Representing Greek: the sequel

Here you can discuss all things Ancient Greek. Use this board to ask questions about grammar, discuss learning strategies, get help with a difficult passage of Greek, and more.
Post Reply
Titus Marius Crispus
Textkit Neophyte
Posts: 78
Joined: Wed Jul 14, 2004 2:00 am
Location: San Antonio, TX, USA

Representing Greek: the sequel

Post by Titus Marius Crispus »

Amazing! Fantastic! Fabulous!

I have just completed the setup of KeyMan with ClassicalGreek as described here.

I can now switch back and forth from the Roman alphabet to the Greek alphabet with a simple keyboard shortcut (ctrl+alt+z for me, but you can pick whatever you like!)

The characters are Unicode, of course, so they won't be visible to users without Unicode support.

Testing:
αβξδεφγηιξκλμνοπθρστυνωχψζ

Can you see those?

User avatar
klewlis
Global Moderator
Posts: 1668
Joined: Tue Jul 29, 2003 1:48 pm
Location: Vancouver, Canada
Contact:

Post by klewlis »

i can see them. :)
First say to yourself what you would be; then do what you need to do. ~Epictetus

Bert
Textkit Zealot
Posts: 1889
Joined: Sat May 31, 2003 2:28 am
Location: Arthur Ontario Canada

Post by Bert »

What is the reason that two keys produce a xi and two keys a nu?

Emma_85
Global Moderator
Posts: 1564
Joined: Thu Jul 03, 2003 8:01 pm
Location: London

Post by Emma_85 »

I seen them too
:-)

Paul
Textkit Zealot
Posts: 708
Joined: Sun Jun 15, 2003 4:47 pm
Location: Maryland
Contact:

Post by Paul »

Hi Titus,

I've used Keyman for a few years now. Rarely is there a font problem in the 'normal' greek unicode range (your test case). But here's an example from the 'extended' greek unicode range - lower case alpha with acute.



It looks to me, at least in 'Post a reply' preview, like a 'box' glyph. This is standard browser behavior when the character, decimal 8049 in this case, does not map to any glyph in the effective font.

Please note that I run my IE with a unicode font (Arial Unicode MS). I am regularly able to see greek glyphs, both in the normal and extended greek unicode ranges. But here, at least in preview, I do not.

We'll see what it looks like after I post it.

Cordially,

Paul

Emma_85
Global Moderator
Posts: 1564
Joined: Thu Jul 03, 2003 8:01 pm
Location: London

Post by Emma_85 »

Still looks like a box to me...

Titus Marius Crispus
Textkit Neophyte
Posts: 78
Joined: Wed Jul 14, 2004 2:00 am
Location: San Antonio, TX, USA

Post by Titus Marius Crispus »

I see the alpha with an acute. I'm using Firefox with WinXP.

Paul
Textkit Zealot
Posts: 708
Joined: Sun Jun 15, 2003 4:47 pm
Location: Maryland
Contact:

Post by Paul »

Hi,

Yeah, Firefox on Win 2K also shows proper glyph. IE6 on Win2K does not. (A strange reversal; usually IE gets the fonts right and Mozilla doesn't).

But the larger problem is that Textkit-ers have different browsers across different platforms. Not all of us can see Unicode characters glyphs correctly.

However unpleasant SPIONIC looks (and its appearance, too, varies by browser/platform), it remains the more democratic of the several solutions.

Cordially,

Paul

Eureka
Textkit Zealot
Posts: 741
Joined: Tue Feb 17, 2004 3:52 am
Location: Melbourne, Australia

Post by Eureka »

Surely, if we can all install SPIONIC, then we could all install Gentium instead (it has all the accent markers).

Then all we’d need is for the font to be added to TextKit…

Paul
Textkit Zealot
Posts: 708
Joined: Sun Jun 15, 2003 4:47 pm
Location: Maryland
Contact:

Post by Paul »

Eureka wrote:Surely, if we can all install SPIONIC, then we could all install Gentium instead (it has all the accent markers).

