GNU bug report logs - #75448
[PATCH] Fix documentation of Iroquoian input methods

Previous Next

Full log

View this message in rfc822 format

From: Kierin Bell <fernseed <at> fernseed.me>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 75448 <at> debbugs.gnu.org
Subject: bug#75448: [PATCH] Fix documentation of Iroquoian input methods
Date: Fri, 10 Jan 2025 13:11:17 -0500

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Kierin Bell <fernseed <at> fernseed.me> Date: Wed, 08 Jan 2025
>> 19:39:12 -0500
>> 
>> This patch fixes a typo in the documentation for Iroquoian input methods
>> for the Oneida endonym, propagated by copy-paste into comments,
>> docstrings, and the NEWS entry. The previous spelling 'Onʌyota:ká:' was
>> missing the third syllable and also used an idiosyncratic orthography.
>> There are two common orthographic variants that I've seen:
>> 'Onʌyote’a·ká·' (common in community language programs) and
>> 'Onyotaˀa·ká·' (used in Karin Michelson's 2002
>> 'Oneida-English/English-Oneida Dictionary'). I opted for the former.
>
> Since these are very delicate matters, I'd like us to go by the
> orthography that is really widely accepted.  In Wikipedia
> (https://en.wikipedia.org/wiki/Oneida_language), I see "onʌjotaʔaːka"
> and "onʌjoteʔaːkaː", which is different from what you show.  I wonder
> what should we use, given so many different variants.
>

Thank you Eli for your careful attention to detail! I'm glad that Stefan
applied the patch as a stop-gap, but I agree that we should go by widely
accepted orthographies. The issue with Oneida in particular is that
there is a relatively high amount of orthographic variation, and there
is definitely no single accepted orthographic variant for this term.

The orthography you cite from Wikipedia is actually a phonetic
transcription that would not be used in prose, although an orthographic
transcription is given also: 'Onʌyotaʔa꞉ka'. This appears highly
idiosyncratic, with no citation given, and is halfway between the more
common orthographies 'Onʌyote’a·ká·' and 'Onyotaˀa·ká·' cited above.

Because there is so much variation, I reached out to a number of
teachers and friends involved in Oneida language revitalization. A
language teacher from Oneida Nation, New York, United States (one of the
three main Oneida communities), who prefers to remain anonymous, has
responded, and I quote from her response here (with permission):


"I understand your concerns about choosing the Ukwehuwehnéha name for
Oneida. It's essential to do what one can to represent our culture and
language, and I appreciate the feedback. However, I believe this name
holds significant meaning and reflects our heritage, where people like
to disagree on specific issues. Believe me, itʼs a battle; no one is
perfect, which is all part of being an u:kwé̲.

It's fascinating how different dialects can influence the spelling and
pronunciation of words, especially in languages like Oneida. The
variations are intriguing, such as Karin Michelson's use of
"Onyotaˀa·ká·" compared to the "Onʌyoteˀa·ká·" that you've encountered.
Here in New York, we aim for consistency in spellings, often writing it
as "Onyotaʼa:ká:" while being mindful of the specific geographical
regions and their diacritical marks. 

It makes sense that there could be regional differences, like those
between Wisconsin and London. Language is dynamic, and these variations
enrich it. It would be beneficial to have more discussions about these
differences to enhance our understanding of the dialects and diacritics
within the Oneida language. Indeed, the variations in dialects depend on
speaker differences and their phonological backgrounds.

I understand your point about the differences in formatting between
Emacs maintainers and Wikipedia. Variations in notation can be
confusing, especially when it comes to unconventional spellings. I
personally prefer consistency and tend to lean towards one style over
another. It's interesting to see how these choices can impact clarity
and readability.

A central aspect of my experience is the online certificate program I
took through the University of British Columbia last year. It has been
helpful in understanding Indigenous languages. One important topic
addressed in the program is avoiding "language mixing" while remaining
within a specific community. It is also crucial to be mindful of the
geographical regions where the language is being taught."


One insight that she provides is that she uses 'Onyotaʼa:ká:', an
orthographic variant of 'Onyotaˀa·ká·'. The latter is common in
dictionaries and academic writing, but it would be pronounced exactly
the same.

She relates another very important point about avoiding language mixing
(preserving dialect and orthographical differences) and aiming for
consistency in orthography within each community.

The orthography used in my patch, 'Onʌyote’a·ká·', appears to be
representative of different dialect variant, with a different
orthography to represent vowel lengthening. But it appears to be the
most common orthography used by language programs from communities in
Wisconsin, NY and Ontario, Canada (going by community resources from
classes that I myself have taken).

So, what I would like to suggest, if we can practically accommodate it,
would be to list both of these two forms whenever one would be used,
e.g.: 'Onʌyote’a·ká·/Onyota’a:ká:'.

The most important place where the endonym occurs from a user-standpoint
is in the initial line of the docstring for the `oneida-postfix' input
method, which is also displayed as completion metadata for
`set-input-method'. I don't think that this would make that line too
long in this case, but that is what had prevented me from suggesting
this type of approach earlier.

Another issue is that there are orthographic variants in relatively
common use for the Mohawk and Onondaga endonyms. These are purely
orthographic and do not represent actual dialect differences as in the
Oneida case, but if it can be done for Oneida, a similar combined
endonym listing in docstrings for these input methods would make sense
for consistency, as well.

>> There is also a small fix for an incorrect input key given in one of
>> the input method docstrings.
>
> I agree that factually the input method behaves as you describe in
> your patch, but is it not more reasonable to change the code to behave
> as the documentation says?  Why should '-' after 'n' produce ñ?  It
> sounds like using '~' for that is better mnemonically?
>

The only issue that I take with using '~' over '-' is that it is quite
unergonomic to enter this repeatedly (at least on a QWERTY keyboard).
However, if the convention used by other input methods is to prioritize
mnemonics in cases like this, then I would be in favor of changing this
in the next version of the patch.

> Thanks.

Thanks!

Kierin

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.