Re: [gentoo-dev] Re: Agenda [WAS: One-Day Gentoo Council Reminder for June]

[Ciaran McCreesh]

On Thu, 12 Jun 2008 01:40:06 -0700
Brian Harring <ferringb@×××××.com> wrote:
> > Are you seriously suggesting that the portage and pkgcore developers
> > think that they should be able to release a package manager that
> > claims to support an EAPI when it in fact doesn't?
>
> When paludis hit the tree, it claimed to support eapi0.  Did it
> fully?
>
> No, bugs existed.
>
> Via your logic, paludis should've never been in the tree.
>
> See the failing here?  Bugs occur, you're claiming perfection is
> required when your own code hasn't met said standards.

Except that there's no well defined way of testing EAPI 0. There is a
well defined way of testing EAPI 1.

> That's not acceptable in any form.  Actual bug reports, for ebuild
> support bugs turn around (including release) for pkgcore is typically
> within same day.  I give a *damn* about compatibility, even if it
> means enabling paludis to grow (thus providing more power for your
> insepid games).

If you care, why don't you write simple test cases?

> The fact that the -r0 incident occured out of the blue a month or two
> back isn't exactly heartening either- proving it was intentional
> breakage admittedly is not possible.  However considering the
> behaviour displayed here, it's a pretty logical assumption to presume
> the -r0 was an intentional breakage for yet more discrediting BS.

And you accuse us of spreading FUD?

If anyone really wanted to break a package manager, there'd be much
more spectacular ways of doing it...

> You pulled a pretty major no-no here, and the fact you can't admit it
> is pretty fricking sad.

No, *you* 'pulled a pretty major no-no' by refusing to do basic
testing, and the fact that you're trying so hard to make it look like
someone else's fault is pretty fricking sad.

-- 
Ciaran McCreesh