Best font selection for a menu?

On Fri, 21 Sep 2007, Andy Dingley wrote:

Why specify Times and Times New Roman together?
The synonym mechanisms will merge those.

Which "synonym mechanisms"? - Oh, perhaps you mean some entries
in the so-called registry of MS Windows. But there are other
operating systems than MS Windows and they don't have such
"synonym mechanisms".

Scripsit Andreas Prilop:

On Fri, 21 Sep 2007, Andy Dingley wrote:

Why specify Times and Times New Roman together?
The synonym mechanisms will merge those.

Which "synonym mechanisms"? - Oh, perhaps you mean some entries
in the so-called registry of MS Windows. But there are other
operating systems than MS Windows and they don't have such
"synonym mechanisms".

Sorry, I have no idea of what either of you is saying here. In CSS,
specifying font-family: Times, Times New Roman means suggesting Times as the
primary font, Times New Roman as secondary. IE naturally gets part of this
wrong, but that's a different story, and the font names are not synonyms.

--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/

On Fri, 21 Sep 2007 13:58:01 +0200, Andreas Prilop
<Prilop2007 (AT) trashmail (DOT) net> wrote:

On Fri, 21 Sep 2007, Andy Dingley wrote:

Why specify Times and Times New Roman together?
The synonym mechanisms will merge those.

Which "synonym mechanisms"?

That would be a matter for browser implementations, thus not part of the
spec itself and thus off-charter for this ng.

However I'm sure you'd agree that informally, in general terms of
browser implementations, they are capable of mapping variant forms of
/times( new)?( roman)?/ onto a variant of a "times roman" font that's
available, even when the default stylesheet maps serif to a seriffed
font other than these "times roman" variants.

Scripsit Andy Dingley:

On Fri, 21 Sep 2007 13:58:01 +0200, Andreas Prilop
Prilop2007 (AT) trashmail (DOT) net> wrote:

On Fri, 21 Sep 2007, Andy Dingley wrote:

Why specify Times and Times New Roman together?
The synonym mechanisms will merge those.

Which "synonym mechanisms"?

That would be a matter for browser implementations, thus not part of
the spec itself and thus off-charter for this ng.

What? You must have confused this with the www-style list, or something. The
charter is rather abstract: "This unmoderated news group is intended for the
discussion of Web style sheets", followed by general advocacy of style
sheets. Check the FAQ for clarifications:
http://www.dev-archive.net/articles/...-mFAQ.html#C02
Note that "Bugs and limitations in implementations" have been explicitly
mentioned.

Besides, if synonym mechanisms were something off-topic, why did you mention
them? And you're still refusing to give us any kind of definition.

However I'm sure you'd agree that informally, in general terms of
browser implementations, they are capable of mapping variant forms of
/times( new)?( roman)?/ onto a variant of a "times roman" font that's
available, even when the default stylesheet maps serif to a seriffed
font other than these "times roman" variants.

How would that relate to _synonyms_, and what on &Planet; do you _mean_?
Please do not hesitate to give an example (a URL, with specific explanation
of what we should see there, please).

--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/

in comp.infosystems.www.authoring.stylesheets, Michael Stemper wrote:

In article <46f26505$0$281$dbd4d001 (AT) news (DOT) wanadoo.nl>, www.FocusOnPanama.com writes:
"Andy Dingley" <dingbat (AT) codesmiths (DOT) com> wrote


font: 100% sans-serif

What about normal body text? I had some fonts selected but apparently some
people here had some objections. I modified it and now it only shows
hideously large fonts that look like a primary school book

That is a problem with the configuration of your PC or workstation. On
my PC, 100% fonts have upper-case letters approximately 1/8" high, which
is as small as I can comfortably read. If your browsing situation has
100% fonts too big for your tastes, fix *your* browsing situation; don't
try to break mine.

His default font is serif, most likely Times New Roman (default font of
99% browsers). So if you change font to sans-serif (most likely Arial or
Helvetica, or even Verdana), it does look much bigger. Even if it has
same font size.

The height of upper case letters are exactly same. But only about 2% of
characters are uppercase, so that doesn't really matter. What does matter
is that lower case letters are considerable amount bigger.

For normal body text, just don't set any font size, style, weight or
family.

--
Lauri Raittila <http://www.iki.fi/lr>

On Fri, 21 Sep 2007, Jukka K. Korpela wrote:

Oh, perhaps you mean some entries
in the so-called registry of MS Windows. But there are other
operating systems than MS Windows and they don't have such
"synonym mechanisms".

Sorry, I have no idea of what either of you is saying here.

FontSubstitutes entries in the Windows registry
http://www.google.com/search?q=FontSubstitutes
allow you to map (unavailable) font names to (available) font
names. For example, Times=Times New Roman
means that you have a typeface named "Times" in all programs
- but the operating system will actually use Times New Roman
for display. Other examples are Helvetica=Arial and
Courier=Courier New .

Scripsit Andreas Prilop:

On Fri, 21 Sep 2007, Jukka K. Korpela wrote:

Oh, perhaps you mean some entries
in the so-called registry of MS Windows. But there are other
operating systems than MS Windows and they don't have such
"synonym mechanisms".

