[Spellyans] Front unrounded vowels, was: The quantity system

[Michael Everson]

At 10:51 +0200 2008-06-25, Daniel Prohaska wrote:

>"There are many fonts and keyboard drivers which 
>do not offer y-circumflex or y-acute."
>
>I should say there are probably many fonts and 
>keyboard drivers that don't offer y-diaeresis.

No, because when people make fonts they typically 
put in all the "standard" characters. Y-diaeresis 
is "standard" in both Mac Roman and Windows 1252. 
That letter is safe. Y-acute is simply not 
available in many fonts. And many fonts are Mac 
fonts. Choosing y-acute is simply guaranteed to 
cause problems for people. Choosing y-diaeresis 
is the better option.

Keyboard inputting software is a different 
matter. Each computer user could download a 
special Cornish keyboard driver. Easy. But 
thousands upon thousands of fonts do not have 
both characters. Having missing characters is 
MUCH WORSE than

>Same problem. I don't think font and keyboard 
>limitations should put us off designing a 
>workable orthography.

There is nothing unworkable about ë/ÿ in terms of 
universal availability in fonts. There *is* 
something problematic about both ê/–y and é/´y.

And there's nothing wrong with the diaeresis.

>We'll just have to make sure to explain to and 
>provide users with the necessary software. Also, 
>a majority of users has Microsoft, and designing 
>an orthography to Mac specifications, because 
>you happen to work with that system, is a 
>little, well, should I say, self-centred ;-)

I am not thinking only of myself. And I am as 
unimpressed by the "Microsoft Is Big" argument as 
I was by the "Kebmyn Is Big" argument.

In point of fact Macs are used very widely in the 
production of published materials. And market 
share is growing.

>"As I said, consider <schön> which is pronounced 
>[Sø:n] in Standard German and [Se:n] in certain 
>dialects (and Yiddish). We're not marking the 
>length in <bës>. We're indicating that it can be 
>pronounced [be:z] or [bi:z]."
>
>I know, and I know it works, but that doesn't 
>mean I particularly like it. I prefer the 
>circumflex because for me that is an unambiguous 
>length marker, while the diaeresis isn't.

Actually, Dan, we are using the circumflex for 
something else: stret is [strEt] and strêt is 
[stre:t]. So it the circumflex is not even 
*available* for use for the bÿs/bës words.

>In the case or bys/bes, we are marking length; 
>bys "unil" has a short vowel by(diaeresis)s 
>"world" has a long vowel.

We are only marking length in bÿs because of the 
i/y distinction in monosyllables; bes and bës are 
both inherently long so length per se is not 
marked in bës.

>"The fact that SWF bes [be:z] 'world' (which has 
>an alternate pronunciation [bi:z]) is not 
>distinguished from SWF res [re:z] 'necessary' 
>(which has no alternate pronunciation) is an 
>error in my opinion. (That is why we had KS16 
>beis.)"
>
>We also had earlier KS with just bes. It wasn't 
>an error then, and I don't believe it is now. In 
>fact it just shows the later merger of the 
>bys/bes-vowel with the res-vowel.

That's not really true. That version of KS 
rejected [bi:z] as an option entirely. That was a 
mistake; it was not inclusive.

>"The fact that SWF bys [bi:z] 'world' (which has 
>an alternate pronunciation [be:z]) is not 
>distinguished from SWF bys [bIz] 'until' (which 
>has a completely different pronunciation) is an 
>error in my opinion. (That is why we had KS16 
>bìs.)"
>
>Again, I disagree. Bys "until" is an unstressed 
>proclitic and has by this definition a short 
>vowel. It's the same as the unstressed y in 
>benyn "woman", or the unstressed a in war "on".

In the SWF one cannot distinguish bys 'until' 
from bys 'world' or war 'beware' from war 'on'.
-- 
Michael Everson * http://www.evertype.com