Salut a omnihom! I'm 3 years too late for this group, and over a hundred years too late for Idiom Neutral, but what the hell. I'm an auxlang enthusiast, and I've come to opine that Idiom Neutral is structurally the best candidate, at least from a European perspective. I've been reading Louis Couturat's Histoire de la Langue Universelle recently, and as a service to anyone who's interested (especially those who either can't or won't read the book) I'm posting here a summary of his Idiom Neutral material. The first part is his criticisms of the original 1902 version of the language, and the second part is his record of the numerous reform ideas that were presented between then and 1907.
It seems to me that Idiom Neutral embodies a journey that began in the late 1880s with Schleyer's Volapük and ended before the First World War when the bulk of the Neutralists jumped ship to Latin Sine Inflexion and thence Occidental. Despite the existence of definite 1902 and 1912 versions, Idiom Neutral was always a language in flux, with the early manifestations resembling Volapük and the late manifestations resembling Occidental/Interlingua. The "true" version must presumably lie somewhere in between, but anyone who holds up the 1902 manifestation as this true version must face the fact that even before the ink was dry, the then Neutralists had lost faith in it and were scrambling to redress its defects. I tentatively suggest that the true form of Idiom Neutral may be actualised via a careful, non-radical reform of the 1902 version, one that corrects the genuine flaws without pushing the language too far backwards (Volapük) or forwards (Occidental).
I enumerate Couturat's criticims of 1902 Idiom Neutral as follows:
(1a) Lots of roots are visually mangled, when compared to their natlang originals.
(1b) There are bloated consonant clusters in roots and compounds, which as well as being hard for some people to pronounce, will often require the insertion of schwas, which different speakers place differently.
(2) There are no definite or indefinite articles.
(3a) There is no part-of-speech marking for root nouns, adjectives, and verbs.
(3b) There is no direct derivation between these parts of speech. He remarks that "-asion" could be replaced with the noun form of the verb root, if such a form existed.
(3c) The possessive adjectives are not derived from the personal pronouns.
(3d) When a noun is used as a verb, there's no way of inferring what the verb means exactly.
(4) There is extensive ambiguity, apparent and actual, caused by the choice of roots and suffixes.
(5a) Some derivatives are mangled/ugly, and in places there are new roots instead of derivatives in order to avoid this ugliness. These surplus roots mean there's more to learn.
(5b) The "-ik/-al" conflict could be solved by means of a "neutral" suffix.
(6a) The prefixes "ne-" and "no-" are too similar, and are inconsistently used.
(6b) The prefix "re-" has the meanings of repeating and redoing; these meanings should be distinguished.
(7) There are compounds that are confusing because of having been directly copied from natlangs.
(8) Ambiguity between subject and object is possible in relative clauses due to there being no accusative marking.
At the end, he opines that the language tries too hard to be naturalistic/romance, and that the more naturalistic an auxlang becomes, the more complex and irregular it becomes, and thus the harder to learn, especially for people unfamiliar with romance languages.