lTibetan font .. problems in IE and Netscape on Mac
Andreas Prilop
In article <i3wU6.72268$Be4.23...@news3.rdc1.on.home.com>,
"Patricia Warwick" <pwar...@home.com> wrote:
> I'm trying to develop some Web pages using lTibetan (a Mac font).
The World Wide Web does not rely on special fonts. It makes no sense to say
"develop Web pages using the blahblah font".
> When I insert two characters they get generated as ¤ and as &Mac221;
> I do this using Adobe GoLive.
¤ means the currency symbol, a circle with four strokes. On a PC,
for example, you can type it with alt+0164. Since &Mac221; does not
exist in HTML 4.0, I conclude that Adobe GoLive is severely broken.
> After I insert these characters they look Ok
> briefly but then the change into incorrect characters, not the ones I
> intended.
Which characters did you intend? Tibetan characters?
> The other 10 characters that I insert on the same page behave
> correctly. I have heard that there are some problems with the iso-8859-1
> charset on the Mac and that 14 characters do not display correctly.
Assuming you want to write Tibetan text, ISO-8859-1 should be of no use
for you since it includes only West European Latin characters.
I recommend <http://ppewww.ph.gla.ac.uk/~flavell/charset/> as a starting
point to further readings in the strange world of character sets and
encodings.
Patricia Warwick
Your comments are not entirely accurate, but I may not have explained
myselft correctly ... so I'll address them in the body. Thanks, Patricia
"Andreas Prilop" <andreas...@altavista.net> wrote in message
news:110620011810034654%andreas...@altavista.net...
> [[ Note F'up ]]
>
> In article <i3wU6.72268$Be4.23...@news3.rdc1.on.home.com>,
> "Patricia Warwick" <pwar...@home.com> wrote:
>
> > I'm trying to develop some Web pages using lTibetan (a Mac font).
>
> The World Wide Web does not rely on special fonts. It makes no sense to
say
> "develop Web pages using the blahblah font".
That is probably because I didn't use the correct technical jargon, but I AM
developing Web Pages that do use a particular font (LTibetan). The <font>
tag specifies LTibetan for some text and the default font for the rest of
the text. The charset is defined as iso-8859-1.
>
> > When I insert two characters they get generated as ¤ and as
&Mac221;
> > I do this using Adobe GoLive.
>
> ¤ means the currency symbol, a circle with four strokes. On a PC,
> for example, you can type it with alt+0164. Since &Mac221; does not
> exist in HTML 4.0, I conclude that Adobe GoLive is severely broken.
Correct, but even if I specify Ý (which is what I think GoLive intended)
it still does not display the character correctly)
>
> > After I insert these characters they look Ok
> > briefly but then the change into incorrect characters, not the ones I
> > intended.
>
> Which characters did you intend? Tibetan characters?
Correct, I am specifying particular Tibetan characters. The Web pages are
intended to teach someone to read Tibetan.
>
> > The other 10 characters that I insert on the same page behave
> > correctly. I have heard that there are some problems with the iso-8859-1
> > charset on the Mac and that 14 characters do not display correctly.
>
> Assuming you want to write Tibetan text, ISO-8859-1 should be of no use
> for you since it includes only West European Latin characters.
That is only the case if it is not overridden by specifying the <font> tag.
>
> I recommend <http://ppewww.ph.gla.ac.uk/~flavell/charset/> as a starting
> point to further readings in the strange world of character sets and
> encodings.
Thanks, I'll take a look at this.
Patricia Warwick
However, I started this whole exercise without being a font developer.
Because of the various problems I have encountered I now understand a few
more of the issues. But the question is, what is the best way to proceed
given that I am not a font developer? Even if I were an expert in this area,
it is not clear to me that there is a real standard to use. And each Browser
has its own issues. Should I develop a Unicode font myself?
I'm certainly open to suggestions ... including even that this is the wrong
time to attempt to do what I am attempting to do. As an irrelevant aside -
a long long time ago I was given a project to help a music professor analyze
the musical style of Palestrino using the primitive computer technology of
the time (IBM S360). I eventually came to the conclusion (based on advice of
experts) that it was better to take a musician and train them to be a
programmer to do this than to take me (a programmer with basic knowledge of
music) and learn enough about music theory to develop the programs. Perhaps
the same applies here.
"Andreas Prilop" <andreas...@altavista.net> wrote in message
news:110620011810034654%andreas...@altavista.net...
