Hi Subhashish,
This is a really hard question, and I don't think there's a good, specific
general answer. A non-specific general answer is that you have to find
something that works reasonably well, takes advantage of any pre-existing
resources, is as easy and practical to use as possible, and that people are
willing to use. That almost sounds like a description of the problem, more
than an answer.
IPA might not actually be a good answer, depending on the variety of
dialects and the phonology of the language in question. For example, some
speakers of Hawaiian pronounce w more or less like English w, others
pronounce it like English v. Do you change the writing based on who's
talking? Another example is English t—for many speakers, the t in "top" is
different than the t in "stop" and both are different from the t in
"bottle"... but in the mind of an English speaker, they are all t. IPA can
be *too* detailed.
You can also lose relationships between words if you spell only according
to sound (again, depending on the language). English "photograph" and
"photography" might be spelled more phonetically as "fotəgræf" and
"fətagrəfi", but that decreases the obviousness of the relation between
them. Is it worth it? It's a judgement call.
As for using the script of a/the dominant language of the region where the
language is found, it depends on a lot of linguistic and cultural factors.
Is the relationship between the language groups neutral or positive? Are
people already generally literate in the writing system of the dominant
language? Are the languages fairly closely related? Yes to any of those
makes the writing system of the dominant language a better choice—though it
still may not be a *good* choice if the answer to any of them is no.
Are there closely related languages (linguistically, not necessarily
geographically) that already have a well-designed writing system that could
be borrowed and adapted? For example, a lot of the Turkic languages have
adopted fairly similar versions of the Latin alphabet. That way some
problems only have to be solved once, and it can also be easier to read a
closely related language.
Another technological issue—are input devices (keyboards—either physical
or virtual, like on a phone) that cover all the needed symbols readily
available? (That's another reason not to use IPA—I think it's actually not
too hard to learn the subset relevant to a language you speak, but it is
often *really* hard to type.)
Ideally, a writing system should be devised over some time, with heavy
input from speakers of the language and guidance from linguists who have
experience with related languages, the development (historical or
practical) of writing systems, or both if available.
A gateway wiki might be a good place to experiment with a new writing
system, but it could also end up generating too much inertia if good
changes are proposed that require re-writing almost everything. I really
don't know.
It's an interesting and difficult question you have. I hope it generates
some fruitful discussion.
—Trey
PS: Also read up on spelling reform in various languages for related ideas
and possible problems.
Trey Jones
Sr. Software Engineer, Search Platform
Wikimedia Foundation
UTC-4 / EDT
On Thu, Aug 22, 2019 at 3:26 PM Subhashish Panigrahi <psubhashish(a)gmail.com>
wrote:
Dear Wikimedians,
Some of you might be recovering from the Wikimania fatigue. Those of you
who have already recovered, I wanted to pick your brain about something
that came up multiple times during discussions but none really seem to have
a clear answer.
Which script (writing system) an oral language speaker would use for
creating an entry on (gateway [1]) projects like Wiktionary or Wikibooks or
even uploading a list of words on Commons using a tool like Lingua Libre?
Will it be the script used for the official language of the region where
the former language is from?[2] This is a bit controversial as native
speakers of many indigenous languages would see this as a form of
colonization. Will it be the w:International Phonetic Alphabet (IPA)? This
is probably the least controversial but a common and average user might not
be able to read IPA as the latter was created by linguists and was created
for linguistic and scholarly studies rather than for everyday use.
Wikimedians who are native speakers of languages with less
written/recorded documentation and individuals who work on such languages
are more encouraged to share their inputs based on past experience.
1. Gateway project: This is a made-up term to define the Wikimedia
projects that are more welcoming to newbies and do not require stringent
citation as almost all oral languages would lack that. It was fascinating
to see Amir challenging that it only takes about 30 seconds to add an entry
to Wiktionary (
https://commons.wikimedia.org/wiki/File:Amir_Aharoni_demonstrating_how_to_a…
)
Subhashish
_______________________________________________
Languages mailing list
Languages(a)lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/languages