Sorry, I have no idea of what either of you is saying here.

FontSubstitutes entries in the Windows registry
http://www.google.com/search?q=FontSubstitutes
allow you to map (unavailable) font names to (available) font
names.

I have a vague idea of that, but how does that relate to CSS?

For example, Times=Times New Roman
means that you have a typeface named "Times" in all programs
- but the operating system will actually use Times New Roman
for display. Other examples are Helvetica=Arial and
Courier=Courier New .

Well, maybe. If such definitions exist and take effect system-wide, they are
comparable to creating copies of fonts under different names. It can be
regarded as misleading and confusing, but I don't think is particularly
relevant to CSS authoring. It's comparable to the fact that the same name
can refer to different fonts in different computers (at least different
_versions_ of a font).

Besides, I just checked the settings in my current computer. Regedit tells
that there are mappings for Times and Helvetica, as you wrote (though not
for Courier). Yet, when I test an HTML document where I set font-family in
CSS, text set to Helvetica is different from text set to Arial (though
similar), and ditto for Times and Times New Roman. So the settings aren't
that system-wide and penetrating after all.

Thus, at least when taken as a general statement, Andy's claim
"Why specify Times and Times New Roman together? The synonym mechanisms
will merge those."
was not correct, in CSS authoring, even if we limit ourselves to the Windows
world.

P.S. Followup-To overruled. Please don't set Followup-To without saying so
in the message body. And in this case, I don't think there was any need to
change group, since we were discussing fonts in the CSS context.

--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/

On Mon, 24 Sep 2007, Jukka K. Korpela wrote:

FontSubstitutes entries in the Windows registry
http://www.google.com/search?q=FontSubstitutes
allow you to map (unavailable) font names to (available) font
names.

I have a vague idea of that, but how does that relate to CSS?

I didn't come up with these "synonym mechanisms". ;-)
Authors on MS Windows might think it is unnecessary to include
Helvetica or Times because the operating system already cares
about these.

It can be
regarded as misleading and confusing, but I don't think is
particularly relevant to CSS authoring.

I didn't come up with ... oh, I said this already.

Thus, at least when taken as a general statement, Andy's claim
"Why specify Times and Times New Roman together?
The synonym mechanisms will merge those."
was not correct, in CSS authoring, even if we limit ourselves to
the Windows world.

Quite so.

On 24 Sep, 20:59, "Jukka K. Korpela" <jkorp... (AT) cs (DOT) tut.fi> wrote:

Thus, at least when taken as a general statement, Andy's claim
"Why specify Times and Times New Roman together? The synonym mechanisms
will merge those."
was not correct, in CSS authoring, even if we limit ourselves to the Windows
world.

Perhaps not, but it's (as ever) a pragmatic approach.

What else are you going to do? Specify _every_ variant naming of "TNR/
Times/New/Roman/Rmn/Ramen" to try and make one stick? It's impossible
to predict just how this "TNR" font has been named on an unknown
platform. There's already a generic fallback mechanism available,
that's enough to stop things "breaking" (by which I mean "failing to
deliver some serifed font on a platform that has one").

For practical purposes of "Catching a Times Roman variant in favour of
another serifed font" the CSS value
<< 'Times Roman', serif >> is adequate to cause this to happen on a
useful share of platforms that could possibly do it, to a level that's
better than << serif >> alone. A verbose attempt to list "everything"
works no significantly better.

Andy Dingley wrote:

On Fri, 21 Sep 2007 13:58:01 +0200, Andreas Prilop
Prilop2007 (AT) trashmail (DOT) net> wrote:

On Fri, 21 Sep 2007, Andy Dingley wrote:

Why specify Times and Times New Roman together?
The synonym mechanisms will merge those.
Which "synonym mechanisms"?

That would be a matter for browser implementations, thus not part of the
spec itself and thus off-charter for this ng.

Would like to know more of this "synonym mechanisms" and which border
supports such font substitutions. As to having to specify each variant
of a font name, all I can say is it appears that you must use the name
exactly, partial names just won't do, Times, "Times Roman", "Times New
Roman"...

It would be *nice* if Times or "Times Roman" would do but I have not
witnessed such mechanism on any browser on Windows (any version) do it.

Example is a Mac T1 font named "Papyrus" but the Windows TTF version is
named "Papyrus LET". If you have the TTF installed on your Windows box
and have

body { font-family: papyrus; }


You're going to see your default font, most likely "Times New Roman" and
not "Papyrus LET" unless your have

body { font-family: "papyrus let"; }

So I would assume for both Mac and Windows boxes with Papyrus installed
on both you would need

body { font-family: papyrus, "papyrus let"; }

However I'm sure you'd agree that informally, in general terms of
browser implementations, they are capable of mapping variant forms of
/times( new)?( roman)?/ onto a variant of a "times roman" font that's
available, even when the default stylesheet maps serif to a seriffed
font other than these "times roman" variants.

Which ones?

--
Take care,

Jonathan
-------------------
LITTLE WORKS STUDIO
http://www.LittleWorksStudio.com