STIX Fonts are putting together a universal font set that already has the special characters built in and can be used from writing to publishing. The fonts that result from the project will be made available for free." The site says that "In particular. the STIX project will create a TeX implementation that TeX users can install and configure with minimal effort." The licensing for these fonts (discussed in the FAQ) will allow free use, but not modification.

Comments

Or alternatively

[gowen]

... you can just use any of the usual TeXfonts, which just work.

Re:Or alternatively

[Gamasta]

So right, LaTex is already enough when you want to write math-stuff. Well, I don't know how to use, compile and whatever, but I believe it's close to that staroffice formula program.

Re:Or alternatively

[fiziko]

In its most basic form, LaTeX is similar to StarOffice. I think it's more correct to say that StarOffice's equation handling techniques are very similar to LaTeX's basic layouts. I could go on for a long time about this, but suffice it to say I used StarOffice for a year. Then I discovered LaTeX, and I haven't looked back since. It's a fantastic and versatile software package that does not require a mouse at any stage of the operation, and has very low system overhead.

Distortion

[ThereIsNoSporkNeo]

Now, supposing I download myself a scientific document that uses this font. I, being a non-scientific type would not have this font. Does this mean I would end up with a bunch of ASCII 1-beginning of lowercase scrambled sentances?

Knowing my luck they'll figure out the meaning of life mathematically, and I won't have the font to understand it.

Re:Distortion

[Anonymous Coward]

if it's a .tex or .dvi file, you'll need the stix package to compile/view it. If it's a .ps file or .pdf file, it will probably have the font embedded in it.

If it's a web page, and you're using IE, you'll be prompted if you want to d/l the font.

And if it's slashdot, well, the <font> tag isn't allowed, and unicode is stripped.

Is something wrong with TeX?

[lonenut]

Is there an issue with using the fonts which come with TeX? This is a standard tool for creating papers in math and physics departments everywhere (many also use LaTeX).

I never had to pay Donald Knuth to use TeX, and I certainly never licensed the included fonts. Is there a legal issue with using the fonts included with TeX, or is this all an attempt to make some free fonts which are friendly to non-markup text processing tools like MS Word?

If the latter is the case, what is the point of releasing these fonts in a TeX usable form at all?

Re:Is something wrong with TeX?

[larry bagina]

well, the stix project isn't just for tex, it will also include a postscript type 1 and truetype font, be unicode compatable, and will have 7,000+ symbols.

TeX contains most math symbols, but is missing a lot of scientific symbols, and some people want to write their scientific paper in MS Word.

Re:Is something wrong with TeX?

[qbed]

Speaking from the biased position of a mathematician, noone in their right mind would go back to using MS Word after using Tex/LaTeX. However, Meta fonts are enormous (these are the fonts used by TeX).

Consider that TeX does not allow for typesetting in all languages at the moment. Only the latinised languages are supported properly, though there are some hacks for similar non-latinised languages.

There is a project that aims to remedy this called Omega (see the CTAN archive) with its own macro wrapper called Lambda. But one of the things that is needed for this is a solid set of fonts. Maybe (though I feel I should be somewhat cautious here) these STIX fonts will be it.

Re:Is something wrong with TeX?

[Anonymous Coward]

You can write scientific papers in word. I've seen someone do a whole astro-physics thesis in word. This, however, required that she break it up into a bunch of little sub-documents 'cause word wasn't up to handling 350 pages with lots-o-figures, hundreds-of-equations and 200 references. She ran screaming into the night when she went to turn in the written copy and the pointed out that it had to be double-spaced ... she had to fix every one of the freakin' documents by hand and futz with the figure alignment, fix the separate un-correlated list of figures. Wow, I'm glad I wrote my thesis in tex. Chapters written using LyX, 15 lines of perl to strip the pre-ramble's off each chapter, a top level .tex with some /makelistoffigures etc and /include's to bind it all together and a Makefile. Nothin' like sending the command "make doctor" and walking over to put the fancy paper in the printer.

Good But

[4of12]

How's this better than Blue Sky's donation of PostScript Type 1 versions of TeX's Computer Modern fonts that previously were only available in MetaFont and bitmapped versions?

I am glad an effort is being made to make the fundamental building blocks of language a truly open and free commodity. Commonly used fonts should be an open standard - except for exotic artistic fonts, proprietary holds on these building blocks are an undue brake on free expression and human communication. What's next, computer speech enunciations that are copyrighted and impossible to use without payment?

Re:Good But

[apsmith]

It's very similar to Blue Sky's donation - in fact inspired by it. What's different is:

1. Times-Roman instead of Computer-Modern style
2. Many more symbols.
3. OpenType and Unicode support.

Re:Good But

[hagarum]

It is always welcome to see high quality fonts
freely contributed to the community. While I
think the type1 CM fonts (and relatives) from
the American Math Society and from others are
quite good, and I'm not really interested in
using Times based font set myself, this is
never-the-less, a useful contribution.

A distressing number of societies are asking
their authors to submit in Times/Times Roman
and its ilk, despite the lack of beauty in these
fonts.

I am currently assembling a proceedings, and we,
at our sponsoring societies request, asked
authors to use Times. There are basically two
reasons for this:

It is deemed important to have a unified look,
and Times is most easily available to a wide
range of platforms and software.

Most authors are unfortunately creating their
papers in MS Word (and converting to PDF, our
required format), and the default font for a
Word installation is Times. Few authors ever
bother to change this or look at other fonts.
You can criticize them if you wish, but they
are not paid to fuss with fonts, and if they
can say what they need to with minimum fuss
using the defaults, they have no motivation
to consider alternatives.

