On Wed, 2011-02-02 at 18:47 +0000, "Jóhann B. Guðmundsson" wrote:
>> Have the installation issues been cleaned out?
> I don't know. as Bruno said, the intention is for the
testing to be
> possible (almost) entirely without installation, and that should be the
Of course those that are using USB Flash drive can create one with
persistent storage to get a closer to actual install behaviour but we
loose any early Gnome Shell adopters which might shed light on some
usage bugs that the test case don't cover which you only find by an
Sure. There's nothing we can do about that as the desktop team / QA
team, though. Really all you're saying is 'I wish the anaconda team
would fix the installer faster', it has nothing in particular to do with
this test day or with GNOME in general.
I'm pretty sure alot of bugs filed against Gnome components will
related to UI Designs decisions rather than actual code breakage by
I have added several warnings to the Wiki page to try and avoid this,
take a look.
for instance somehow the UI Designers forgot all about those
that close the lid and leave the computer on for download or are forced
to close the lid in class when the teacher or professors ask them too
and then need to get immediately back on to type something down thus
cant wait for suspend/resume time/breakage which is a *Design feature*
as Richard mention on his blog "In GNOME 3.0, we’re defaulting to
suspending the computer when the user shuts the lid, and not providing
any preferences combobox to change this. This is what the UI designers
for GNOME 3.0 want " I for example would file a bug against this.
Please start a new thread about this, I'd rather not derail the test day
discussion. (FWIW, I agree with you to a degree: I do not want to
suspend my laptop when it's plugged into the mains and I close the
screen. I just do it to keep things tidy on my desk, and to turn off the
panel to save a bit of power and not uselessly run the panel and reduce
its lifespan and potentially cause screen burn. You will be able to
configure this from dconf, though, I'm guessing. I'd say it'd be
perfectly legitimate for people to file a bug on this, actually. I don't
think the design has been well thought through, if it applies when on AC
as well as when on battery power.)
Another example of a potential bug or a *Design feature* is when you
to system settings --> User Account and you create a new user and are
forced to choose the login name for the user from a list of user names
based on the full name you enter for the user or is this an actual bug
that you cannot type whatever username you like for the user? Bug vs
Design battle. .. .
As I advise on the page, the best thing to do is ask in IRC before
filing a bug; I'm hoping there will be a few GNOME devs around all day
to help answer such questions. (If you're a Fedora / GNOME dev on this
list, please do try and find some time to hang out in the channel to
help with this, it'd be really useful!)
Anyway back on track we might want to downgrade smolt since it's
working due to bug 672971 that is if smolt profile is needed ( which I
expect to )
I've fixed this (applied jlaska's fix and pushed a fixed build through
to Rawhide), so it should make the test day image.
another thing was mentioned on our QA IRC from a reporter
was a lack of IRC client in current Gnome Live but I think the reporter
did not realize that you can do irc with empathy.
Reporters have already begun jump starting the test day
and reported some early birds bugs.
Indeed, which is great, but I hope they all come back and test again
when the full test case set is up!
I've written two new test cases so far:
and am working on the others. If anyone wants to help out, please do: my
scratch list of test cases to work on is http://fpaste.org/N8si/
please just find me on IRC to co-ordinate and make sure we don't
duplicate effort, and the test cases are clean and added correctly to
the test day page. thanks!
We also need to have any updates to current packages present as
tomorrow and not just be on specifically composed test day LiveCD to
prevent/reduce any duplication and reports of bugs that may already be
fixed it's always better that reporters are testing as close to the same
yup, that's below getting the live image right on my priority list,
though - I anticipate most people will test using the live image, it's
usually the case, especially before we have an Alpha out. I'll try and
have instructions on the page for installing all necessary updates that
don't make it into tonight's Rawhide compose.
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org