Lets try to make a test plan that fits all the desktops we ship ( KDE LXDE XFCE etc.. )

That is not the scope of this ticket. This ticket is just for implementing a test plan for the F13 criteria, which intentionally cover only GNOME (or, more correctly, the desktop spin and the default DVD installation package set, both of which happen at present to be GNOME).

I'm working on this, I have about half the tests done so far. Category is https://fedoraproject.org/wiki/Category:Desktop_Acceptance_Test_Cases .

Can we call the QA:Testcase_<DE>Desktop<test> like "QA:Testcase Gnome Desktop audio basic" we can then easily reuse the test cases that get written and just adjust it to *DE and thus have all Desktop test cases in the same category ( Desktop_Acceptance_Test_Cases ) ?

No, because I don't want to do it that way. The semantic difference I mentioned above is actually an important one: we're writing a test plan for the default desktop, not for GNOME specifically. It's just that right at present the desktop filling the role of 'default desktop' happens to be GNOME. If you look at the test cases, I have written them as far as possible to be generic - they are not in fact specific to GNOME, and are intended to be written such that they would still apply if the default desktop were something different.

Gnome is what we currently ship as the default we just point testers to pick all Gnome testcase or whatever DE if it changes in the future? I did take a look at your test cases and they did contain Gnome specific direction hence they are in fact specific to Gnome otherwize we only would need to make one set of Testcases and point all DE Testcases to those right... If they would not have I would not have recommended we mentioned the DE in question in the Testcase. Should we then change the category to Default_Desktop_Acceptance_Test_Cases then create a separate category for each DE ( Gnome Desktop KDE Desktop etc ) Have to sets of direction one for default and another one Gnome and where to we put Xorg and Window manager etc that we ship? I failed to se the logic in what you are doing so that I can understand you better I ask you this Why dont you want to do it that way and why does it bother you so much calling the Testcases Gnome since they are Gnome specific?. And we can never be sure that the same bits are installed on both the dvd and the Desktop live cd ( the image are controlled by 2 separate groups ) + the Installer team might decide to leave nothing hashed and ask the end user to select the DE he wants installed from dvd who knows..

Can you please write in paragraphs? My brain melted and dripped out of my ear somewhere around line 16.

The whole point is that I'm not writing GNOME test cases. I'm writing desktop test cases, for the Fedora desktop...however the Fedora project is choosing to provide its desktop at the point in time when you're interested in using the test cases. I don't want to call them GNOME test cases because that's not what they are. I don't see any element of the four test cases I've written so far which is in fact specific to any particular desktop environment.

In https://fedoraproject.org/wiki/QA:Testcase_audio_basic "How to test" Line 2 3 and 5 are indeed Gnome specific.

The more correct way to test this would be to ask the tester to open up a terminal window and run either aplay or paplay /usr/share/sounds/alsa/Front_Center.wav

This is not DE depended and is the first thing to test to establish if sound works or not.

After this test testers can start testing application that support audio and or video playback regardless what they are called or in which DE/Spin that application resides.

In https://fedoraproject.org/wiki/QA:Testcase_desktop_browser "How to test" Line 6 ( which is a bit Mozilla favored ) and Expected Results" Line 3 I recommend not having any extension install testing since we cant guarantee that the extension being installed is not broken.

In https://fedoraproject.org/wiki/QA:Testcase_desktop_updates "How to test" Line for "(e.g. system menu > System > Administration > Software Update)." This is a bit Gnome favored is it not?

Regardless of that I recommend removing that since we never can be sure that this is the same path to update application across DE and especially with our ( Gnome ) Desktop team which is known through out the intertubes for what they call "tidying up the menu(s)" ( rearranging removing applications in menus which is like rearranging removing the furniture's for the blind end user).

I didn't write testcase_audio_basic for this set of tests, it comes from a test day. It was just handy to re-purpose it as part of the desktop acceptance tests. The aplay / paplay test may be a good one to add, but I do want to check that a typical desktop pipeline (including gstreamer) is working in that test.

Browser extension testing was specifically requested by the desktop team when I asked them to give me a set of baseline 'is firefox basically working' tests.

The system menu path given is an example (that's what e.g. means, it's different from i.e.). It's not intrinsic to the process, and can always be updated if necessary.

Nice work Adam. Looks like a list of test cases has been defined (https://fedoraproject.org/wiki/Category:Desktop_Acceptance_Test_Cases) and submitted for review on the mailing list (http://lists.fedoraproject.org/pipermail/test/2010-January/087842.html).

Just pinging status on some tickets. Is there any additional action planned to complete this task?

I'm planning to keep it open till we have the test matrix up and available in final form (whatever that turns out to be).

Replying to [comment:12 adamwill]:

I'm planning to keep it open till we have the test matrix up and available in final form (whatever that turns out to be).

https://fedoraproject.org/wiki/Test_Results:Current_Desktop_Test

Okay to close this ticket?

yup!