May 09, 2009
On 09.05.2009 00:29, Christopher Wright wrote:
> Open source projects tend to have to support wider ranges of compilers.
> I've seen OSS projects where they prioritized issues specific to
> gcc-3.4.2 as highly as any other issue, even if the developers typically
> used the 4.x branch.

That's probably because they wanted to be cross-platform, since 3.4.2 was the latest GCC supported by minGw for a long time.
May 09, 2009
Lutger wrote:

> - these other languages go out of their way to make new releases backwards
> compatible, sometimes at great costs. They almost never completely succeed
> though.

And D1 makes an effort to follow the spec, and will not incorporate
breaking changes. D2 is, in many ways, a whole new language. Sure,
you could argue that D1 code should be compilable in D2, but it
would matter no more than the argument that C code should be
compilable in D1.

--
 Simen
May 09, 2009
grauzone wrote:

>> Nearly all of those are due to inadvertent reliance on bugs in D1. You see this quite a bit in the C++ world. Every time g++ gets updated, I have to tweak something in my sources.
>
> In the case of D, the problem is that sometimes you don't really know if it's a bug or a feature.

So you file a bug, and if nothing else, the spec should be clarified.

--
 Simen
May 09, 2009
On 5/9/2009 3:38 AM, Steve Teale wrote:
> Walter Bright Wrote:
>
>> Steve Teale wrote:
>>> This is the sort of answer that will kill D. The guy comes back after
>>> 2 years, asks a straight question, and get's told "business as usual,
>>> we're still arguing among ourselves about what it should be".
>>>
>>> Maybe Tiobe is right! Lots of others may not even bother to ask. They
>>> just visit the newsgroup, read a page of it, and conclude "same old,
>>> same old", and go away.
>>>
>>> D should be D, not maybe 1.043, or let's wait a while and see what
>>> happens with D2. Potential real users hate uncertainty. If they are
>>> going to commit, then D must do so too.
>> What bothers me about this sentiment is that every other mainstream
>> language undergoes revision, sometimes major ones, but that never seems
>> to be an excuse for people to not use it.
>>
>> For example, C++ is quite in flux with C++0x.
>>
>> The only languages that are not undergoing revision are dead ones.
>
> Yes Walter, but C++ went through a tedious standardization process - itself a long argument. So there was some basis for people to think that it had 'got there'.
>
> But to come back after 2 years and find the same stuff still going on is depressing - been there done that. OK, it didn't put me off, I'm still in there, but it bothers me. I'd be dishonest if I said otherwise.
>
> I am not criticizing you. I think you are doing a great job under the pressure of a slew of suggestions. But maybe a line in the sand at some point?
>
> OK so for those who crave stability there is D1.x, but when all the focus appears to be on D2, what level of confidence is afforded to D1 users. Can a project Manager cross his heart and say that D1 will still be alive and well in five years time?
>

This is quite interesting! Before there was D1, I can remember the same requests being submitted over and over again. Tool developers complained that they could not use D for anything serious because of its constant state of flux. They continuously requested that a line be drawn. Shortly after that line was drawn in January 2007, a good majority of the community complained that it wasn't exactly what they asked for and a majority of the most active members of the community disappeared into thin air.

The sad thing is, even after D1 became stable following a number of bug fixes, very few people used it for much of anything. Those "advocates of stability" all disappeared; abandoning their tools/projects in the process.

The same thing is beginning again and it seems to me that instead of something that is actually useful, people a waiting for MicroMoney or some other name brand begins to invest in their own D compiler. As far as I see it, there will come a time when D language will soar above all the rest (not only in capability, which it already does, but also commercial usage), I will continue to support it until then and well beyond and I'm quite sure that I'm not the only one. For all the Naysayers out there... Keep saying nay and go the hell away. D2 is just where it is supposed to be. Let’s not end up in the same mess we did by trying to make impatient people happy and releasing D1.

Andrew
May 09, 2009
> beyond and I'm quite sure that I'm not the only one. For all the Naysayers out there... Keep saying nay and go the hell away. D2 is just where it is supposed to be. Let’s not end up in the same mess we did by trying to make impatient people happy and releasing D1.

