Todd Moody wrote:

> Indeed, and the history of auxlangs shows the compromises.
> Esperanto's daft hats were a compromise between phonetic
> spelling, with no digraphs, and recognizability.  Ido's
> accusative rules are another compromise.

I do not see any compromise between Esperanto
and Ido. New auxlangs are made exactly the
same way as artlangs -- there is no planning
before a new auxlang is made, so the maker
of the new auxlang does not know whether his
new auxlang will be better than other auxlangs.
For example, Larry Sulky does not know why he
made Ilomi, why he abandoned it, and why he is
now working on Lume.

Languages are tools, so they should be designed
in a rational process, like all the other tools.
For example, when a Honda engineer is about to
design a new car, his first step is a review of
new materials and new technologies, and an estimate
how they can be incorporated into the new design.
Before he designs the new car, he has rough idea
how much better the new car will be, so he knows
that he is designing better car than the old

If the Honda engineer had the mindset of an auxlanger,
he would say that he cannot design a better car
because he does not know if he should design a
truck, or a passenger car, he does not know how
powerful the engine should be, and he has absolutely
no idea what body color is the best... 

If an auxlanger had the mindset of the Honda engineer,
he would design the auxlang like Adolf Hitler's
Volkswagen car, or like Mikhail Kalashnikov's AK-47
rifle -- something that is simple and can be used by
ordinary people.

>You know, W. John Weilgart claimed that aUI could be "learned" in
>minutes, because that's all it takes to memorize the basic elements.
>And he claimed that the meanings of words could be worked out without
>much memorization.  And I'll give him this much credit: he did the work
>of creating an "encyclopedia" of aUI, and translated some literary works
>into it.  Did anyone ever learn aUI well enough to communicate orally in
>real time?  I tend to doubt it.

We know that aUI compound words are extremely
vague and that its long and short vowels sound
alike. These are fundamental flaws.

>Is Ygyde really easy to learn and understand?

I like its extremely simple grammar, but I
have concerns about the ease to memorize its
predefined words. Its 90 adjectives, 90 nouns,
and 90 prepositions are mutually related to
make them easier to remember (mnemonic roots)
but I am not yet happy with the result. I
thought of importing English roots, but they
do not fit into the two letter roots. At best,
the English roots could be used as alternative,
probably four-letter long roots (Long Ygyde).

>Is it expressivly adequate?  (After all, a language could
>be very easy to learn and understand precisely because it
>*isn't* expressively adequate) You claim to have a scientific
>approach to these things.  If so, then you know that the
>question can't be answered until there are speakers of Ygyde.
>To claim that a language with zero speakers is both easy and
>expressively adequate is simply to talk rubbish.

Do you have good ideas?

The main purpose of any auxlang is to convey
simple ideas efficiently.