Symbol font does not work in Netscape

[Achernar]

The Symbol font does not now work in Netscape, and it never has. Anything rendered in the Symbol font shows up in the default font, which is for me (and probably most people) Times New Roman. I am sure that this is not a problem with the way that the font is installed on my machine, and it's not a problem with my settings. (Anyone else with a copy of Netscape may prove me wrong by checking it.) I may be the only person now who uses this browser, but I think more people will eventually change over. And anyway, it's poor practice to post something that some people can't read.

So I suggest people stop using the Symbol tag. There is a perfectly good alternative for the Greek alphabet. Use α (α) through ω (ω) for the lowercase letters, and Α (Α) through Ω (Ω) for the uppercase letters.

By the way, when I say Netscape here, I of course mean the browser called Netscape, not the browser made by a company called Netscape (which would be Navigator). Netscape 6 is not an upgrade to Navigator 4.x - it is a different browser, based on a different engine. I believe that Mozilla has the same engine, so the problem may lie there too.

[Arnold Winkelried]

FYI - I don't see greek characters in Opera 5.0 on MacOS 9.2 - I see square boxes.

[Achernar]

Okay, then I guess there is no good alternative. I personally feel less guilty about my way, because the people in charge of telling browsers what to do say that's the way you should do it.

In that case, I suggest people forgo the Greek alphabet entirely, and spell out the letters instead. Others should know what you're saying.

[Arnold Winkelried]

Yes, one can always use something like what you suggest, e.g. use {pi} for the greek letter pi:
The area of a circle with radius r is {pi}r²

[Achernar]

Yeah, that's a good suggestion. I should probably also point out that I personally have no problem with the current situation - I know the Greek alphabet well enough to see what's going on. However, I'm sure other people just won't realize what's happening, and maybe not understand what you've written.

[waterj2]

And anyway, it's poor practice to post something that some people can't read.

I think the fault would lie with Netscape or whoever made the browser for not following the standards. I'm still trying to figure out how they could disable "symbol" while still allowing other font. Or does Netscape force everything to be in the default font?

I think the usage you suggest would be more of a problem, as it would not be compatible with older browsers. I would consider it better practice to make things unreadable to newer browsers that deviate from established standards (and I think the font tag is well established) than older browsers which follow what the standards used to be.

[Achernar]

I agree that the fault lies with Netscape. However, two things.

First, Netscape is following the standards. It's everyone else who's gotten it wrong, in this regard. Apparently Symbol font is not an ISO-8859-1 encoded font, and as such, it shouldn't (technically) be rendered.

Second, I don't know which all browsers support my method (which you'll notice I've abandoned) but I don't think that Opera 5.0 is all that old. If it is an old browser/new browser issue, though, then while I agree with you that my method is more problematic today, I'm trying to think ahead. I can't imagine people hanging on to old browsers for a terribly long time after they've been made obsolete.

The font tag, also, is deprecated as of HTML 4. If it were up to me we wouldn't be using it at all.

[Urban Ranger]

Originally posted by Arnold Winkelried
FYI - I don't see greek characters in Opera 5.0 on MacOS 9.2 - I see square boxes.

That seems to be a problem with Opera then, since those seem to be standard (i.e. W3C) character entities. Maybe the newer Opera 7.0 has better support?

[Urban Ranger]

Originally posted by Achernar
The font tag, also, is deprecated as of HTML 4. If it were up to me we wouldn't be using it at all.

Can't really define CSS classes on a BBS.

[Mangetout]

Originally posted by Arnold Winkelried
Yes, one can always use something like what you suggest, e.g. use {pi} for the greek letter pi:
The area of a circle with radius r is {pi}r²

Maybe we need a pi smiley...

[Achernar]

Originally posted by Urban Ranger
Can't really define CSS classes on a BBS.

Why not? The following tags have the same effect:

<font face="arial">
<span style="font-family: arial">

You're also right about Opera being deficient when it comes to rendering special characters. Both Opera and IE are not as robust as Mozilla/Netscape when it comes to this.

[mnemosyne]

I use Mozilla, and I was both able to see your symbols, and unless the only (few) math/physics threads i've read all used your method, I was able to make out the symbol font. I think the font my browser defaults to is Arial or Helvetica. I don't really understand it, but in my Fonts section, if says serif fonts are times, sans-serif are helvetica (the SDMB does not have serifs for me), cursive and fantasy fonts are both set to "avant-garde gothic" and monospace fonts are courrier. I also have the "allow documents to use other fonts" box checked, so maybe that's why I see the symbols. I don't know. Either way, every font on the list of available ones are all identified with that 8859-1 tag.

I have no idea if that helps or not. I NEVER use the symbol font, and I either write out my greek letters (e.g. alpha) or I substitute something else in, like ug for micrograms. Close enough for me.

[mnemosyne]

Oh, and I use the Linux-based Mozilla, not the Windows one. I don't think that makes a difference, but it might. I do have mozilla in Windows, but I never boot into it, and so its doubtful that I'll check!

[Chronos]

Apparently Symbol font is not an ISO-8859-1 encoded font, and as such, it shouldn't (technically) be rendered.

On the other hand, aren't fonts supposed to be referenced by name? If a web page tells my browser that something should be in the font called "symbol", and I have a font called "symbol" on my computer, then I darn well expect my browser to figure it out. Heck, if a web page asks for a font called "xyzzy", and I happen to have a font by that name, I'd expect it to make an honest effort on that, too. I recognize that my computer might not have that font, in which case I have no specific expectations as to its behaviour, or that my computer might even have a different font with the same name, in which case difficulties could obviously arise. But that's not an argument for browsers to restrict themselves in what fonts they support; it's an argument for web designers to be careful in what fonts they use. I imagine, though, that most modern computers have a font called "symbol", which appears essentially the same for each, so this is not an issue.

As for the font tag being depreciated, what's that supposed to mean? It's obviously still widely used, and it's part of the official standard (even if an old part). That sounds to me like it's still official in every sense of the word.

[Achernar]

Originally posted by Chronos
If a web page tells my browser that something should be in the font called "symbol", and I have a font called "symbol" on my computer, then I darn well expect my browser to figure it out.

Not really. Ideally, user agents should conform to the specifications for the document type that they're rendering, and page designers should stick within these specifications as well. But this immediately raises the question, why should the specifications only allow approved fonts?

I don't know for sure, and I can't find a good answer on w3.org; however, I can take a guess. It's a matter of accessibility. If your page is supposed to be a web page, then it needs to be accessible, which means that if someone doesn't have that font, the content should remain effectively the same. Fonts are a matter of style, not content. So a font gets ISO approval that an A actually looks like an A, and so on. If you have a font that intentionally doesn't follow this convention, like the Symbol font, then it's even worse. How is somebody using a non-visual browser supposed to know what you're trying to say?

Here's what deprecated means, from the HTML 4.01 specs:

A deprecated element or attribute is one that has been outdated by newer constructs. Deprecated elements are defined in the reference manual in appropriate locations, but are clearly marked as deprecated. Deprecated elements may become obsolete in future versions of HTML.

User agents should continue to support deprecated elements for reasons of backward compatibility.

It's in the official standard only in the sense that it's officially outdated. It does not appear in the HTML 4.01 Strict Document Type Definition, and your page won't validate as such if you use it.

Now, in my opinion, none of this has any bearing whatsoever on whether we should use the Symbol font on this board. My complaint with it was motivated only by the fact that it doesn't work, not whether it should work or not.