Sorry about the delay--hectic weekend that looks likely to go into overtime.

On 6/7/14, Scott Raney <[log in to unmask]> wrote:
> STEVE!  Good to have you back.  Don't know if you've seen my
> proposals, but I'm a fan of Inlis and especially of the
> analysis/rationale/motivations behind it.  Do you have a more complete
> description of the current status of the language than what I found on
>   Regards,
>     Scott

I've managed to render all my documentation obsolete--and sometimes
I'm not sure myself where I'm at. But I think I've had my last bout of
Brilliant Ideas That Change Everything, and I'm repairing the damage.
I'm also conducting some translation tests. (For example, it turns out
that there's not much difference between translating from Basic
English and regular English into Inlis. I was mildly surprised, but BE
is structurally too different from Inlis for its limited vocabulary to
help.) If I can unify my word lists and update the other documents,
I'll be about at a point to start a group and annoy the masses.

Originally, I envisioned a simple task--just relexing Interglossa to
English (or more nearly Bislama). But I kept tweaking, and my first
word list, based on the Interglossa list, has given way to one based
on VOA's Special English, which unfortunately has more synonyms than I
like. I've changed the orthography, phonology, and importation rules
somewhat, and the grammar rather more, but it should still achieve its
design goal of making Paul Bartlett hack up hairballs.

> PS: In response to your comment, what I'm thinking of is something
> like (which I'm really liking!).  The source
> sentence is presented in several languages, along with several
> candidate auxlang translations of it (different spellings, vocabulary,
> different modifier words, etc.) each with a description of how it is
> parsed (maybe colorize by parts of speech? Is there a standard for
> this?).  People (hopefully with different native languages) read the
> parses, see if the corresponding auxlang translation then makes sense
> to them, and the rate each candidate (not sure of the criteria yet,
> but intelligibility will of course be the big one).  After doing that
> for awhile (bring on the Mechanical Turk!) we'll start to get some
> ideas of which things work and which things don't.  The ones that
> don't will be dropped, and variations on the ones that do will be
> introduced to see if people like one of them better.

It's not a bad idea, especially for worldlangs, where it could ensure
neutrality. The problem will be getting enough people from diverse
enough backgrounds to make it work. They would also need enough
background in languages (not necessarily linguistics proper) to
understand the questions. The IALA tried something vaguely similar
after deciding to create their own auxlang, mailing a questionnaire
with several proposed auxlangs (ranging from something like Esperanto
or Ido through something roughly like Novial all the way to something
like Interlingua. They didn't get much response, but this was well
before the Internet.

I'm still not sure it would work. On the one hand, if you present
mock-ups of actual systems (the IALA approach), respondents will
probably vote for individual features they may not like because the
overall system seems good. But if you poll on features rather than
systems--phonology, orthography, word order, basic grammatical
features, etc.--you may get conflicting results and will likely lose
respondents because of the lengthy development period. Still, it's
worth a try.