[gentoo-dev] Re: The fallacies of GLEP55

[Steven J Long]

Ciaran McCreesh wrote:

> On Sat, 16 May 2009 00:28:36 +0530
> Arun Raghavan <ford_prefect@g.o> wrote:
>> As I've stated a long time ago, I'm for this solution. My
>> understanding is that there are 2 objections to this:
>
> 3) It doesn't solve the problem. It doesn't allow things like version
> format extensions.
>
> That's the big one, not the other two.
>
Debian-style epochs, which were mooted to this list over a year ago, do
however, given that we already have SLOT.

They're also a lot simpler and do not 'require' a potentially unlimited
set of new extensions for every "new format" (look, shiny!) that a herd
might experiment with.

And again, you haven't explained why an internal format specification
should allow NN variants on the format (beyond your usual sniping at
the level of developer expertise, which you propose to address by making
it easy to mess up the spec, instead of simply expecting people to learn;
which they certainly appear capable of doing, as the ebuild tree attests.)

In summary, the proposed benefit doesn't seem like one, and certainly not
enough to justify the fundamental noob mistake of breaking encapsulation.

Personally I remain unconvinced this is even enough to merit the use of
epochs. I'd much rather see them kept in reserve for a real issue, and
an innovation which actually solves a problem our end-users face. So no,
not a "big one" at all; just yet another student attempt at coercion
cloaked in obfuscation. No more, no less.

'Sometimes I find it hard to believe that people can attach so much of
their ego to one particular design, even when the obvious flaws are
pointed out. As a wise man once said to me:
"There also comes a point where you realise that no one can know
everything, so it's not a problem to ask someone or on occasion be
wrong.."'[1]

Think about it. No? Ah well never mind, didn't really expect you to ever
accept you might still have something to learn. No doubt there'll be
another 50 or so emails from you next time I catch up on the list. And
of course if each one doesn't have a detailed objection, it's enough for
you to claim that you've run it past the list. And if they do, it's
simply because the person doesn't understand (despite your clear and
lucid explanations to the list.) </sarcasm>

[project]
You are aware that many Gentoo developers simply cba to argue with you?
Since the stuff you're proposing often makes no effective difference to
what they're doing anyway (it's /that/ useful), it can seem easier simply
to let you have your way. Believe it or not, that's how I feel; I'm only
speaking now as your GLEPs are so massively flawed, that I've been told
I would have to maintain a patched portage were they to go ahead.

Happily the only patches required would be to get rid of the crap you're
proposing. I *still* resent the extra workload, and the fact that it's
only necessitated because you haven't got over the rejection of being the
only developer ever to be kicked in such a fashion. Twice.
[/project]

[1] http://igliot.blogspot.com/
--
#friendly-coders -- We're friendly but we're not /that/ friendly ;-)