On 12/13/05, Brett Sanger <[EMAIL PROTECTED]> wrote:
> On Tue, Dec 13, 2005 at 01:50:58PM -0500, Rob Kinyon wrote:
> > This is an internals change, not a new framework. This change is
> > completely and utterly backwards compatible.
> >
> > I'm not sure why people are complaining so much.
>
> If I had to guess, I think it's because of one or two reasons:
>
> 1) THe change would render no benefit, but woudl create new ways for a
> enwbie to mess up.

If you don't use any plugins, you'll get the current CA functionality.
If you do want to use plugins, you can override any single
functionality that CA uses. By itself.

> 2) The change would mean that someone could make plugin Fubar that
> relied on a different set of assumptions for the pipeline, and the odds
> that two plugins would not be compatible would increase.

Anytime you have pieces of software written by people who have never
communicated with each other attempting to work in the same space, you
have issues. For example, CDBI and Test::MockObject as seen in
http://www.perlmonks.org/?node_id=516372. That's not intrinsic to or
exacerbated by this proposal.

Personally, I think it's because I didn't do any legwork before
throwing this completely random idea out there. I know it'll work
because I've been involved in the internals before. I know it'll help
implement certain things because it's a cleaner design. But, I haven't
made a concrete proposal, which is why people are having issues. It's
all too handwavy and metameta. And, unfortunately, I don't have time
to do more than that right now.

Rob

---------------------------------------------------------------------
Web Archive:  http://www.mail-archive.com/[email protected]/
              http://marc.theaimsgroup.com/?l=cgiapp&r=1&w=2
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to