Hallo conlangers!

On 02/01/2018 20:03, Dirk Elzinga wrote:

> Any writing system is fundamentally nothing more than a set of mnemonics.
> Deciding what you need to be reminded of will tell you what the system
> should look like. That’s just as true for the scratched characters on the
> Shang dynasty oracle bones as it is for the most hyperspecific IPA
> transcription you can produce.

Yep.

> That said, I work in IPA on my projects, and here’s why.
>
> * I take the sound design of my languages seriously. By that I mean not
> just a carefully crafted inventory, but rather the overall sonic effect of
> spoken language. The IPA is a flexible, standardized tool that helps me do
> this.

Right. I also often use IPA during the design of a phonology, but tend
to develop a suitable romanization when it is done because I find such
romanizations more convenient. They can be designed to be easy to type,
and I feel that IPA is somewhat overfeatured for transcribing a single
language (and sometimes, one still has to resort to diacritics, e.g.
with the not-so-rare aspirated stops, where you can easily find more
convenient solutions in a custom-made romanization). If you ask me, IPA
is what you use to specify sounds, but not always the right thing to
transcribe your language.

Most of my conlangs are "lostlangs"; they exist in a version of the
consensus reality with a few languages added - tiny linguistic
minorities or ancient written languages representing old lineages that
in reality died out completely. Old Albic has a writing system of its
own, but is intrafictionally transliterated by scholars, based on the
orthography of one of its daughter languages, Avalonian (the Old Albic
phoneme inventory is a proper subset of the Avalonian one; the Avalonian
orthography in turn is inspired by Insular Celtic conventions such as
_c_ for /k/ in all positions, and h-digraphs for spirants, e.g. _ch_ for
/x/). Most others of these lostlangs would have been romanized by local
schoolmasters, parsons, doctors or the like (most of these languages are
located in Western or Central Europe), and it is these transcriptions
that I use to transcribe them. Intrafictionally, these transcriptions
are mostly from the age of the typewriter, and designed to be easy to type.

> * I am familiar with the IPA and I’ve been trained in its use, so using the
> IPA is convenient for me. (I have a keyboard layout that I’ve been using
> since grad school, and it has become second nature to me. I can type pretty
> quickly in IPA now.)

Typing IPA is actually not very difficult: http://www.i2speak.com/

But it is still a degree of inconvenience, and not really helpful in
writing HTML pages, where you need those &#x....; things instead because
the "real" characters break too easily! And you can always build a
romanization where everything you need is readily accessible on the
keyboard you are using!

> * I rarely, if ever, design writing systems. It’s just not my thing. I do
> design romanizations but only for convenience, and only after I have a
> fairly good understanding of the sound system. So while I can appreciate
> the aesthetics of a carefully designed romanization, that’s not how I want
> to spend my time. Using the IPA lets me do what I want to do without
> getting distracted by other things.

I sometimes do design writing systems, such as the Old Albic one:
http://www.joerg-rhiemeier.de/Conlang/albicfixed.png

And I feel that once I have a good easy-to-type phonemic transcription,
I can work on the morphology, syntax and lexicon of my conlang at least
as well as if I was transcribing everything in IPA!

--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html
"Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1