Then all we’d need is for the font to be added to TextKit…
Hi,

Unless I am sorely mistaken - always a possibility - it's not that simple. The 'beauty' of SPIONIC is that it knows how to convert betacode (or nearly betacode) to the appropriate greek glyphs. That is, anyone using an ASCII text editor can create greek glyphs.

There is no such capability in the several Unicode fonts, e.g. Gentium. Gentium can properly show the glyph 'greek small alpha plus oxia' in response to the character 8049, but it can't convert the two ASCII characters a/ to the same glyph.

Try it yourself. Create a syntactically correct HTML file with this SPAN tag in the BODY:

<SPAN STYLE="font-family:spionic">a/</SPAN>

Open it in, say, IE. You should see alpha with an acute accent. Now change the tag to this:

<SPAN STYLE="font-family:gentium">a/</SPAN>

and reload the web page. You should see 'a/'.

In fine, not only is it a matter of having the right font, but of being able to generate the proper characters. NB: 'character' does not mean 'glyph'; it means the underlying numeric value which is mapped to a glyph by a font.

Cordially,

Paul

Eureka
Textkit Zealot
Posts: 741
Joined: Tue Feb 17, 2004 3:52 am
Location: Melbourne, Australia

Post by Eureka »

Oh. Although I've never programmed HTML in my life, I can see what you mean.

However, I would think that, rather than entering it as:

<SPAN STYLE="font-family:gentium">a/</SPAN>

You would enter it as:

<SPAN STYLE="font-family:gentium">ά</SPAN>

[Obviously except with ά being the glyph (or moreover its unicode number), rather than the two SPIONIC characters.] :)


But then again, I don't really know.

Paul
Textkit Zealot
Posts: 708
Joined: Sun Jun 15, 2003 4:47 pm
Location: Maryland
Contact:

Post by Paul »

Eureka wrote: You would enter it as:

<SPAN STYLE="font-family:gentium">ά</SPAN>
Yes, exactly. And that's my point. To effectively use a Unicode font, Textkit-ers would need to be able to enter Unicode data.

I suppose everyone could avail themselves of Keyman or something else like it. But I don't think everyone will be happy about this. I know there are some Unicode issues in older Windows (98, ME). Also, I don't know if this approach lends itself to non-Windows platforms.

Personally, I'd be much happier with a Unicode based solution. In fact, when I was new to Textkit I raised this very issue. But now it seems to me that SPIONIC is the simplest, one size fits all solution.

Cordially,

Paul

Geoff
Textkit Fan
Posts: 345
Joined: Fri Aug 22, 2003 2:30 pm

Post by Geoff »

Unicode will come.

You will be assimilated. Resistance is futile. 8)

eris
Textkit Member
Posts: 119
Joined: Thu May 13, 2004 8:19 am

Post by eris »

Just wanted to let you know that I see your Greek characters, including Paul's
lower case alpha with acute. I'm using Netscape on a Mac (currently ver. 10.2.8

Thanks for the link to KeyMan. I'll try out the program at home with my pc. I don't think it'll work for Macs.

ThomasGR
Textkit Enthusiast
Posts: 444
Joined: Mon Aug 23, 2004 8:49 pm

Post by ThomasGR »

Isn't better to drop all these dasia and oxia etc. and use only one sign for the accent? Than you wouldn't have any problems with Unicode. I think, the ancient Greeks didn't use these dasia and oxia nonsense but it is a late invention of the Alexandrian grammarians.

chad
Textkit Zealot
Posts: 757
Joined: Tue Jul 22, 2003 2:55 am

Post by chad »

hi thomas, that would be fine for people who use the stress accent to pronounce greek.

the accents represent different pitches though for people who use the pitch accent pronunciation... it wouldn't make sense to use one symbol for 3 different pitch accents...

ThomasGR
Textkit Enthusiast
Posts: 444
Joined: Mon Aug 23, 2004 8:49 pm

Post by ThomasGR »

It is much more useful to learn vocabulary and grammar, then waisting my time with pitches and how to pronounce them all. I haven't met anyone who could manage it, e.g. when to use circumflex and how to speak it. It's a totally waist of time and perhaps also money. That explains why the ancients didn't bother too much with that and soon dropped them all to one single sign for the stress accent. I myself often do not write even the stress accent.

Post Reply