Daniel Tobias
>
> > Assuming you want to write Tibetan text, ISO-8859-1 should be of no use
> > for you since it includes only West European Latin characters.
>
> That is only the case if it is not overridden by specifying the <font> tag.
No, it's always the case, in accordance with the standards; it's only
due to bugs in some browser versions that anything else might sometimes
be displayed. The proper standards-compliant manner of using "special"
characters is to use the Unicode values for them -- see the Unicode
site:
http://www.unicode.org/
I have some more information about characters in my site:
http://www.dantobias.com/webtips/char.html
--
--Dan
Dan's Web Tips: http://www.dantobias.com/webtips/
Thanasis Kinias
> "Andreas Prilop" <andreas...@altavista.net> wrote in message
> news:110620011810034654%andreas...@altavista.net...
>> The World Wide Web does not rely on special fonts. It makes no sense to
> say
>> "develop Web pages using the blahblah font".
>
> That is probably because I didn't use the correct technical jargon, but I
> AM developing Web Pages that do use a particular font (LTibetan). The
> <font> tag specifies LTibetan for some text and the default font for the
> rest of the text. The charset is defined as iso-8859-1.
[...]
>> Assuming you want to write Tibetan text, ISO-8859-1 should be of no use
>> for you since it includes only West European Latin characters.
>
> That is only the case if it is not overridden by specifying the <font>
> tag.
Patricia,
Andreas is quite right here, actually. The <font> element (which is
deprecated in HTML4 and XHTML1, and has been removed entirely from the
latest specification, XHTML1.1) can _suggest_ a font which the user agent
_might_ choose to use in rendering the text, but that is all. The user
agent is under no obligation actually to use that font, even if it is
available.
The font doesn't get delivered with the page, like it does in a PDF
(Acrobat) file. If the end user doesn't have the font installed on his
machine, he will see not what you intended but the literal ISO-8859-1
characters rendered in the default font on his machine (typically Times
Roman or some variation on that theme).
I am writing this on a Linux box. I have no LTibetan font, nor could I
install it if I wanted to. The same goes for Windows boxes, which are the
overwhelming majority of users (whether we like it or not). Even if I were
using a Macintosh, the odds that I would have such an esoteric font are
slim. So no matter how carefully you do your design, if I tried to view
your page, I would get nonsense.
There's nothing wrong with using a font like this, but not in a Web page.
Try distributing PDFs, for example, with the font embedded.
Alternately, use UTF-8 encoding. That's the only real way of doing Tibetan
text, AFAIK.
Regards,
--
Thanasis Kinias
Optimal LLC -- Accessible Web Design
Arizona, U.S.A.
Andreas Prilop
"Patricia Warwick" <pwar...@home.com> wrote:
> But the question is, what is the best way to proceed
> given that I am not a font developer?
Since Tibetan support in browsers/operating systems/typefaces
is close to zero, consider to work with images for each Tibetan
character. Many web sites that explain Arabic and Hebrew do so--
and Arabic/Hebrew support in current browsers/operating systems
is much better.
And have a look at <http://www.xs4all.nl/~wijnands/nnq/nquote.html>
Your articles will become more readable if you follow that guidelines.
Andreas Prilop
"Patricia Warwick" <pwar...@home.com> wrote:
> Your comments are not entirely accurate,
Which one?
> The <font>
> tag specifies LTibetan for some text and the default font for the rest of
> the text. The charset is defined as iso-8859-1.
This is a misunderstanding. You could use <font face> to specify a
_typeface_ but not different _characters_. Please read carefully
<http://ppewww.ph.gla.ac.uk/~flavell/charset/fontface-harmful.html>
> Correct, but even if I specify Ý (which is what I think GoLive intended)
> it still does not display the character correctly)
Ý is displayed correctly if and only if you see a capital Y with
acute accent.
<http://www.unics.uni-hannover.de/nhtcapri/multilingual2.html>
An example for a Tibetan character is ཀ .
> > Assuming you want to write Tibetan text, ISO-8859-1 should be of no use
> > for you since it includes only West European Latin characters.
>
> That is only the case if it is not overridden by specifying the <font> tag.
No, no! You could write
<font face="Helvetica">Ý</font>
to suggest that "capital Y with acute accent" be displayed in Helvetica.
But the semantics of Ý cannot altered by <font face>.
Patricia Warwick
> I have read the article you referenced and I understand your points.
