Something similar happened to me. I was using it as a backup strategy and then one day, I didn't have a backup.

The STL does one thing very well: it's very predictable performance-wise while being reasonably useable.
When you use it, you know perfectly what the performance is going to be.
It also offers some occasionally-useful features (std::weak_ptr for instance). c++11's move constructors and rvalue references are very good for squeezing out the last bit of performance where possible and for ensuring exception safety to certain operations; although it's mostly useful for very low-level, entrenched libraries such as the STL. Lambdas are syntactic sugar, but a well flavored one.

c++ is now a very different beast than it was in the 90s. If used properly, it can be very effective (performance-wise) in complex projects. But it can also give programmers tons of rope to hang themselves with.

In my testing, yes. I just generalized his patch so that it could be used when starting any thread rather than just the specific thread that he found.
jojelino finds some interesting stuff but his implementation often leaves something to be desired.

When bash takes that long to parse a .bashrc it's usually because you have the bash_completion package installed. That takes a long time under
Cygwin. It would take a long time under msys too.

And, of course, the "translation system" is just cygwin + hacks.

You're misinterpreting the email. An equivalent patch was added. It just wasn't done in the manner proposed by the original author.

It should be no surprise that most of the comments in this thread are rehashes of 12+ years of history of "Why don't they just do ...?"

We'll happily accept a patch which implements fork using low-level NT semantics since all of our efforts in finding the right sequence of calls to make this work have proved to be for naught.

We "stopped at 95%" because we were supporting Window 9x. Once we stopped doing that we could make use of the features available in more sophisticated versions of Windows.

But, of course, there are some things that are just very hard to do in Windows no matter what version. That's why some things are not implemented.

The new 1.7.1 version of Cygwin allows multiple versions of Cygwin to coexist.

This wouldn't really make sense for many users, such as Free/Open Source Software.

On the other hand, it would make sense to tie the value to the cost of LICENSING some intellectual work.

So those who have high-value intellectual works would have to overvalue it, in order to avoid being compelled to license it to unwanted parties, and thus pay.
On the other hand, if someone wanted to license their work for free, they'd have to pay nothing and still be protected.

Here's the link to the cygwin donations page.