I don't think speakers of the language that invented Worcester Sauce are in a position to give prescriptive logical advice to other languages.
Languages are the products of a sort of evolutionary processes. Rarely do the conventions appear to be the best possible solution at first sight. But it isn't uncommon for convincing arguments to appear later, showing that some seemingly arbitrary rules are almost beautifully constructed to make languages efficient. Not only do common words tend to be shorter than uncommon ones; I've also seen examples of word pairs that would be dangerous to confuse to be further apart in spelling and pronunciation than statistics would suggest (I believe sailing was the example I saw: single words can mean life or death, and are often spoken under difficult circumstances).
> I don't think speakers of the language that invented Worcester Sauce are in a position to give prescriptive logical advice to other languages.
Of course they are. They didn't invent it, and even if they did they could still be right on _this_ argument.
>Not only do common words tend to be shorter than uncommon ones; I've also seen examples of word pairs that would be dangerous to confuse to be further apart in spelling and pronunciation than statistics would suggest
And there's a whole bunch of counter-examples for that. One is the german "zwei" (the number 2) and "drei" (3). The german military commonly uses "zwo" for 2 because those two are so easily confused.
It's not like our languages are willfully poorly designed, but there's still a bunch of bad decisions in there, and we might be in a position to fix it, especially when it comes to things like spelling (which is mostly decreed from on high, since what schools teach matters most).
