[dmd-beta] getting ready for beta - but we have lots of regressions

[Nick Sabalausky]

On Mon, 14 Oct 2013 05:38:02 +0200
Martin Nowak <code at dawg.eu> wrote:

> On 10/03/2013 06:47 AM, Brad Roberts wrote:
> > I really want the various package makefiles to have a target that 
> > constructs a directory of the results of the build that are
> > suitable for directly being bundled up into a release. 
> That sounds like the way to go, how feasible is it to do this before 
> this release?
> Especially considering that there are some dreaded dependencies on 
> external binary
> not in the git repo.

Or we could use something that doesn't involve keeping per-platform duplicates in-sync, has a workaround for the external dependencies and already exists. *cough* But then what would I know about it? ;)

This makefile stuff was already discussed over four months ago and AFAIK hasn't gone much of anywhere since. If any of it ever actually materializes then even I'll grant that maybe it might actually be worth using. But vaporware and discussions that chase after ideal approaches are useless.

This has become an irritating pattern with D. These releases, assertPred, @nogc/heap, and no doubt plenty of other stuff I haven't even been paying attention to: First shit gets discussed. Nothing happens. More discussions occur. Again, they go nowhere. Someone finally stands up, decides "fuck this shit" and actually writes REAL CODE. Aaaand...the new pull request languishes in debate hell because people would rather keep chasing rainbows then let D ever actually progress.

Forget letting "perfect" be the enemy of the "good". We let "perfect"
be the enemy of accomplishing anything at all.