std.experimental Timeline

[Dicebot via Digitalmars-d]

On Saturday, 2 January 2016 at 04:27:02 UTC, Jack Stouffer wrote:
> On Friday, 1 January 2016 at 20:10:22 UTC, Dicebot wrote:
>> Thus the original formal process was looking like this in my 
>> head:
>> ...
> Can this be adopted formally?

Depends on if you want to volunteer to manage the process and 
communicate with all interested parties to make it reality - and 
keep maintaining process in future. Or convince someone else to 
do the same.

>> Idea of any "automatic" inclusion is very harmful because by 
>> current standard std.experimental modules are not required to 
>> have Phobos quality and can be very far from being good enough.
>
> My proposal was based on the quality of std.logger and 
> std.allocator when they were merged. I wasn't aware that 
> modules that were unfinished could be merged into 
> std.experimental.

Haven't found any issues with std.allocator so far but std.logger 
definitely is not Phobos ready per my requirements. I have been 
recently re-evaluating it as possible replacement for old Tango 
logger we use and found that in several places it forces 
unnecessary GC allocations and/or immutability (i.e. 
https://github.com/D-Programming-Language/phobos/blob/master/std/experimental/logger/core.d#L742 requires that log message must be formatted into immutable string before it can be passed to another logger). This is a deal breaker that can possibly require major API change to fix.

>> What makes things most complicated is that I don't do review 
>> bookkeeping anymore and no one else (including Andrei himself) 
>> have been willing to take control of the process since than.
>
> This can be supplemented by people asking for review managers 
> like they do for initial inclusion into std.experimental, no?

As far as I understand, there is no semi-persistent review 
manager at all right now. Some people volunteered to handle 
specific voting threads but to make sure new formal process is 
accepted one needs much more than that - provide formal spec, 
ensure approval of language authors, keep poking all involved 
developers to actually follow the process etc. It doesn't take 
that much time but it must be dedicated long-term commitment to 
work.