Hacker News Reader

Comment by hugh4life | Hacker News Reader

I really wish more effort was put into D... I think that's really the best path to freedom vs Java and .Net... I wish Nokia would would adopt D because nobody wants to use C++ and Python is slow...

I just saw this on twitter...

http://twitter.com/casperbang/status/27134952930

"Java has an official open source reference implementation, but is not free. C# has only unofficial open source implementations, but is free!"

http://www2.research.att.com/~bs/bs_faq.html#Java

"Java isn't platform independent; it is a platform. Like Windows, it is a proprietary commercial platform. That is, you can write programs for Windows/Intel or Java/JVM, and in each case you are writing code for a platform owned by a single corporation and tweaked for the commercial benefit of that corporation."

View on HN

Which D, 1 or 2? I have used D for several years and know a few people that also used D for a longer period of time. Consensus is that D is dead for all practical purposes (especially D2).

Could you expand D being practically dead? or link to some sources on that? I haven't heard such strong statements about D before, and I'm curious to learn more.

Pretty much sums it up: http://h3.gd/devlog/?p=22 D1 and D2 are like different languages. D1 went to a dead end and D2 got nowhere, it got into a conundrum feedback loop. It's a bit hard to explain everything in this post since there is a lot to cover, but most of the community that is years into this is aware of all of it.

Also, good luck with any kind of support that is not in the interest/hobby of D developer. I once wrote a fairly large program that was for a commercial purpose (wrote as an order) - and I got stuck on a fairly large bug in standard library which I couldn't resolve myself because I have lacked the expertise to do so, I even think it was not open sourced then also (basically free() never returned memory to the OS). That bug is still open after several years. I offered money, anything basically to get this resolved since my application crashed due to it - if there weren't for helpful Tango guys I would have to rewrite the damn thing from scratch in another language.

D1/Tango had so much potential, but it got nowhere, unfortunately. I am still bitter I spent so much time and effort in D at all, just to return to C (and some C++).

"Which D, 1 or 2?"

D2 of course... "Consensus is that D is dead for all practical purposes (especially D2)." BS..

There is a chicken and egg situation with D... if somebody big doesn't put some backing behind it then it will go nowhere...

There is no chicken and egg situation with D anymore. It has been around for 10 years. D1 at least. D1 and D2 are different beasts.

Tomasz wrote it better than I can right now http://h3.gd/devlog/?p=22 and I respect his view because, if nothing else, he is the man that wrote the largest code base there is written in D (for his Nucleus engine and tools). Main problem, among MANY, is that people who are working on D these days are not coding in D at all, not serious projects at least - otherwise they would see fairly quickly what is wrong with it. This topic has been covered among D community by large now, you can dig up fallout of tango team from D2 and numerous sagas alike on digital mars newsgroup.

Most people will see it as C++ with a somewhat nicer syntax. That does not provide a lot of incentive to switch to D. You need more than marginal benefits to cause a huge shift.

Additionally, it doesn't help that D is torn. There used to be the Phobos/Tango split, now there is yet another split: D1/D2. The second is definitely a more serious competitor to C++, but there is no up-to-date fully open source compiler available. Digital Mars's D2 compiler has a proprietary backend, gdc (a GCC frontend) is outdated, and ldc (LLVM frontend) does not have usable D2 support.

Oh, and if you want a 64-bit D2 compiler, even proprietary, you are out in the woods (yes, work has started).

Even the average Scheme compiler is in a more usable state ;).

How a language with a proprietary reference implementation and library, namely D, could ever be a suitable solution to the Java trap? This is utter nonsense.

How a

Reply on news.ycombinator.com