Beefy Boxes and Bandwidth Generously Provided by pair Networks Russ
Don't ask to ask, just ask
 
PerlMonks  

Re^15: The current state of Perl 6

by TimToady (Parson)
on Apr 22, 2010 at 01:06 UTC ( #836170=note: print w/ replies, xml ) Need Help??


in reply to Re^14: The current state of Perl 6
in thread The current state of Perl6

This seems to deserve a serious answer.

The original question was about the "Freezing the Spec" and not accusing chromatic of something bad. Chromatic is a great guy!
I agree about chromatic. As for Freezing the Spec...
But how many projects have you worked on where the spec evolves towards infinity without a stop? What happens to such projects? Isn't it wise to avoid a second system affect which Perl 6 seems to be experiencing.
I've been on many different kinds of projects with many kinds of development cycles. I've never seen a spec evolve toward infinity without stop. I have seen specs that were frozen too soon, which is why Perl has always used "slushes" rather than "freezes". The phrase "evolve toward infinity without stop" implies that such a process must be divergent, but you can also evolve toward infinity without stop while also asymptotically approaching a stable value. There are many graphs in analytical geometry, and the typical development curve has many S-curves with stable plateaus where we release intermediate versions. Perl 5 has followed this course, so it's "evolving toward infinity without stop" too. The question is really one of convergence vs divergence, something we think we understand with Perl 6.

As for second system syndrome, that usually bites you when you are working under a deadline. We don't do deadlines; we do convergence.

You can't make systems without making mistakes, and if you avoid making mistakes you will never build a big system!
If you actually look at most of the recent design decisions for Perl 6, you'll discover that most of them are actually to correct earlier mistakes that make it harder to implement. If any new powerful features sneak into the spec, it's generally a result of simplification, not complexification. So in analyzing the current design process, please consider convergence vs divergence, as we do. We want this as much or more than you do, or we wouldn't be working on it so hard.


Comment on Re^15: The current state of Perl 6

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://836170]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others taking refuge in the Monastery: (8)
As of 2014-04-19 13:43 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    April first is:







    Results (480 votes), past polls