> However, I started this whole exercise without being a font developer.
> Because of the various problems I have encountered I now understand a few
> more of the issues. But the question is, what is the best way to proceed
> given that I am not a font developer? Even if I were an expert in this
area,
> it is not clear to me that there is a real standard to use. And each
Browser
> has its own issues. Should I develop a Unicode font myself?
>
> I'm certainly open to suggestions ... including even that this is the
wrong
> time to attempt to do what I am attempting to do. As an irrelevant
aside -
> a long long time ago I was given a project to help a music professor
analyze
> the musical style of Palestrino using the primitive computer technology of
> the time (IBM S360). I eventually came to the conclusion (based on advice
of
> experts) that it was better to take a musician and train them to be a
> programmer to do this than to take me (a programmer with basic knowledge
of
> music) and learn enough about music theory to develop the programs.
Perhaps
> the same applies here.
>
> news:110620011810034654%andreas...@altavista.net...
> >
> > In article <i3wU6.72268$Be4.23...@news3.rdc1.on.home.com>,
> > "Patricia Warwick" <pwar...@home.com> wrote:
> >
> > > I'm trying to develop some Web pages using lTibetan (a Mac font).
> >
> say
> > "develop Web pages using the blahblah font".
> >
> &Mac221;
> > > I do this using Adobe GoLive.
> >
> > ¤ means the currency symbol, a circle with four strokes. On a
PC,
> > for example, you can type it with alt+0164. Since &Mac221; does not
> > exist in HTML 4.0, I conclude that Adobe GoLive is severely broken.
> >
> > > After I insert these characters they look Ok
> > > briefly but then the change into incorrect characters, not the ones I
> > > intended.
> >
> > Which characters did you intend? Tibetan characters?
> >
> > > The other 10 characters that I insert on the same page behave
> > > correctly. I have heard that there are some problems with the
iso-8859-1
> > > charset on the Mac and that 14 characters do not display correctly.
> >
> > for you since it includes only West European Latin characters.
> >
Patricia Warwick
>
> Your comments are not entirely accurate, but I may not have explained
> myselft correctly ... so I'll address them in the body. Thanks, Patricia
>
> news:110620011810034654%andreas...@altavista.net...
> > [[ Note F'up ]]
> >
> > In article <i3wU6.72268$Be4.23...@news3.rdc1.on.home.com>,
> > "Patricia Warwick" <pwar...@home.com> wrote:
> >
> > > I'm trying to develop some Web pages using lTibetan (a Mac font).
> >
> > The World Wide Web does not rely on special fonts. It makes no sense to
> say
> > "develop Web pages using the blahblah font".
>
AM
> tag specifies LTibetan for some text and the default font for the rest of
> the text. The charset is defined as iso-8859-1.
> >
> &Mac221;
> > > I do this using Adobe GoLive.
> >
> > ¤ means the currency symbol, a circle with four strokes. On a
PC,
> > for example, you can type it with alt+0164. Since &Mac221; does not
> > exist in HTML 4.0, I conclude that Adobe GoLive is severely broken.
intended)
> it still does not display the character correctly)
> >
> > > briefly but then the change into incorrect characters, not the ones I
> > > intended.
> >
> > Which characters did you intend? Tibetan characters?
>
> intended to teach someone to read Tibetan.
> >
> > > correctly. I have heard that there are some problems with the
iso-8859-1
> > > charset on the Mac and that 14 characters do not display correctly.
> >
> > Assuming you want to write Tibetan text, ISO-8859-1 should be of no use
> > for you since it includes only West European Latin characters.
>
tag.
> >
> > point to further readings in the strange world of character sets and
> > encodings.
Mike
> There's nothing wrong with using a font like this, but not in a Web
> page. Try distributing PDFs, for example, with the font embedded.
> Alternately, use UTF-8 encoding. That's the only real way of doing
> Tibetan text, AFAIK.
That last bit may be a good thing to look into.
I'm feeling the need to jump to Patricia's defence here; she asked a
pretty simple question (which unfortunately nobody was really able to
answer) and got told to stop trying to use html for this as not everybody
will have this specific font. However, the reason she's doing it is to
teach Tibetan, and if there's a font with the right characters in it that
does the job, it sounds like a pretty developer- and user-friendly way of
doing it. She never stated that she was making her home page and expected
the entire world to come see it in Tibetan; for all we know it's a closed
project with a closed user group who will have the font, because it's
needed for the pages and they want to learn Tibetan.
