some smaller issues in english rules

[moritz-gross]

for fun, I asked Claude to find any mistakes in Rules/Languages/en.

That's a solid yield I think. In particular the "ln" and copy-paste errors. At least I think "ln" should be without a space, as "gcd" is too for example.

[NSoiffer]

It's a good idea and Claude found some spacing inconsistencies, but as the comment at the top of the file says "for readability, spaces can be used around any of the delimiter characters", so they don't matter. Still it is good to be consistent. It would probably be a good idea to turn the words describing the format into a BNF-like format (probably as an addition rather than a replacement). Maybe that would help AI. Could you do that?

The "l n" fix is wrong -- the letters want to be pronounced separately.

Claude did find bugs in the phrases, but they don't affect speech, just attempts at translations. It is good to fix those.

I will fix up "l n" diff and commit.

[moritz-gross]

Yeah that's what I figured, but doesn't the same apply to "gcd" then, or what am I missing?

I'll look into the format idea.

[NSoiffer]

That's a good point about "gcd" and "ln". My memory is that at least one speech engine said something like "lynn", but I just retried the four that I know are used in JAWS and NVDA and they all spelled the letters out. So either they have improved in three years or my memory is faulty... or both :-)