James Antill wrote:
> Are you really arguing that you never make mistakes?
No, that's not at all what I'm saying! I'm arguing that problems of the
"works on Fedora n, doesn't work on Fedora m" type are extremely rare and
that it's usually safe to assume that testing on one version of Fedora is
sufficient. It doesn't depend on the packager. I just took my own updates as
an example because I definitely know whether they're broken or not as I get
all the complaints if they are.
> Or to put it another way, I assume you've read Paul Wouters's excellent
> post in this thread ... so are you arguing that he's just stupid? Or
> maybe that he didn't care?
His update wasn't of the "works on Fedora n, doesn't work on Fedora m" type,
so this has nothing to do with the particular message you were replying to.
He screwed up. It can happen. We're all human. Life goes on. It seems
another update has already been issued to rectify this. So why is this
causing that big a stir?
Now we can look into figuring out what exactly characterized this update as
"dangerous" so we can make sure future updates of the same type can get
closer scrutiny. (Apparently one characteristic was touching config files,
which seems to be a flag to me, config files by definition vary from system
to system.) But banning all direct stable pushes surely isn't the answer.
Kevin Kofler
--
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/devel
