Intro - Chapter: 0:History 1:Design 2:Basics 3:Var 4:Op 5:IO 6:Block 7:Sub 8:OOP 9:Rx 10:Meta
Overview - Appendix: A:Index B:Tables C:Cook D:Delta E:Best of F:Links G:Glossary

Larry Wall: Perl 5 was my rewrite of Perl. I want Perl 6 to be the community's rewrite of Perl and of the community.

Why Perl 6 ?

Back in the year 2000, things didn't look so good. The buzz around Perl was gone, because everybody had decent Regexes, even if theirs were not as tightly integrated as in Perl. And many of the other Perl 5 features could be also found in Python or Ruby. Naturally, also, new shiny things appeared in mainstream languages, that couldn't so easily be pasted into Perl. The Perl syntax, especially for Regexes, is very rich, therefore you could hardly add something in a meaningful manner without changing a handful of others. Some features like "do {} while"-loops were just bolted on and other feature could not even be plugged in. A complete revamp was also more and more necessary because things like the OOP and missing signatures made problems to the Perl beginners. Other things like formats just had to move out of the interpreter into a module because they weren't that important as in the early days. And finally a lot of cruft had also to go, that was deprecated since Perl 5.0 like the ol' package'sub syntax.

Much of these issues could be dealt within the perl 5 arena, and is recently done by the way, but if you would look into the perl internals, you would see a speed racer, highly tuned but with a lot of dirty tricks your cs-prof would't tell. In that magic world, were many things are mingled together merely invisibly, it is really hard, to bring in some cascades of changes. And even that would have some natural limits. Some nice features like lispish macros, which are really powerful and a way cleaner and faster in execution than perl 5 sourcefilters and a nice XS replacement, which would largely ease Perl's extensibility, are impossible to achieve with that codebase. It would be a complete rewrite anyway.

But it was not only the language and his interpreter, also perl people themself needed some new toy. Something to brim over with enthusiasm for. Perl 5 just worked, which is good, but sometimes boring if there i not much you expect to happen. Especially if there are problem like described above. And Larry had the strong feeling that the Perl community had to open up anyway.

So it had to be a complete new language with a brand new interpreter and Perl would be again the most cuddlesome thing around. As the Perl 1 manpage tells you: Perl was always about to have all the cool features of other languages in one place. That isn't for the purpose of bling-bling. (You couldn't brag with lazy evaluated arrays in a hip hop video anyway) But it's about getting the task done the way you prefer, so only the very best could be enough.

How it all started

The initiative to develop Perl 6 in a process run by the Perl community was announced at the fourth Perl Conference on July 19th, 2000.

Parrot

* Perl 6
* Parrot
* Ponie
* PCT
* NQP
* Rakudo
* Pugs
* elf
* SMOP
* mildew/mold
* mp6
* kp6
* v6
* MAD
* STD
* sprixel
* gimme5
* niecza

Even the brave and higly skilled pumping nicholas the first wrote in the p5p list that in the regex engins were dragons he was afraid to wake up.

Intro - Chapter: 0:History 1:Design 2:Basics 3:Var 4:Op 5:IO 6:Block 7:Sub 8:OOP 9:Rx 10:Meta
Overview - Appendix: A:Index B:Tables C:Cook D:Delta E:Best of F:Links G:Glossary