Date: Wed, 11 Apr 2018 16:39:26 +0000
Yep. And speaking as someone with 250M lines of legacy, I'm well aware of
the need to simplify and the need to leave things behind. See, for
instance, http://wg21.link/p0684 - but it has to be a balance.
On Wed, Apr 11, 2018 at 12:37 PM Boris Kolpackov <boris_at_[hidden]>
wrote:
> Titus Winters <titus_at_[hidden]> writes:
>
> > Yeah, one of the recognized strengths of C++ is that it doesn't leave all
> > of the legacy code behind, generally speaking.
>
> But there is a cost to this legacy support both in time it takes
> the committee to figure out the battalions of special cases as well
> as the resulting complexity which some of us have no capacity to
> handle and which will most likely limit the "toolability" of the
> end result.
>
> Boris
> _______________________________________________
> Tooling mailing list
> Tooling_at_[hidden]
> http://www.open-std.org/mailman/listinfo/tooling
>
the need to simplify and the need to leave things behind. See, for
instance, http://wg21.link/p0684 - but it has to be a balance.
On Wed, Apr 11, 2018 at 12:37 PM Boris Kolpackov <boris_at_[hidden]>
wrote:
> Titus Winters <titus_at_[hidden]> writes:
>
> > Yeah, one of the recognized strengths of C++ is that it doesn't leave all
> > of the legacy code behind, generally speaking.
>
> But there is a cost to this legacy support both in time it takes
> the committee to figure out the battalions of special cases as well
> as the resulting complexity which some of us have no capacity to
> handle and which will most likely limit the "toolability" of the
> end result.
>
> Boris
> _______________________________________________
> Tooling mailing list
> Tooling_at_[hidden]
> http://www.open-std.org/mailman/listinfo/tooling
>
Received on 2018-04-11 18:39:42