If you must attack people's methods, at least try to find out the exact
situation. If this is for example for a small network of machines in a
classroom as a teaching aid, where all machines will have the desired
font installed, what's the problem? I know I'm assuming a lot of things
here, and if she really is trying to publish a webpage on the net for
consumption by the general population, then yes, it's a Bad Idea (TM). I
got the distinct impression that this wasn't the case, though.
As for the actual problem - have you tried editing the html outside of
GoLive? Do the offending charactesr actually get replaced by something
else, and if so does changing them back fix it? This is possibly only a
partial solution at best, but I don't really have any ideas. Sorry :-)
--
Mike.
Remove "-spam" to mail me.
Etan Wexler
<rapaci...@ii.net> wrote:
> [She's] doing it is to
> teach Tibetan, and if there's a font with the right characters in it that
> does the job, it sounds like a pretty developer- and user-friendly way of
> doing it.
Exactly.
However, the LTibetan font does not have the right characters.
LTibetan has Latin characters mapped to glyphs for Tibetan characters.
This is a violation of character identity, which becomes apparent as
soon as one tries to change font, search the text, copy text, collate
the text, break lines, break words, and so forth.
> She never stated that she was making her home page and expected
> the entire world to come see it in Tibetan; for all we know it's a closed
> project with a closed user group who will have the font, because it's
> needed for the pages and they want to learn Tibetan.
If the material is for a non-specialist audience, using bitmap graphics
of Tibetan glyphs is the most sensible approach. A specialist audience
may be expected to configure their system to handle properly Tibetan
characters.
This business of mismapped fonts is a stopgap acceptable for print
jobs. Once printed, all that we have to identify characters are the
glyphs, so what was once, say, a mismapped Latin character in memory
has become a Tibetan character on paper. For the interchange of
electronic text, mismapped fonts are a bad idea and a last resort.
Mike
>> [She's] doing it is to teach Tibetan, and if there's a font with the
>> right characters in it that does the job, it sounds like a pretty
>> developer- and user-friendly way of doing it.
> Exactly.
> However, the LTibetan font does not have the right characters. LTibetan
> has Latin characters mapped to glyphs for Tibetan characters. This is a
> violation of character identity, which becomes apparent as soon as one
> tries to change font, search the text, copy text, collate the text,
> break lines, break words, and so forth.
Okay, good points. I'll crawl back into my little corner and be quiet now.
Alan J. Flavell
> However, the LTibetan font does not have the right characters.
I _don't_ recommend what I'm about to say. However, it is sort-of
theoretically respectable.
If you have a font which implements a private character coding, you
_could_ put 8-bit characters into your document and advertise them as
charset=x-user-defined , or even make up your own name for this
non-standard coding.
It's discussed towards the end of this page
http://ppewww.ph.gla.ac.uk/~flavell/charset/fontface-harmful.html
(the name of which should be self explanatory ;-)
What you can't respectably do is to pretend that you're changing fonts
within a document when in reality you are changing your character
coding.
Of course, what I really recommend is the principle that was
introduced with RFC2070, studiously ignored by certain wowser
developers, re-asserted by HTML4 and finally now supported, at least
in principle, by recent browsers. A pity that it doesn't work 'out of
the box' yet for Tibetan, though.
> > She never stated that she was making her home page and expected
> > the entire world to come see it in Tibetan; for all we know it's a closed
> > project with a closed user group who will have the font, because it's
> > needed for the pages and they want to learn Tibetan.
>
> If the material is for a non-specialist audience, using bitmap graphics
> of Tibetan glyphs is the most sensible approach.
With alt="&#bignumber;" for those browsers which can do it (and
indexers which support the alt attribute), you're probably right just
now.
all the best
Alan J. Flavell
> I'm feeling the need to jump to Patricia's defence here;
I'm not aware that anyone here was attacking her.
> she asked a pretty simple question
She asked what might seem to be a simple question, but based on a
misunderstanding of how HTML is designed to work. HTML's answer is
also relatively simple, but unfortunately the two things have a
tendency to be mutually incomprehensible without an in-depth study,
which makes for frustration on both sides. However, my impression is
that Patricia quickly got up to speed on that. I worry that you are
trying to pander to perceived ignorance, and that she is already way
beyond where you credited her with being.
