On 11/11/15 23:34, Walter Bright wrote:
On 11/11/2015 3:03 AM, Lionello Lunesu wrote:
Will share how it's received in an hour ;)

This is great! How did it go?

Actually went much better than expected! Very engaging crowd, with good questions. I think the way I positioned D was being "just another cool tool in a programmer's toolbox" really resonated with them. It certainly prevented the usual Go vs D vs Rust kind of competitiveness.

Few of the questions I got were

* Why doesn't D explicitly specify the exceptions that can be thrown? (To which I answered that I never saw the point in Java and only found it tedious. This did not convince the person.)

* Can I use D to build for ARM? (I said that the usual ARM code generators of GDC and LDC could be used, and that the runtime had less and less [if any] x86 dependencies left.)

* Why the difference between byte[] and char[]? (I explained it as being a semantic difference, with no difference in memory layout. One can be indexed meaningfully, the other can't.)

* Can I create an array of shared ints (as opposed to a shared array of ints)? I said that one could by using shared(int)[], although I have yet to try this myself.

* Does D use SIMD? (Yes, and it's leveraged automatically when you do operations on arrays.)

* How do the libraries available to D compare to the others like Go and Python? (I replied by introducing C/C++ compat and code.dlang.org)

* What if a global function with argument A has the same name as a member of A? (I talked about function hijacking and how D prevents it.)

I might remember more tomorrow. Will share if I do.


Reply via email to