On Monday, 12 October 2015 at 07:44:47 UTC, Andrei Alexandrescu
wrote:
Instead of assuming my purpose here is to pull tricks on you
and manipulate the dialog politically, it's more productive to
just stick to the technical discussion. I only started the
dialog to get more informed about technical things.
I'm not assuming this. I'm fairly confident that you have the
best intention for D in mind rather than winning internet debate
points. So do I.
Still you are pulling the above mentioned trick. i don't think
you did it on purpose. I think you did it because you are
convinced that DIP25 is great, in fact so convinced that you take
thing for granted that aren't.
It is a natural tendency we all have to lower the standard of
evidence we require when we are already convinced. You, me,
everybody.
I don't think grepping for return will have a good noise to
signal
ratio.
Sorry, I meant to git grep for "return ref".
So I did. I found 2 uses of return ref in current master that
aren't in unitests. These are both related to unicode decoding.
These are legit use of DIP25, and I recognize their value.
Still I don't think this is valuable enough to have its own
syntax. I also, on the other hand presented cases of things that
people thought could be done with DIP25, but ended up to be
either not possible or to have a disappointingly low complexity
to added value ratio.
I can agree on something: DIP25 is the right direction. But it is
too little to pay for itself and I haven't seen anything here
that would suggest otherwize.
You also mentioned several time that you have good evidence
that
DIP25 rox. Yet, every time you post that without any evidence,
I'm a bit
more convinced that none exists.
The motivation is in the document and follows many discussions
derived from it. It's all about functions and particularly
member functions returning up references to data safely.
Amaury, you and Timon are probably the most competent PL
theorists in this forum. He did great work: found real problems
with DIP25 that need to be looked at. In that light, spending
time protesting and yelling figuratively at people is a
distinctly unproductive way to spend your time as a very
talented contributor. Do great work. It will be recognized.
Don't point me at your past posts. They are not great work and
you know it. Don't point me at those related DIPs. They are not
great work and you know it. Don't find reasons to not do great
work because it'll be wasted on my ego. Do great work and you
will prevail.
This tone of discussion has carried its course. I'm done
arguing so if you want to continue arguing, great - last word
is yours. In the recent times I've done my best to reduce my
participation to unproductive discussions in forums, and the
added perspective and time for real work have been very
valuable. I suggest everyone to try it.
Ok, I'll write a DIP.
