Has D failed? ( unpopular opinion but I think yes )

Guillaume Piolat first.last at gmail.com
Sun Apr 14 10:17:17 UTC 2019


On Sunday, 14 April 2019 at 09:42:01 UTC, Laeeth Isharc wrote:
> Brutal upfront sometimes, but the upfront emotional experience 
> is the least important part of the cost of developing and 
> maintaining software.

100% agree.

That's exactly my experience, the upfront payment of creating D 
software was sometimes brutal the first year - even for a 
seasoned D programmer - and afterwards it yielded dividends in 
lower Cost Of Ownership quite evidently.

And a lot of it is thanks to DUB/SemVer I must say, without it I 
wouldn't even dare maintaining that much projects else, and 
another big part is thanks to the community that comes with a 
particular culture. It really changes your perspective over time.

It's very important to keep the complexity down (not only code) 
in an organization and D is an integral part of this **because it 
scales from disposable code to production code**.

My _solo_ organization has 62 projects in 4 years, all in D 
(dub.json):
- 2 are open-source to eventually keep costs down, if other 
companies have similar views eventually,
- about 20 of which have to be maintained privately with  (Cost 
of Ownership),
- 27 are only maintained lazily when needed, they are absolutely 
needed for experimentation
- about 15 were thrown away and won't ever have to be maintained

If the rate of maintained-LOC _production_ was higher than 
that(it's hard not to produce code), it would lead to a worse 
outcome in the exponential curve of software size.

The future of D is perhaps more shared costs between D users, 
hiring reputed library designers, with a shared understanding 
that value must flow back to the D Foundation.

Only this can solve software size.



More information about the Digitalmars-d mailing list