What about the people who want to use D for something actually useful?
May 09, 2009
Tyro[a.c.edwards] wrote:
> On 5/9/2009 3:38 AM, Steve Teale wrote:
>> Walter Bright Wrote:
>>
>>> Steve Teale wrote:
>>>> This is the sort of answer that will kill D. The guy comes back after
>>>> 2 years, asks a straight question, and get's told "business as usual,
>>>> we're still arguing among ourselves about what it should be".
>>>>
>>>> Maybe Tiobe is right! Lots of others may not even bother to ask. They
>>>> just visit the newsgroup, read a page of it, and conclude "same old,
>>>> same old", and go away.
>>>>
>>>> D should be D, not maybe 1.043, or let's wait a while and see what
>>>> happens with D2. Potential real users hate uncertainty. If they are
>>>> going to commit, then D must do so too.
>>> What bothers me about this sentiment is that every other mainstream
>>> language undergoes revision, sometimes major ones, but that never seems
>>> to be an excuse for people to not use it.
>>>
>>> For example, C++ is quite in flux with C++0x.
>>>
>>> The only languages that are not undergoing revision are dead ones.
>>
>> Yes Walter, but C++ went through a tedious standardization process - itself a long argument. So there was some basis for people to think that it had 'got there'.
>>
>> But to come back after 2 years and find the same stuff still going on is depressing - been there done that. OK, it didn't put me off, I'm still in there, but it bothers me. I'd be dishonest if I said otherwise.
>>
>> I am not criticizing you. I think you are doing a great job under the pressure of a slew of suggestions. But maybe a line in the sand at some point?
>>
>> OK so for those who crave stability there is D1.x, but when all the focus appears to be on D2, what level of confidence is afforded to D1 users. Can a project Manager cross his heart and say that D1 will still be alive and well in five years time?
>>
> 
> This is quite interesting! Before there was D1, I can remember the same requests being submitted over and over again. Tool developers complained that they could not use D for anything serious because of its constant state of flux. They continuously requested that a line be drawn. Shortly after that line was drawn in January 2007, a good majority of the community complained that it wasn't exactly what they asked for and a majority of the most active members of the community disappeared into thin air.
> 
> The sad thing is, even after D1 became stable following a number of bug fixes, very few people used it for much of anything. Those "advocates of stability" all disappeared; abandoning their tools/projects in the process.
> 
> The same thing is beginning again and it seems to me that instead of something that is actually useful, people a waiting for MicroMoney or some other name brand begins to invest in their own D compiler. As far as I see it, there will come a time when D language will soar above all the rest (not only in capability, which it already does, but also commercial usage), I will continue to support it until then and well beyond and I'm quite sure that I'm not the only one. For all the Naysayers out there... Keep saying nay and go the hell away. D2 is just where it is supposed to be. Let’s not end up in the same mess we did by trying to make impatient people happy and releasing D1.
> 
> Andrew

Very well and heartfelt put.

D needs many things, but probably above all it needs the dedication of bright people.

I've had a great time at BoostCon, and enjoyed many interesting conversations with as many bright people. Beman Dawes, the originator of Boost and an extraordinary gentleman, gave me a little history of Boost. Essentially Boost has been toiling in anonymity until Dave Abrahams came about. Dave is a design, coding, and organizational powerhouse all in one. He not only contributed a great deal of design and code to Boost, but also managed organizational and logistics aspects, in addition to exercising the charisma necessary for attracting other extraordinarily qualified people to dedicate their prime to Boost. And it has snowballed ever since. It was amazing just being at Boostcon to see how much expertise punch those people packed.

We are lacking such a powerhouse for D. Walter is a great lone wolf kind of coder, but not a team organizer and charismatic leader. Things have been changed to the better since he has made Phobos accessible on dsource.org and since he released the source of the compiler. Integration with druntime was another great step forward (thanks Sean!) Soon we may convince Walter to put the entire dmd source on dsource.org, which would improve things even more. But the role of "Dave" for D is still up for grabs.

Of course it's not that easy to find that, but in the meantime there's a lot of good things to be done. If you do something great, sooner or later it will be picked up. The perfect example is Don, who became a prominent contributor solely through the sheer quality and quantity of his work. (There are many others that come to mind, but please allow me to stay with one because as soon as I mention more than one, I'd feel I'm unjust to others that I might even forget for the moment.)

