Lack of update information

[Matthew Woehlke]

Jesse Keating wrote:
> If everybody is just throwing the latest from upstream in whatever the
> feel like it, what keeps a release from being more stable than rawhide?

Aside from upstream (hopefully) not being a mess, there is a difference 
between "this seems to work" and "what justification, besides that 
upstream has fixed bugs and added features, can I give to push this update"?

If a package seems to be non-broken and fixes bugs, that's good enough 
for me to update. (Of course, I'm also not in a limited-bandwidth 
situation, but...) Just because I haven't been bitten by a particular 
bug /yet/ doesn't make updating useless; maybe it saves me from being 
bit tomorrow.

Also, notice I didn't actually object to asking for better "what's 
changed" information. I objected to putting bureaucracy in the way of 
what Fedora currently is; a distro that tends to closely track upstream.

> On Mon, 2009-01-26 at 17:29 -0600, Matthew Woehlke wrote:
>> Should we not release any updates without a Fedora bug being filed 
>> asking to upgrade to the latest upstream?
> 
> That's actually not unreasonable.  The update process should be user
> driven, as in a user needs or wants something specific from the new
> upstream code, we don't just install a bot to throw whatever falls out
> of upstream directly at our users whether they want/need it or not.

...except now I have to run around opening a ticket every time KDE bumps 
its requirement on CMake version, or libical version, or...

No, thanks.

-- 
Matthew
Please do not quote my e-mail address unobfuscated in message bodies.
-- 
find / -user your -name base -print0 | xargs -0 chown us:cats -- Unknown