>> Somehow they ended up with about half of the language being case sensitive and half being case insensitive.
Why you think it's because of the "lack of forethought"? Case-insensitive methods/classes were decided to be so for a reason (since it made easier to write bigger projects in PHP). You may not like this decision, but it doesn't mean everything you disagree with is because nobody thought about it.
>> Somehow they ended up not considering the implications of "case insensitive" with regards to internationalization, and thus ended up with utterly bizarre rules where code works or doesn't work depending on the locale it's run in.
Again, the only reason why that bizzare situation with Turkish locale exists is because it wasn't noticed. There's no deep and mysterious reason rooting in flawed PHP design why it can't be fixed - as far as I can see, the fix is not hard either, it just that nobody got to fixing it.
2. Changing language implementation with potential of breaking of 99% of existing code is not a "bugfix". If you don't understand this, maybe it's too early for you to lecture people on how to design languages. It can be done, but there's no reason to - it would lead to massive breakage with no upside in any functionality.
3. "The issue itself is not that important" - BINGO! You said it. If the issue is not that important, why you're wasting time dwelling on it and make it sound as if this is the most important thing ever? You wrote a long post supposedly about deep flaws in PHP design and process - but the only thing you actually wrote about is an obscure bug that you yourself admit is not important! Maybe it's time to discuss something that is important then?
1. Ad hoc language design process (if you can call it that at all) resulted in a serious lack of forethought in the language's case sensitivity rules. Somehow they ended up with about half of the language being case sensitive and half being case insensitive. Somehow they ended up not considering the implications of "case insensitive" with regards to internationalization, and thus ended up with utterly bizarre rules where code works or doesn't work depending on the locale it's run in.
2. Extreme emphasis on backwards compatibility means an unwillingness to fix the bug by e.g. making the language fully case sensitive, or regularizing the rules for case-insensitive comparisons among language identifiers to be locale-independent. This could be phased in over a long period of time with progressive deprecation to ensure that anyone relying on the old behavior has years of notice to fix it.
3. Somehow, the broken case insensitivity of these identifiers also breaks case identical lookups for certain names. Despite the bug existing and being discussed for ten years, this still doesn't work. No matter how bizarre your case-insensitive comparison rules are, there is absolutely no reason whatsoever that looking up a class using the exact same byte-for-byte name should fail. That this hasn't been fixed means that either the PHP team doesn't care, has no idea how to fix this, or the code base is too broken and crufty to make a fix practical. None of these alternatives says anything good.
The issue itself is not that important (unless you're Turkish, or running on Turkish servers, or somehow end up running in a Turkish locale), but the fact that it exists and has not been fixed in a decade means something is seriously wrong.