I must comment also on the matter of embedding
fonts. I think all fonts should be embedded
in Type 1 or similar outline format. It
is the only way to be sure. A number of papers
that we have received as PDFs do not display
quite right because their authors have different
fonts than I do---even slightly different
"Times" fonts. And papers from Asia, even when
in English, invariably contain glyphs from
Asian fonts---apparently even spaces, digits and
some symbols like = signs might be pulled out
of an Asian font and sneak into a submission with
the author none the wiser.

There is no way for authors to know what fonts
their readers will have, either today or 10
years or 50 years from now. There is no chance
that any set of "standard fonts" will meet all
needs, and there is little chance that most
authors whose primary job is science will take
the time to become familiar with what fonts are
"standard" and which are not. They frankly assume that *whatever* they have, everyone will
have.

If a PDF paper is to replace a paper paper
it should contain as much information as
reasonable to describe itself. Embedding
outline (not bitmapped) fonts does not result in
unreasonably large PDF files, and it ensures
that the readers will see what the author
intends.

Yes, there is.

[g4dget]

TeX fonts are not PostScript or TrueType fonts. That causes all sorts of practical problems.

For example, there are lots and lots of LaTeX papers out there in PDF format that were created with dvips and pstopdf, and the results are just awful. Papers with embedded TeX fonts are also very large.

In fact, many people nowadays just use PDFTeX, and it would be good to have fonts that go with that natively.

Now, you could convert TeX fonts to PostScript or TrueType. But typographically, they are not really all that nice.

Re:Yes, there is.

[rjw57]

TeX fonts are not PostScript or TrueType fonts. That causes all sorts of practical problems.

Actually you can get a set of Adobe Type 1 fonts for LaTeX from the AMS. You can then make TeX use these standard PostScript Type 1 fonts in the PDFs it produces. More details here and here. You can also make LaTeX use the standard PostScript fonts for its body by using the \usepackage{times} directive in the preamble.

computer modern, and missing characters

[apsmith]

Since I'm involved in the project I guess I'll comment...

We would happily use TeX for everything, except for two fundamental problems, and a few more superficial ones. First the fundamental problems:

• The standard TeX fonts are Computer Modern (Knuth's invention) - these are lighter and "looser" than the standard text typefaces (Adobe Times Roman in particular), and text becomes uglier and slightly harder to read when the two types of fonts are used together in the same document.
• The standard Tex fonts are also missing a lot of special symbols that scientific publishers have, over the years, had to create for themselves. It's impossible to capture every special character that a mathematician or physicist may decide to invent for some particular purpose, but we've managed to include in the description of the font, and in the latest Unicode version (3.2) essentially all the special characters we could find that have received repeated use in scientific communications. If more appear later we'll get them added to the font. The idea is to be comprehensive.

The more superficial problems are, first that TeX distributions suffer from wild inconsistencies in what particular fonts are available - early experiences with making font-less PDF files (or even DVI files) that relied on users having a TeX font distribution available to them for display were essentially total failures. And second, TeX is unfortunately not all that widely used even by scientists and engineers... Another side problem is the way TeX is limited to a small number of 256-character font files (with further restrictions on tfm files) - what we're planning is to have all 8000 or so glyphs available in 4 OpenType fonts (regular, bold, italic, bold italic) - of course to use with TeX it will have to be broken out into a few dozen 256-character Type-1 font files.

The goal here is to be able to distribute scientific content in XML format, with the mathematical content marked up in a standard way such as MathML, and special characters treated properly as entities within Unicode, and then have essentially any conforming application (web browsers, Star Office, MSWord one hopes...) display the content correctly and reliably.

Re:computer modern, and missing characters

[nagora]

...lighter and "looser" than the standard text typefaces (Adobe Times Roman in particular),

Times is not a standard text typeface, it is a standard newspaper typeface and is buck-ugly and tiresome to read on normally proportioned pages. NEVER use Times as a body face unless you're being charged a great deal too much for your paper or postage!

TWW

Re:computer modern, and missing characters

[apsmith]

Sounds like you haven't been reading any scientific journals lately :-)

Embedded fonts

[nucal]

If it's a .ps file or .pdf file, it will probably have the font embedded in it.

This is probably the key point ... all of the sponsors of this font are scientific publishing houses that are already heavily into using Abobe .pdf files for online publishing. Having a standardized scientific font is one more step towards a uniform online publishing format. Whether this should have been based on .pdf files or not is another issue - it is certainly not my favorite format, but it looks like we're stuck with it.

New Century Schoolbook

[Anonymous Coward]

Use it if you have it. Otherwise, try Bookman.
Headers can be Ariel Bold. No more than one serif
font and one sans serif font in a document please!

TeX distributions; source; sans serif

[bcrowell]

What you're doing is great -- there is strong need for this!

Are you using MetaFont to create the fonts? If so, will the sources be available?

Could you explain a little more about the rationale for creating an entirely new TeX distribution? teTeX has had trouble staying up to date, but what's wrong with TeXLive? Does it really make sense to reduplicate their effort?

Lastly, I'm disappointed that there's no mention of creating a sans serif math font. Any chance that you'll make one?

No...

[SpiffyMarc]

What's next, is if Microsoft throws their muscle into the font arena with the new FontAsAService technology:

Monthly Font Summary
You looked at Comic Sans MS 24 times
You looked at Tahoma 13 times

You owe: $12.35

Man, Adobe would be all over that!