Mailinglist Archive: opensuse-testing (24 mails)
| < Previous | Next > |
Re: [opensuse-testing] Building the OpenSUSE testing team
- From: Andreas Jaeger <aj@xxxxxxx>
- Date: Thu, 19 Mar 2009 10:49:44 +0100
- Message-id: <200903191049.45429.aj@xxxxxxx>
On Thursday 19 March 2009 10:32:55 Tilman Schmidt wrote:
We need both IMO. We need on the one hand a defined set of use-cases that the
distribution is tested against - so that in the end we can say: All use-cases
were tested and that's their status so we know exactly which fail and can base
a decision of releasing on that.
On the other hand, we need to test new features as well.
So, let's look at other ways of coordinating as well. IRC can help a lot but
should not be a prerequisite for coordinated testing.
If you have any concrete examples, let's discuss these.
Andreas
--
Andreas Jaeger, Director Platform / openSUSE, aj@xxxxxxx
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
Maxfeldstr. 5, 90409 Nürnberg, Germany
GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126
Satoru Matsumoto schrieb:
However, during the 11.1 release process, there was a call for feature
testing.
http://lists.opensuse.org/archive/opensuse-testing/2008-11/msg00005.html
But you can see most of the Test Results for listed features remain
"Idle". http://en.opensuse.org/Testing:Features_11.1
We could know *what to test and when* in this case, but as far as I can
see, not so many people participated the testing.
I think we have to analyze why many people don't have keen interest to
testing. Do we need to spread the word much more ? Do we need to
clarify what and how things to be tested and where the result should be
reported ? Or, do we need to prepare benefits for dedicated testers ?
This particular call for testing had two reasons why I didn't consider
participating:
a) It called for testing new features, but I am much more interested,
and consider it much more important, to make sure the old features
still work correctly, ie. there are no regressions.
We need both IMO. We need on the one hand a defined set of use-cases that the
distribution is tested against - so that in the end we can say: All use-cases
were tested and that's their status so we know exactly which fail and can base
a decision of releasing on that.
On the other hand, we need to test new features as well.
b) "Basic reqirements:
* install openSUSE 11.1 beta 4
* meet in #opensuse-testing on irc.freenode.org"
I do not use IRC.
So, let's look at other ways of coordinating as well. IRC can help a lot but
should not be a prerequisite for coordinated testing.
Apart from that, I must admit that my testing enthusiasm has been
somewhat dampened by a recent series of WONTFIX resolutions for
problems I reported, a few of them with closing remarks that seemed to
imply what I reported was "my problem". So I guess positive feedback
from developers that test reports are appreciated, taken seriously,
and the reported issues addressed, would go a long way in encouraging
more testers.
If you have any concrete examples, let's discuss these.
Andreas
--
Andreas Jaeger, Director Platform / openSUSE, aj@xxxxxxx
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
Maxfeldstr. 5, 90409 Nürnberg, Germany
GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126
| < Previous | Next > |