Andre Poenitz <[EMAIL PROTECTED]> writes: > PS: As a bold but reasonably educated guess: We could get rid of 90% of > the code in support/* _and_ reduce compile and link times by a third (on > top of that what we gained lately _and_ get a decent socket _and_ get > a decent concurrency framework if that mindset would undergo a reality > check...
And we could spare even more code if we worked on kword instead. JMarc