From: James Kass (jameskass@att.net)
Date: Mon Apr 25 2005 - 09:18:25 CST
"Arcane Jill" writes,
> Er ... because it doesn't work. That's exactly what I'm saying. I'm saying you
> CAN'T use the Unicode characters for musical symbols. Believe me, I've tried
> it. All that happens is that 99% of all web browsers on the planet display
> unknown-character glyphs in place of each Unicode musical character. And that
> doesn't help anyone.
Is there an on-line example of one of your efforts?
> I don't agree that that's a safe assumption at all. Besides, please don't
> strawman me - I never said anything about Limbu or Buhid. (In fact, I've never
> heard of them).
They are Unicode script ranges in the BMP. Improperly configured
browsers and/or poorly thought-out HTML files may well display
missing glyphs when presented with Limbu or Buhid text.
> Fact is, when I publish a web page, I want it to be viewable by
> /everyone/. Why? Because when, as a surfer of the internet, I stumble across a
> page which won't display properly in my browser, and I read something like "you
> need to download and install such-and-such in order to view this page", I tend
> to think: "Well I won't bother then". As, I believe, do most people.
Indeed, for various reasons most people probably don't bother. Which
may mean that most people miss out on some pretty cool stuff. (And
also avoid some real junk, naturally. There's no sure-fire way to know
the difference in advance.)
> Yes, of course. And I /want/ to use Unicode. Badly.
Why not use it well, instead? (BIG SMILE)
> But I'm not going to if my
> web site won't display properly for 99% of all users who look at it.
This may be a slight exaggeration, but your point is taken.
> > So, every time I look at the Klingon page, my system will automatically
> > download and install a font I already have? Wouldn't that slow everything
> > down? I already seem to be on 'molasses dot com'.
>
> I don't see why it would do that. That would be terribly bad design.
Then you *do* see.
> > There really should be no problem. Provide your users with font links.
>
> Well, actually, this /is/ the problem. If my web page says "In order to view
> this page, you must first download and install such-and-such-a-font", then most
> people will simply go away. I know I would.
Then you don't get to see the page. And, I confess, sometimes neither
do I. Whether I bother with the download depends on how much I want
to view the page and the nature of the download.
> > Or, get a font and embed it.
>
> Okay, now we're getting there. How?
> (Actually, I think Jon has already answered that).
He has. One solution to two proprietary embedding methods is to
make two of everything. That's a work-around solution rather than
an ideal one, though.
> > PDFs work, too.
>
> True, but kinda not revelant.
But, Unicode plane one musical notation characters can be properly
embedded in PDFs. Of course, the user must have downloaded and
installed a PDF viewer in order to see the file.
> I suspect that you may have misunderstood me. I /don't/ have a practical
> problem which needs a workaround. Rather, I cited a /hypothetical/ problem, as
> an /illustration/ of a reason of why one might choose not to place uncommon
> Unicode characters in an HTML web page.
Since you'd mentioned musical notation symbols specifically, I was kind
of hoping you had a practical problem needing resolution.
> Sorry if that wasn't clear. I don't buy
> "throw away HTML and use PDF instead" as even remotely practical - but I'm not
> looking for a fight.
If the problem had been practical, I'd point out that PDFs
work now, and there'd be no need to wait for browsers to
catch up.
http://home.att.net/~jameskass/musictest.pdf
> I'm happy with Jon Hanna's answer, which is basically that
> the W3C are on the case, and we just have to wait for browsers to catch up.
I'm happy with Jon Hanna's answer, too. However, by the time
browsers catch up with those recommendations, we may find that
core fonts on popular operating systems have caught up, as well.
Best regards,
James Kass
This archive was generated by hypermail 2.1.5 : Mon Apr 25 2005 - 09:18:41 CST