>(which unfortunately nobody was really able to answer)
I was away at the time that the question was raised, but on reviewing
the answers that were posted, I would say there were some much better
answers given than I had been accustomed to seeing 1 or 2 years back.
> and got told to stop trying to use html for this as not everybody
> will have this specific font.
It sounds to me as if you are still misunderstanding the answer.
> However, the reason she's doing it is to
> teach Tibetan, and if there's a font with the right characters in it that
> does the job,
But it doesn't "do the job" in HTML's terms at all. It only gives a
visual illusion of working, relying on what can really be categorised
as browser bugs, while being unsound at HTML's fundamental level.
> If you must attack people's methods, at least try to find out the exact
> situation.
The "exact situation" is HTML, RFC2070 and open published interworking
specifications. That is why we are here, presumably.
If she wanted a non-standard solution, she would have come to the
wrong place. Offering a nonstandard solution might appear to give more
immediate gratification, but in the long term would be a disservice
IMHO.
I'm not familiar with Tibetan myself, but that unfamiliarity does not
in any way affect the principles on which HTML4 is based. Arial
Unicode MS is said to include Tibetan characters, but for some reason
my MSIE5 does not seem to want to display them. Netscape 6 PR3 (I'm
using that on Win95 for example) seems happy to display plausible
characters - whether they would be acceptable to someone familiar with
Tibetan typography, I honestly don't know, as I say. Sample material
at http://www.hclrss.demon.co.uk/unicode/tibetan.html or
http://ppewww.ph.gla.ac.uk/~flavell/unicode/unidata0F.html
best regards
Bertilo Wennergren
"Alan J. Flavell":
> I'm not familiar with Tibetan myself, but that unfamiliarity does not
> in any way affect the principles on which HTML4 is based. Arial
> Unicode MS is said to include Tibetan characters, but for some reason
> my MSIE5 does not seem to want to display them.
My MSIE6 does, and I'm sure MSIE5 would too, on my system. Your
problem is most probably due to not having installed the entire
Multilanguage support of Windows (check in the Control Panel).
Or perhaps you have not installed all the relevant modules of
MSIE5 (it's highly modular...).
> Netscape 6 PR3 (I'm
> using that on Win95 for example) seems happy to display plausible
> characters - whether they would be acceptable to someone familiar with
> Tibetan typography, I honestly don't know, as I say.
The same goes for me. What I see with MSIE6 on your page
<http://www.hclrss.demon.co.uk/unicode/tibetan.html>
looks very much as plausible Tibetan characters.
Arial Unicode MS is really a great thing. Any Windows user intersted
in reading "exotic" languages really ought to get it and install it.
I do not know however if it's usable on Macs...
--
#################################################################
Bertilo Wennergren
<http://purl.oclc.org/net/bertilo>
<bert...@bertilo.se.fm>
#################################################################
Eric Jarvis
wrote...
>
> Arial Unicode MS is really a great thing. Any Windows user intersted
> in reading "exotic" languages really ought to get it and install it.
> I do not know however if it's usable on Macs...
>
may I put in a special plea for the inhabitants of Hong Kong and
Japan to do this...since those seem to be the only places where
something other than unicode has become the standard :)
--
eric
"live fast, die only if strictly necessary"
Alan J. Flavell
> My MSIE6 does, and I'm sure MSIE5 would too, on my system. Your
> problem is most probably due to not having installed the entire
> Multilanguage support of Windows (check in the Control Panel).
> Or perhaps you have not installed all the relevant modules of
> MSIE5 (it's highly modular...).
My experience in the past has been that if a component is missing,
which is needed for browsing a particular page, then IE5 offers to
install it when you try to browse that page. But this time it didn't
do that.
> > Netscape 6 PR3 (I'm
> > using that on Win95 for example) seems happy to display plausible
> > characters - whether they would be acceptable to someone familiar with
> > Tibetan typography, I honestly don't know, as I say.
>
> The same goes for me. What I see with MSIE6 on your page
> <http://www.hclrss.demon.co.uk/unicode/tibetan.html>
That's Alan Wood's page actually...
> looks very much as plausible Tibetan characters.
Seems that way to me.
> Arial Unicode MS is really a great thing.
Yes, but MSIE doesn't really need it, because it can pick and choose
from a selection of fonts.
