On Sun, 2007-02-25 at 13:26 -0800, chromatic wrote:
> On Sunday 25 February 2007 12:56, Geoffrey Broadwell wrote:
> > As I mentioned in another thread, but didn't make clear in that email: I
> > don't need a "finished" spec.  I need the *current* version of spec to
> > actually be mostly implemented.
> The implementors want the same thing.
> > And if it's not fun, I won't end up putting any of my very 
> > limited free time into it.
> Neither will the implementors.

My apologies if what I said came across as critical; it wasn't intended
that way.  I was merely trying to point out that there is at least one
user here that does not believe lack of a "finished" spec is the biggest
blocker for people wanting to switch to Perl 6.  I don't believe that
the core team should be rushing the spec at all.  That's just pushing
any possible issues from before first release to after first release,
and any team manager can tell you that's a recipe for making broken
things considerably more difficult/expensive to fix.

In the early days, the extreme flux of the spec was a bit of an issue,
because being away for a couple weeks meant that things had changed A
LOT in the intervening time.  That doesn't seem true any more.  Even for
a user with limited time, the current pace of changes don't seem
daunting to keep up with.  Rather, the problem is that many things have
yet to work at all.

I'm not trying to say that the implementors should rush either, nor am I
complaining about current status; I grok the dynamics of volunteer code.
I merely disagree with the "spec is all-important" crowd.  I personally
have a preference for "rough consensus and working code", and I wanted
to make sure that viewpoint wasn't lost.


Reply via email to