On 2021-04-17 12:08, Christopher Dimech wrote:
Thomas,
So we are decided? I am not pushing anybody down the cliff - rms, you
or anybody. I simply wish that after
a few world wars, people start seeing the light and things will be
somewhat blissed out working on free software.
In a lot of ways this is pretty simple for me. Having observed a few
weeks of the FSF's ham-fisted response to the ham-fisted way in which
they handled a this entire matter, makes it very easy for *me* to arrive
at the stance that the *current state* of the FSF is such that I don't
personally believe it has the ability to be a good steward of the work
that is assigned to it going forward, and so *I* am not going to do that
thing in particular as long as that remains the case.
I'll see my work in GCC11 through (there's one remaining patch review to
address this week); I don't like leaving things in a half assed state if
I can avoid it. The work to finish out C++20 library support features
which passed through the Concurrency and Parallelism study group (SG-1)
in WG21 on their way to being standardized will be, for now, done in a
public repo with GPL license sans-FSF assignment. Other work which I
have initiated to replace the dependency on Thread Building Blocks
within the Parallel STL algorithms (PSTL); something required for this
part of libstdc++ to no longer be marked 'experimental' will not be done
with a GPL license and will not, as a result, be assigned to the FSF.
It is my hope, and expectation, that that work will become part of GCC12
and GCC13 respectively, and I will know in the fullness of time if that
expectation is to be met.
