Re: Symbol font with all four weights
- From: "Alan J. Flavell" <flavell@xxxxxxxxxxxxxxxxx>
- Date: Tue, 6 Jun 2006 22:11:39 +0100
On Tue, 6 Jun 2006, Alan Wood wrote:
Alan J. Flavell wrote:....
Where /can/ one find updated versions of the Symbol font
correspondence table that's at
http://www.unicode.org/Public/MAPPINGS/VENDORS/ADOBE/symbol.txt ?
I have a page that may be suitable at:
http://www.alanwood.net/demos/symbol.html
Thank you! The data on that page will be very useful, and duly
acknowledged where used.
I have identified what I hope are the proper Unicode equivalents for
all except one of the Symbol characters.
I was expecting at least three - the ones which disambiguate
copyright-serif/sans, trademark-serif/sans, and registered-serif/sans;
but, instead, you seem to have merged those, and come up with another
problem case (:-)
I will take some time to mull this over, and let you know how well it
falls into place.
So, I'll go over that data - but meantime, there's the more general
issue that I feel I need to take up. No offence intended, but, the
way that I read it, anyone who has been already fooled by the false
gold of <font face="Symbol"> is at risk of remaining fooled after
reading the text of your page. Your "not very reliable" really ought
to make it clear that it's absolutely contrary to published
specifications, and thus it can only 'work' (I mean, "do what the
misguided author intended") in so far as it relies on browser bugs.
So, when you say:
|For Windows, Netscape 4 and Internet Explorer allow the font face to
|be specified in HTML, and by specifying Symbol the basic Greek
|alphabet and many other special characters can be displayed. This
|technique is not very reliable and is deprecated in the HTML 4.0
|Specification
then I'm afraid I would have to rate that as *far, far* too low-key,
and even slightly off-target...
The fact that the <font...> markup is deprecated in HTML4 does not
seem to me to have *anything* to do with its misuse for faking
non-standard characters. <font face...> was codified in
HTML/3.2(retch) for the purposes of proposing cosmetic changes to the
font. It was /never/ proposed by the W3C for the purposes of
extending the character repertoire - because the W3C already *had*
spelled out the right way to extend the character repertoire, and
RFC1866, RFC2070, and subsequently the i18n sections of HTML4, were
the specifications for how to do that.
<font...> is deprecated in HTML4 *because the presentation belongs in
the stylesheet*, via { font-family: ... ;}. It's not deprecated there
in order to rule out the use of symbol-type fonts, **because they were
never ruled-in**.
So, the presentation belongs in the stylesheet. But to misinterpret
that as using { font-family: Symbol; } in CSS would be *at least as
bogus* as using <font face="Symbol"> in old HTML, no matter if it
happened to produce the intended effect in some sloppily-implemented
browser.
By "sloppily-implemented browser" I don't only mean the operating
system component that thinks it's a web browser - some other browsers
apparently feel constrained to break with the specification in order
not to upset the misguided users of that thing - at least in their
"quirks" mode... but one would be foolish to rely on that, when
there's a properly documented way to do it. *I really must state in
the strongest terms that we would be remiss if we were to allow others
to think we condoned such foolishness*, which is why I'm upset by the
low-key nature of the warnings on your page.
Right from the start of published specifications for HTML, i.e RFC1866
for HTML/2.0, it was specified that characters in HTML represented
what they *meant* in iso-8859-1 (in HTML2), and said that it would be
iso-10646 for the future, as indeed (modulo the evolution of
iso-10646/Unicode) it has turned out since. RFC2070 spelled out the
significance of this, and its author produced the seminal document
"font face considered harmful" around the same time - which I see has
re-appeared at http://alis.isoc.org/web_ml/html/fontface.en.html - and
a good thing too (I must update my references).
The use of e.g <font face="symbol">a</font>, as exemplified in your
page, *is simply bogus*, since the Symbol font does not contain a
letter "a": any specification-conforming client agent would either
rate the rendering as impossible - since the font does not contain the
required character - or would go and find some better font which *did*
contain a letter "a" in order to render the content "according to
specification".
So, going back to your quote, I have to say that I would have to
express it somewhat along the following lines:
"By specifying Symbol, some sloppily-implemented browsers could be
fooled into displaying Greek and other special characters, but this
has never been a documented feature of HTML: it has been contrary to
the published specifications from the outset, it was specifically
warned against by the author of RFC2070, and has for some time been
tightened-up in specification-conforming browsers."
I frankly would much prefer it if your table never did show samples
like the cited <font face="Symbol">a</a> - which is rendered correctly
by my browser as a latin lower-case "a" of course, not as the Greek
alpha which it wrongly claims to represent - nor indeed the numerical
character references like <font face="Symbol">Ò</font> - which my
browser correctly renders as O-grave, rather than the misguided
intentions of its author - since, no matter how strongly the
accompanying wording warns against it, the examples are likely to fool
some impressionable readers. (Especially those accustomed to the
standards-nonconforming extrusions from MS Word).
This is a working document (derived from that out of date
Adobe/Unicode mapping table, prior to considering your excellent data)
which tries to avoid any suggestion of the misuse under discussion
here: http://ppewww.ph.gla.ac.uk/~flavell/unicode/symbol.html
Good heavens, it's now at least 8 years that I've needed to rant about
this, as I see from one of my web pages:
___
/
<p> In 1998 I commented here how depressing it was that authors
were still resorting to tricks with non-standard
<code>&#number;</code> usage, or <a
href="fontface-harmful.html">FONT FACE</a>=Symbol, and getting the
visual results they wanted on the mass-market browsers, instead of
writing correct HTML4. But, as expected, browsers are progressively
implementing the published specifications, and some of these tricks
cease to produce the desired results. And be warned again that font
tricks will not be indexed correctly by robots, and probably won't be
handled correctly by screen readers and speaking browsers: in other
words, they only give a visual impression of working, while being
completely unsound at the more fundamental level.
</p>
\___
As I say, this isn't meant to be personal - I've found your pages to
be most useful; but on this issue I'd have to say that the underlying
specifications have to take first place, even at risk of causing
offence. Obviously your pages are your own, and I can't direct you
what to say, but I would hope to see the pages consistent with
published specifications, even in cases where non-conforming usages
were discussed with suitable caveats.
best regards
.
- References:
- Re: Symbol font with all four weights
- From: Character
- Re: Symbol font with all four weights
- From: Alan J. Flavell
- Re: Symbol font with all four weights
- From: Alan Wood
- Re: Symbol font with all four weights
- Prev by Date: Re: gpl or public domain fonts
- Next by Date: Looking for Bickley Font Free
- Previous by thread: Re: Symbol font with all four weights
- Next by thread: Re: Symbol font with all four weights
- Index(es):
Relevant Pages
|
