Re: [firstname.lastname@example.org: Re: Woody retrospective and Sarge introspective]
- To: email@example.com
- Subject: Re: [firstname.lastname@example.org: Re: Woody retrospective and Sarge introspective]
- From: Raphael Hertzog <email@example.com>
- Date: Thu, 1 Aug 2002 12:02:41 +0200
- Message-id: <[🔎] 20020801100241.GB26704@home.ouaza.com>
- Mail-followup-to: firstname.lastname@example.org
- In-reply-to: <20020801040054.GB23775@azure.humbug.org.au>
- References: <20020730235824.GK13609@home.ouaza.com> <20020731000353.GA13359@riva.ucam.org> <20020731085202.GD30556@home.ouaza.com> <20020731105636.GC729@sirena.org.uk> <20020731120602.GA1916@home.ouaza.com> <20020731133048.GB32589@azure.humbug.org.au> <20020731140656.GF1916@home.ouaza.com> <20020731145045.GE32589@azure.humbug.org.au> <20020731154651.GK1916@home.ouaza.com> <20020801040054.GB23775@azure.humbug.org.au>
Le Thu, Aug 01, 2002 at 02:00:54PM +1000, Anthony Towns écrivait:
> Because otherwise all the packages that depend on the old lib and haven't
> been updated are uninstallable, and testing is unreleasable.
That's only true if you remove the old library. No ?
You could respond by explaining the "testing specification", for example :
"testing has been designed to have packages that are all built with the
same version of the library, we do that by imposing only one version of
a single source package in it".
Is that true ?
Do we have any precise "testing specification" ? or do we only have the
sources of the testing scripts ?
> Sorry, but you're just making things up as you go, now.
Oh, sorry to try to understand the motivations of current choices ...
You skimmed a question that interested me in particular :
would it be (easily) feasible to run the testing scripts on both
t-p-u and unstable at the same time ?
> Yes, and that's the problem: you don't understand the problem well enough
> to know all the basic pitfalls, and when you find some you try to patch
I'm sorry, but if you keep doing so, you and the ftpmasters will always
be the only one knowledgable enough to do anything wrt to release
management. Sure, I don't know all the intricacies but not many know
them all ...
I'm trying to help, no more. If you're not interested by help,
suggestions and so on... just tell me. I'll go do something else
and complain when something doesn't satisfy me instead of
trying to improve the situation.
> over your solution rather than working out what's fundamental.
And what is fundamental in your opinion ?
I suppose, that you did some "auto-criticism" of testing yourself
already. What was the result ? What do you think can be improved ?
If you tell me, I might redirect my researches in a direction that
satisfies you more ...
Raphaël Hertzog -+- http://strasbourg.linuxfr.org/~raphael/
Formation Linux et logiciel libre : http://www.logidee.com