D Language Still Showing Promise, Advancements

[Zorael]

The D devs are very aware of the common points of criticism (in short: dynamic/shared libs, stop-the-world GC), and limiting the amount of allocations in phobos is AFAIU on the todo list.

There was a concurrent gc library for D1 (CDGC) but it hasn't been ported to D2. It looks like it will be, though.

Anyway, the CDGC author's dconf talk: Concurrent Garbage Collection for D by Leandro Lucarella

[plonoma]

You mean the ARC that got added in C++11 in the form of Smart Pointers, Shared Pointers and Widgets? :P

Yeah, something like this kinda. ARC is neat because it pushes things to compile time.
Read up on the C++11 new pointers and have found some problems with it: http://www.codeproject.com/Articles/...Smart-Pointers
Seems like we need some better semantics than C++ can provide, and the earlier attempts is now starting to create legacy stuff.
The behaviour is still not really consistent and you need to take special precautions to not fail.
It's coming closer but it's still not as reliable and error-resilient as I would like it to be.

[c117152]

First off, D isn't a C substitute. It's a C++ substitute. If you're in the market for a C substitute, try golang.

Now, while I detest big OO languages like C++ and Java, D at least does it right. In fact, If you entertain the basic premise that large OO language are some times necessary - say, for closed source game engines where there must be some very good abstractions around some very tough math, D is C++ done right.
D takes every feature C++ has to offer, and redoes it in retrospect properly. The syntax is coherent and consistent. The standard library isn't the random grab bag that C++ is but has some very specific goals and functions. And most importantly D was designed from the grounds up with a vision in mind unlike the randomness C++ has become.

So, if you're a C++ programmer that is working on a pet project, you might as well go with D. You won't regret it. If you're part of some closed source library project that has the leeway to experiment a little, D just might be right for you.
Even if you're a hobbyist that dislikes C++, you can do well to learn some D (like myself) so you can fine tune your hate of the concept rather than the implementation Then you can go ahead and fall in love with golang :P

[GreatEmerald]

Even if you're a hobbyist that dislikes C++, you can do well to learn some D (like myself)

Indeed. Though I like both the concept and implementation. Writing in D is just a lot faster and easier than writing in C (and to a slightly lesser extent than C++), and just like C++ and unlike Java it doesn't force you to use OOP. So in a way it's a good C replacement as well, although C isn't as direct a competitor as C++ is. Also, like Walter Bright said in the keynote, its nice and coherent syntax makes you feel good about writing code in D, because the resulting code is just aesthetic-looking and is very readable, if you use the proper language features (templates, array slicing, contracts, Ddoc, unit tests etc.).

[timofonic]

Phoronix: D Language Still Showing Promise, Advancements

provably correct, and of industrial quality. By provably correct they mean it provides provable memory safety, provable purity and immutability,
http://www.phoronix.com/vr.php?view=MTM5MTg

At first I toght about being "probably", not "provably". Then I found this word exists and as wiktionary puts, it comes from provable -> -> proven -> prove -> proof

I'm not a native speaker, so this amuses me and I always find words I don't know to learn more about the English language

[a user]

So, if you're a C++ programmer that is working on a pet project, you might as well go with D. You won't regret it. If you're part of some closed source library project that has the leeway to experiment a little, D just might be right for you.
Even if you're a hobbyist that dislikes C++, you can do well to learn some D (like myself) so you can fine tune your hate of the concept rather than the implementation Then you can go ahead and fall in love with golang :P

The GC is in every case you mention for me (us) the main point why D is no go. we have a quite constraint memory demand which makes any currently known attempt of GC a deal breaker.

i can't go into details, but in contrary to what somebody else posted here, GC is worse than manual memory management. of course you do not compare a bad implementation of memory management with a good GC. you compare a good memory management with a good GC. the lack of control yields a damn lot of situations where a memory limit will cause big issues with any kind of GC.

GC are in general only for usecase where you actually have either a lot of memory or time enough.... or your customers/clients can live with the drawbacks.

[Ericg]

The GC is in every case you mention for me (us) the main point why D is no go. we have a quite constraint memory demand which makes any currently known attempt of GC a deal breaker.

i can't go into details, but in contrary to what somebody else posted here, GC is worse than manual memory management. of course you do not compare a bad implementation of memory management with a good GC. you compare a good memory management with a good GC. the lack of control yields a damn lot of situations where a memory limit will cause big issues with any kind of GC.

GC are in general only for usecase where you actually have either a lot of memory or time enough.... or your customers/clients can live with the drawbacks.

Automated Reference Counting is a nice middle ground. As long as the object has 1 reference to it, it wont be freed which prevents 'accidental frees' that can happen from human-error in manual memory management, but when it hits 0 references it gets automatically freed.

[mrugiero]

Just as a heads-up, all the conference videos can also be found on YouTube here:

http://www.youtube.com/channel/UCzYzlIaxNosNLAueoQaQYXw/videos

Nope. GC can actually be more efficient than manual memory management. And those who absolutely must go around it, it's easy to either turn it off or simply avoid using functions that invoke it. See this presentation for more information on the subject:
http://www.youtube.com/watch?v=1MF5bcmvJ0o

Is there any written version? I don't really like long videos, I prefer reading. I'm curious about how could it be more efficient than good manual management.

[GreatEmerald]

Is there any written version? I don't really like long videos, I prefer reading. I'm curious about how could it be more efficient than good manual management.

There are slides, but they don't show the point the presenters want to make that well, so you should just watch the video. It's all about the GC in any case.

[ciplogic]

Automated Reference Counting is a nice middle ground. As long as the object has 1 reference to it, it wont be freed which prevents 'accidental frees' that can happen from human-error in manual memory management, but when it hits 0 references it gets automatically freed.

Yes and no: ARC increases the size of the object by adding the reference data into it. I'm (almost) sure that ARC may work great for this guy, as I suppose that the bias against the GC is founded on rumor and "I think", and "these big pauses". D has not a generational GC (because in generationa GCs the freeing of most short lived objects automatically).

Also, ARC is expensive in CPU power, if you have a loop, and in this loop you have a reference increasing and decreasing by one, the same loop can work slower with reference counting than with a proper GC.

At last, but not at least, ARC gives more predictability about freeing of objects (given you don't make reference cycles, meaning leaks), so many pauses are by design very small. This I think is the main advantage of ARC, and to not use a GC. If the user's application creates many long lived objects, the GC pauses can be really big.

But memory constraints? Not sure.