Collections (real collections!) in Turing

[DemonWasp]

Turing seems to be a combination of C (in most regards, including its tremendous inability to do any high-level work) and bits of Pascal. We can thank whatever dieties we believe in that they realised how bad the C++ OO-system was and decided to avoid it. For most of the stuff you see students doing in Turing, you can convert line-for-line into C code.

Turing is slow; at an estimate, it's about 1% of the speed of compiled C++ for most operations, though even that may be generous. It's not the slowest language out there (most Javascript implementations being much slower), but it's also not fast. OpenT would have fixed that (it'd be nearly Java-speed, which is essentially C++-speed).

Python is probably a better first choice; it avoids the nonsense of the machine / OS / hardware but lets the programmer touch on the high-level concepts. It also forces proper indentation, which makes it a lot easier to read code (and therefore, learn how to code).

C++ is, in my opinion, not a terribly great language. It was a good first try, but let's take the lessons-learned from that fiasco and put them to use in our latest set of languages and forget the hulking beast as soon as possible. You might be able to tell that I'm not enjoying the work-in-C++ aspect of my job right now; this is because it won't just BUILD ALREADY.

[Warchamp7]

It's a real shame OpenT died off. It looked promising.