I forgot to mention that even NN4.* (I used NN4.6 this time) happily
displays the Tibetan characters on my test pages, if I configure Arial
Unicode MS as my Unicode font. It also (as usual) displays Alan
Wood's pages _provided_ you manually select utf-8 coding. It looks
however as if it doesn't understand what to do with the "vowel signs"
which I _think_ (Tibetan specialists please comment!) are meant to
work like diacriticals (i.e non-spacing characters).
cheers
Bertilo Wennergren
"Eric Jarvis":
> on Wed, 13 Jun 2001 14:48:04 +0200, bert...@bertilo.se.fm
> wrote...
> > Arial Unicode MS is really a great thing. Any Windows user intersted
> > in reading "exotic" languages really ought to get it and install it.
> > I do not know however if it's usable on Macs...
> may I put in a special plea for the inhabitants of Hong Kong and
> Japan to do this...since those seem to be the only places where
> something other than unicode has become the standard :)
Just to be clear: The font "Arial Unicode MS" is not for Unicode
encoded pages only (UTF-8, UTF-16...). Pages coded in the
encodings commonly used for Japanese and Chinese can be displayed
with that font too. If the browser/systems knows the encoding
being used, it remaps that characters too Unicode before using
the font. (Only if the page uses some character that is lacking
in the font - and thus most probably lacking in Unicode 2 - the
display will fail.)
Bertilo Wennergren
"Alan J. Flavell":
> On Wed, 13 Jun 2001, Bertilo Wennergren wrote:
> > My MSIE6 does, and I'm sure MSIE5 would too, on my system. Your
> > problem is most probably due to not having installed the entire
> > Multilanguage support of Windows (check in the Control Panel).
> > Or perhaps you have not installed all the relevant modules of
> > MSIE5 (it's highly modular...).
> My experience in the past has been that if a component is missing,
> which is needed for browsing a particular page, then IE5 offers to
> install it when you try to browse that page. But this time it didn't
> do that.
You can't rely on that functionality. And if it's the Multilanguage
support of Windows itself that's missing then you won't get any such
offers.
> > Arial Unicode MS is really a great thing.
> Yes, but MSIE doesn't really need it, because it can pick and choose
> from a selection of fonts.
Yes, but Arial Unicode MS acts like a nice basic package of characters
that you can install once and for all without having to go looking
around
for all sorts of exotic fonts.
For those who don't already have it the font can be downloaded here:
<http://office.microsoft.com/downloads/2000/aruniupd.aspx>
MS used to have some legal stuff on the page that said you can only
use the font if you have this or that licence. I can't seem to find
anything of that kind on the new page though. Have they actually
made it free for all?
Eric Jarvis
wrote...
quite possibly
it has been available for free elsewhere for a long time...legal
or not that has pretty much put MS in a situation where they
can't really charge for it
Darrel E. Knutson
Thanasis Kinias <firstiniti...@optimalco.com> wrote:
>Even if I were
>using a Macintosh, the odds that I would have such an esoteric font are
>slim.
If you run Mac OS 9.0 or 9.1 the Language Packs for different kinds of
writing systems can be installed at any time from the System CD. These are
in Unicode, but the closest to Tibetan I have seen is Devanagari, which I
used to write a few texts in Nepali. Again, the Unicode stuff is not
automatically installed and I would assume that even fewer people have done
this as there is a conflict with WorldScript when using 9.0 as the Classic
environment under Mac OS X. Even among Macintosh users I would estimate
that fewer than 1% have a south or central Asian font installed.
Even BBC World Service announces their Nepali radio programs on their site
with Nepali text written in a moderately large GIF image.
Darrel
--
Internet Trainer <http://darrel.knutson.com/>
Roedingsmarkt 14 <mailto:dar...@knutson.com>
20459 Hamburg, Germany GSM/D2: +49 (0)173/2088764
Thanasis Kinias
> Arial Unicode MS is really a great thing.
What about Linux users interested in reading "exotic" languages? I haven't
found any equivalent for this font.
Also, does anyone know of a more attractive font with similar Unicode
coverage? Arial's kanji are _really_ ugly. I'd really like a Roman/Mincho
style.
Code2000 has coverage at least as good as Arial Unicode MS, but its Latin
characters are even uglier than Arial's.
This has gotten a bit O/T from HTML . . .
Bertilo Wennergren
"Thanasis Kinias":
> Bertilo Wennergren wrote:
> > Arial Unicode MS is really a great thing.
> What about Linux users interested in reading "exotic" languages? I
haven't
> found any equivalent for this font.
