Issue 1, issues 2 & 3

Jaakko Leino jaakko.leino at Helsinki.FI
Wed May 31 12:04:55 UTC 2000

Michael Everson wrote:
> >Other things being equal (as if they ever were), I'd choose the latin
> >version--if for no other reason, just because the fonts are probably
> >drawn so that "latin" letters go more nicely together that "latin" and
> >"greek" letters mixed freely. Besides, I agree with Erkki that we should
> >use IPA characters whenever there's no reason not to do so.
> But that isn't what you get with GAMMA, for instance. Latin gamma is
> usually drawn like a v with a ring under it; Greek gamma looks like a
> scythe. Is the Latin glyph acceptable?

I must confess that I wrote my comment above without taking a look at
the actual glyphs. My hedge "other things being equal" was meant to
exclude cases like GAMMA. As a general rule, I'd say we want to have
glyphs that look exactly like the ones that have been used in FUPA.
Anything else is a compromise, and we don't want to make compromises
without (very) good reasons. Just following IPA or the vague name
"latin" is not such a good reason.

> >As for the ENG versus ETA question, I still feel (as I did about a year
> >and a half ago when there was a discussion on the issue on this list)
> >that the ideal solution would be to have them both in "free variation",
> >because that would reflect the actual situation in FUPA.
> Then you have chaos for sorting and searching. We want you to avoid this!

I know. Hence the "would" part of my comment. But one way to get around
this is to include yet another glyph in the system, named FUPA SYMBOL
ENG (or LATIN SMALL LETTER FUPA ENG, or something), which could be drawn
either way by the fontographer. After all, what we are talking about is
not the IPA symbol ENG, but the FUPA symbol ENG. I understand that in
IPA, ETA is strictly wrong, but in FUPA it certainly is not. I'd like to
keep it that way.

> The ETA was chosen because printers had Greek sorts in their boxes of lead
> type.

And ENG came in largely because computers had it in their fonts. No
matter what the reason, ETA was chosen at first, not ENG. That's the

> You'll have fewer problems if you let Greek Eta be a vowel and use
> the Latin consonant which looks like the eta, but sounds and functions like
> what you really want, namely the velar nasal.

Whether ETA is a vowel or a consonant in some other system should not
make a whole lot of difference in terms of FUPA. To take a parallel
case, the letter Y is [used to symbolise] a vowel in a lot of languages
(such as Finnish) but a consonant in some others (such as English). And
yet, we all get to use it the way we're used to and everyone's happy. I
still don't see the problems that would follow if ETA was chosen.

But then, as I said in my previous message, this question is not the
most critical one. The main goal is to get FUPA included in the Unicode
system, and I am very thankful that there are active people who work on
this. I apologise for quarreling about secondary questions.

Best regards,

   +----+----+          Jaakko Leino           +----+----+
  /    /|    |\                               /|    |\    \
 +----+ |    | +     Tutkijakoulutettava     + |    | +----+
 |    | +----+ |    Helsingin  yliopiston    | +----+ |    |
 |    |/ \    \|    suomen kielen  laitos    |/    / \|    |
 +----+   +----+                             +----+   +----+
 |\    \ /|    |  jaakko.leino at   |    |\ /    /|
 | +----+ |    |  |    | +----+ |
 + |    | +----+         191 23327           +----+ |    | +
  \|    |/    /                               \    \|    |/
   +----+----+       Onko elämä neliö?         +----+----+

More information about the Ura-list mailing list