[Dlang-internal] dmd semver experiment

[Johan Engelen]

On Tuesday, 15 May 2018 at 15:17:04 UTC, Walter Bright wrote:
> On 5/13/2018 12:27 PM, Johan Engelen wrote:
>> I feel currently the project is racing at too high speed 
>> pumping out releases, at the cost of quality and carefulness 
>> that I feel is appropriate for compiler/language development 
>> (adding to existing quality-of-implementation problems).
>> Thus I am very much in favor of major releases every half year 
>> (or longer), with only minor releases in-between that do not 
>> add new features or language changes.
>
> There's always a tension between people needing things now and 
> being conservative.

I think the "need" of people is exaggerated, and I feel is often 
argued from purely ones own point of view.
Shouldn't a professional language lean towards "conservative"?
(if "longer than 2 month between major releases" can be called 
"conservative" for a programming language...)

> The best I can do is if you post a list of the bugzilla 
> regressions most irksome to you, we can prioritize them.

Thanks but my point was: that's no use. All those bugs are 
discovered after release, and by the time they are fixed there 
are 2+ new versions out with their own new set of bugs and 
regressions. Most often workarounds have to be added for the 
bugs, only in a few cases is it practical/in-time to fix the 
compiler (feasible because Weka already has its own fork of LDC 
[1]).

The 'best' that can be done from my perspective is: do a major 
release every 6 months. What's the rush? What good have the fast 
cycles brought? Let's provide time for second thought, 
contemplating, restraint, and other things that surely would help 
a mature programming language. Please, currently the taste that 
sticks is that D is just a small group's hobby language...

- Johan

[1] Because a "fix" is needed that introduces a bug elsewhere, so 
can't upstream yet. The _minimized_ testcase is 100kbyte full of 
templates. We are still figuring out how to correctly fix things.