Is it impossible to use TrueType fonts on Linux?
> Also, does anyone know of a more attractive font with similar Unicode
> coverage? Arial's kanji are _really_ ugly. I'd really like a
Roman/Mincho
> style.
I doubt there is even a single font out there that has as impressive
coverage as Arial Unicode MS.
> Code2000 has coverage at least as good as Arial Unicode MS, but its
Latin
> characters are even uglier than Arial's.
Oh, Code2000... Where can I get it?
> This has gotten a bit O/T from HTML . . .
It's still close to the practical side of HTML, but perhaps not
the _authoring_ side, rather the consumer side of it.
Richard Watson
> Is it impossible to use TrueType fonts on Linux?
No it's not at all impossible - it's quite easy.
But MS fonts aren't free except with an MS operating system.
--
Richard Watson GnuPG/PGP:0x55227960 http://www.doilywood.org.uk
e-mail:ric...@doilywood.org.uk jabber:ric...@jabber.doilywood.org.uk
Alan J. Flavell
> I doubt there is even a single font out there that has as impressive
> coverage as Arial Unicode MS.
The design of HTML in this area deliberately decouples the character
representation from the font technology. One of the benefits of that
(unless misguided authors overrule it with inappropriate FONT FACE or
CSS specifications) is that the browser is free to mix and match any
fonts to which it has access, according to their character repertoire.
MSIE takes advantage of this opportunity (as does iCAB on the Mac).
NN4 of course doesn't.
> Oh, Code2000... Where can I get it?
Sounds like time for another trip to Alan Wood's pages. Start at
http://www.hclrss.demon.co.uk/unicode/
where you'll find your way to http://home.att.net/~jameskass/
and a list of links which are all called "Click Here" (grrrrrrrrrrrrrrrr)
George Lund
Alan J. Flavell <fla...@mail.cern.ch> writes
>The design of HTML in this area deliberately decouples the character
>representation from the font technology. One of the benefits of that
>(unless misguided authors overrule it with inappropriate FONT FACE or
>CSS specifications) is that the browser is free to mix and match any
>fonts to which it has access, according to their character repertoire.
Surely it would be sensible if a browser could do the mixing and
matching anyway if a CSS-specified font didn't contain the requisite
glyphs? Perhaps even going through the list of alternatives then
looking back up the cascade until if found a font than did... one day
maybe :-)
--
George Lund
Bertilo Wennergren
"Richard Watson":
> "Bertilo Wennergren" <bert...@bertilo.se.fm> writes:
> > Is it impossible to use TrueType fonts on Linux?
> No it's not at all impossible - it's quite easy.
> But MS fonts aren't free except with an MS operating system.
But it does seem that Arial Unicode MS is now free, and you
can just download it if you want it.
Thanasis Kinias
> "Bertilo Wennergren" <bert...@bertilo.se.fm> writes:
>
>> Is it impossible to use TrueType fonts on Linux?
>
> No it's not at all impossible - it's quite easy.
OK. I haven't tried yet; I'm new to Linux. My Windows installation
committed suicide when I swapped out my Winmodem for a real modem and I
haven't got it back on its feet yet to steal the fonts.
What about OpenType fonts?
> But MS fonts aren't free except with an MS operating system.
Actually, I recently D/L'd the latest versions of Arial Unicode and the
"core Web fonts" or whatever they call Verdana, Georgia, etc. They're now
free online [1]. Note that these are just TrueType, not OpenType (as come
with Windows 2000); I guess they're not really the latest after all.
1. Please note the restraint: I refrained totally from any cynical
commentary on MS's motives for giving them away free.
Thanasis Kinias
> The design of HTML in this area deliberately decouples the character
> representation from the font technology. One of the benefits of that
> (unless misguided authors overrule it with inappropriate FONT FACE or
> CSS specifications) is that the browser is free to mix and match any
> fonts to which it has access, according to their character repertoire.
>
> MSIE takes advantage of this opportunity (as does iCAB on the Mac).
>
> NN4 of course doesn't.
MSIE5 botches it completely in my experience. I did a test page last
winter (no longer available) on name translation which mixed Old English,
Old Norse (Latin, not Fuþark), Latin, Greek, and Russian. I laboriously
told MSIE to use Arial Unicode for every writing system (there's no "use
this for everything" option). And the result was that it mixed up the
fonts, even though one would have sufficed. IIRC all extended Latin
characters were done in Times New Roman. This is unacceptable, when in the
word Mjölnir spelt in Old Norse spelling (hooked o vice ö), one character
is displayed in a different font than the rest of the word.