So what I think D really needs is your heart and soul :o). Let me (re)try to make a step in that direction by reopening the Phobos czar position. Gregor Richards has held that position for a while but hasn't done anything with it. (I'm not saying this with reproach; Gregor has been busy with school and if I don't know what that means...) I can hardly imagine what someone like Dave could do with that role. So if anyone is interested, speak up!


Andrei
May 09, 2009
On Fri, 08 May 2009 16:20:48 -0700, Walter Bright wrote:

> Frank Benoit wrote:
>> can you allow "shared" or "__gshared" in D1 as do-nothing keywords? That
>> will make it more easy to write code that compiles for D1 and D2.
>> Or is there a trick to accomplish this?
> 
> I think that making code that will compile under both is an impractical task.

I'm thinking that D1 is really the prototype edition of D, and D2 will be the first release edition of the "D Programming Language". By this I mean that D1 has worked out to be a sounding board that has helped define what the design requirements for "D" have really been. The process is not over as usage of D1 is still pointing out things that need to be in D2.

This is a good thing. And if we end up with two languages (D1=D_proto and D2=D_proper) in which both are stable, then that is not such a bad outcome.

-- 
Derek Parnell
Melbourne, Australia
skype: derek.j.parnell
May 09, 2009
On 5/9/2009 11:24 AM, grauzone wrote:
>> beyond and I'm quite sure that I'm not the only one. For all the
>> Naysayers out there... Keep saying nay and go the hell away. D2 is
>> just where it is supposed to be. Let’s not end up in the same mess we
>> did by trying to make impatient people happy and releasing D1.
>
> What about the people who want to use D for something actually useful?

Well, for those cases, there is D1. Which, if I may add, is extremely stable. Take a network administrator running his server or firewall on say DragonFly or any of the many flavors of BSD or Linux; he does not upgrade his server with nightly builds, he sticks instead with a "stable," well tested branch of the OS and experiments with "current." When current becomes stable, the OS developer does not stop developing his OS, he continues on with his vision while the "users" further stress-tests the newly stabilized branch and take steps to find and correct any holes in the system. And our system administrator, well, he doesn't switch until he is comfortable that the newly stabilized branch meets all his criteria for upgrade. If they are not met... he waits until the OS branches again.

We cannot expect the compiler developer to stop developing compilers until after the tools developers decide to build tools. He has done his job, which is to develop a stable, documented compiler. Use it to develop your tools which will in turn allow the compiler/language to gain popularity. Report any bug reports on anything that seems out of the ordinary so that it can either be clarified in documentation or corrected with code.

Walter is a damn good Engineer but I don't think his interests lie in writing tools (IDE and the like). So those people that do write those tools may want to take the "stable" compiler and start developing. So what if this feature "that is in D2" is not in D1? Use what you have in D1 and develop your tools and simply stay abreast with the development of D2 so that when it is time to switch, you are not in the dark.

Andrew
May 09, 2009
grauzone wrote:
>> beyond and I'm quite sure that I'm not the only one. For all the Naysayers out there... Keep saying nay and go the hell away. D2 is just where it is supposed to be. Let’s not end up in the same mess we did by trying to make impatient people happy and releasing D1.
> 
> What about the people who want to use D for something actually useful?

They can if they really want, just as much as others can keep on naysaying if they really want :o).

Andrei
May 09, 2009
Simen Kjaeraas wrote:

> Lutger wrote:
> 
>> - these other languages go out of their way to make new releases
>> backwards
>> compatible, sometimes at great costs. They almost never completely
>> succeed
>> though.
> 
> And D1 makes an effort to follow the spec, and will not incorporate breaking changes. D2 is, in many ways, a whole new language. Sure, you could argue that D1 code should be compilable in D2, but it would matter no more than the argument that C code should be compilable in D1.
> 
> --
>   Simen

I'm certainly not making that argument. In the talk Andrei and Walter gave at the D conference they mentioned the design goal 'no issue left behind' This is what makes D evolution so much more interesting.

Yet it has a price. Looking at C# for comparison, it is about the same age (C is not fair, too old): you can easily upgrade from 2.0 to 3.5 and beyond. Most important part of that ease is the ability to use older libraries in a newer environment. Still, lots of issues in C# are already beyond fixing.


1 2 3 4 5 6 7 8 9 10 11 12
Top | Discussion index | About this forum | D home