On Sun, Feb 11, 2001 at 05:03:12PM +0000, Simon Cozens wrote:
> On Sun, Feb 11, 2001 at 11:28:45AM -0500, brian d foy wrote:
> > okay. i quit.
>
> Well, hm. I'd rather we actually made something positive out of this.
>
> There's obvious FUD out there and we don't seem to be giving the impression of
> getting much done, or doing anything to counter it. Part of the problem is
> that we don't currently have anything that we can point to and call progress.
> That's a problem in itself, because if people don't see progress they lose
> interest and go away.
>
> In order to do something about this, I suggest that we should:
>
> i) maintain a weekly summary of what's going on on the mailing lists.
> I'm happy to do this when I do the p5p summary; it could be hosted on
> www.perl.com or www.perl.org, I suppose. I'll post the summaries to
> perl6-meta and people can do what they will with them.
>
> ii) maintain a white paper style document on dev.perl.org detailing what
> we've decided, what we've considered, our rationales and so on.
> Roughly, a distilled summary of *all* of the mailing list traffic, ever.
> I can make a start on that tomorrow. That way people have something to
> look at and see where we're at.
yeah that's a start, but I think you hit it right on the head - until we have
something concrete from Larry, we won't have 'progress', as far as the world
sees it.
Again, We'll have continued discussion, but what the perl development project
needs right now is a swift kick of *direction* from larry. And I'm pretty sure
that he knows this.
Ed
(
ps - and as far as I'm concerned, there are two great decision points that
could be made early on && made public for perl's design:
(1) that perl is a 'micro-architecture' where the perl runtime
environment pretty much dynaloads everything as it needs it (builtin
functions && data-structures). This keeps the kernel as scalable as
possible (both up and down)
(2) that the perl parser is pluggable - so that one can easily write
different parsers via a common mechanism, something like
Parse::RecDescent. In fact, I have high hopes that it is a descendent
of Parse::RecDescent.
Everything above this is window dressing. I've seen sentiments of the above
points being made && lying around places, but I've never seen a place where
Larry has blessed them and published the fact that he has blessed them.
And larry is so tied to perl that whoever is 'PR guy' is really not going to
matter - the world is ultimately going to listen to - primarily - *him*. And
judge our progress (at least in the design stage) based on what he says or
doesn't say.
)