In NN4, of course, there were lots of '?'.
BTW, even Arial Unicode can't handle Fuþark, unfortunately. And Windows
NT's Character Map doesn't even include that section of Unicode.
Andreas Prilop
"Alan J. Flavell" <fla...@mail.cern.ch> wrote:
> One of the benefits of that
> is that the browser is free to mix and match any
> fonts to which it has access, according to their character repertoire.
> MSIE takes advantage of this opportunity (as does iCAB on the Mac).
> NN4 of course doesn't.
I don't quite understand what you mean. In Netscape 4, you can also
choose different typefaces for, say, Cyrillic and Greek.
Alan J. Flavell
> > is that the browser is free to mix and match any
> > fonts to which it has access, according to their character repertoire.
> > MSIE takes advantage of this opportunity (as does iCAB on the Mac).
> > NN4 of course doesn't.
>
> I don't quite understand what you mean. In Netscape 4, you can also
> choose different typefaces for, say, Cyrillic and Greek.
Neither of which will be used when the Cyrillic and Greek text is
presented as utf-8. That was my point. If you want to display a
document that contains both Cyrillic and Greek, then with NN4 you need
to provide a font which contains them both, and configure it as the
"Unicode" font, and present the document as utf-8 (even if it's in
fact coded in us-ascii, but you know that already).
MSIE, on the other hand, can pick the Cyrillic from one font, and the
Greek from another, if you wish, and doesn't need the document to be
presented with utf-8 coding.
Andreas Prilop
"Alan J. Flavell" <fla...@mail.cern.ch> wrote:
> Neither of which will be used when the Cyrillic and Greek text is
> presented as utf-8.
But this applies only to the MS Windows version. Since you referred also
to iCab (which has no Windows version), I assumed that you meant all
operating systems. Before Mac OS X, there were only separate 8-bit
character sets (MacCyrillic, MacCE, ...) and any browser has to compose
UTF-8 pages from different fonts.
Alan J. Flavell
> "Alan J. Flavell" <fla...@mail.cern.ch> wrote:
>
> > Neither of which will be used when the Cyrillic and Greek text is
> > presented as utf-8.
>
> But this applies only to the MS Windows version.
Aha - thanks for pointing that out
> Since you referred also
> to iCab (which has no Windows version), I assumed that you meant all
> operating systems.
Sorry, iCAB is the only vaguely-recent browser version that I have on
a Mac. Prior to that, the beast in question had an obsolete Netscape
version (2.0, I think) as its only web browser (it's a geriatric
68k-based box, which I saved from the scrap heap as I still needed
something to manage the residual Appletalk network).
I thought I'd always made it clear that I referred to you whenever I
needed the Mac version of an answer...
And X Netscape 4 is yet another mystery. So I suppose I should have
referred explicitly to "Win Netscape 4.*" as being the problem with
which I was familiar in this area. Sorry about that.
Richard Watson
> Richard Watson wrote:
>
> > "Bertilo Wennergren" <bert...@bertilo.se.fm> writes:
> >
> >> Is it impossible to use TrueType fonts on Linux?
> >
> > No it's not at all impossible - it's quite easy.
>
> OK. I haven't tried yet; I'm new to Linux. My Windows installation
> committed suicide when I swapped out my Winmodem for a real modem and I
> haven't got it back on its feet yet to steal the fonts.
Xv4 handles truetypes out of the box. Most distributions have some
kind of font manager to add them automagically too.
> Actually, I recently D/L'd the latest versions of Arial Unicode and the
> "core Web fonts" or whatever they call Verdana, Georgia, etc. They're now
> free online [1].
Anyone got an url?
Philip Newton
wrote:
> Thanasis Kinias <firstiniti...@optimalco.com> writes:
>
> > Actually, I recently D/L'd the latest versions of Arial Unicode and the
> > "core Web fonts" or whatever they call Verdana, Georgia, etc. They're now
> > free online [1].
>
> Anyone got an url?
http://www.microsoft.com/typography/fontpack/default.htm
Cheers,
Philip
--
Philip Newton <nospam...@gmx.li>
That really is my address; no need to remove anything to reply.
If you're not part of the solution, you're part of the precipitate.