Will D continu to live after walter death?

Joakim dlang at joakim.fea.st
Mon Oct 16 10:17:21 UTC 2017


On Monday, 16 October 2017 at 09:03:56 UTC, Rion wrote:
> On Thursday, 12 October 2017 at 15:04:57 UTC, moechofe wrote:
>> What is the wanted lifetime of the project?
>> Is D will manage to pass through time?
>> It is valuable to start a 40 years old project using D?
>
> Its a serious topic but that gets way too much joking.
>
> While D is part of the foundation, that is not the issue. It is 
> leadership, focus, goals, long term vision ... There are not 
> that many people that can take over that task successfullly.
>
> I noticed its all a joke to people. Posted in a other topic 
> about the fragmented nature of D and the high focus on solo 
> developers. And the issue of what happens if a main developer 
> of a extension has no more time or god forbids dies. That same 
> applies to D as a language.
>
> That D is in a Foundation means nothing. Apple without Jobs is 
> still Apple but you can see the difference in there products 
> after his dead.

You're right that it's a valid question for the project, as the 
main D frontend is largely developed by Walter:

https://github.com/dlang/dmd/graphs/contributors

Kenji did a lot for 5 years, adding more lines of code than even 
Walter during that time, but appears to have bowed out since 
early last year.

Walter really should be mentoring dmd contributors, and actively 
looking for more. Druntime and phobos, on the other hand, seem to 
be developed by others, and don't depend as much on one person.

To answer the original question, it is almost impossible to plan 
for a 40 year-old project, given how fast tech changes.  You have 
to be prepared to maintain ancient toolchains yourself for such a 
long time horizon, like I imagine COBOL devs do today.

In that case, one of the main criteria should be that the entire 
toolchain is open-source and fairly understandable, because you 
will almost certainly have to maintain it yourself.  I don't 
think you can depend on even mainstream languages like C, C++, or 
Swift being around and having good support in 40 years.


More information about the Digitalmars-d mailing list