incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From NoOp <gl...@sbcglobal.net>
Subject Re: Calling all volunteers: It is time to test
Date Fri, 02 Mar 2012 04:30:07 GMT
On 03/01/2012 11:35 AM, Rob Weir wrote:
> On Thu, Mar 1, 2012 at 2:24 PM, NoOp <snip...configure your client properly> wrote:
> 
> <snip>
> 
>> So in order to avoid further confusion, I still think it would be
>> helpful to clarify which version the 'Call to volunteers' are supposed
>> to be testing (ooo-dev or 'full normal install').
>>
> 
> I think testing either install set is fine.  Most bugs are in common
> or are clear from the context. But I'll try to be more clear with
> future notes.

Thanks, a followup on the user list regarding this is necessary (IMO).

I also suggest that AOO follow the tried & true method of posting
DS/Beta/RC et al on:
http://www.openoffice.org/download/index.html
rather than on a "Calling all volunteers: It is time to test" with a
unclear link to
<https://cwiki.apache.org/confluence/display/OOOUSERS/AOO+3.4+Unofficial+Developer+Snapshots>
and asking users/testers to go there to download rather than some
cwiki.apache.org page.

Further I very much disagree that "testing either install set is fine".
I've already pointed out that a test 'reporter' is not able to
distinquish between the two via the Help|About results.
  Were I a developer and/or QA person I'd definitely like to know if the
tester/reporter is testing/reporting version x (ooo-dev) or version y
(openoffice full standard).

It is obvious that you did exactly what I tried to point out; you sent
out a msg asking users to "If you are ready to test, you can download
the latest developer snapshot build from this page". You then downloaded
the other: OOo_3.4.0_Linux_x86_install-deb_en-US.tar.gz instead &
reported the install (albeit with new tester results from
'/openoffice.org3.4-debian-menus_3.4-9586_all.deb' - it's OK we've all
been there done that).

I've already pointed out that there are differences between the ooo-dev
builds and the ooo 3.4 builds. A tester on the user list has already
reported a printing difference between the two... see: "Printing problem
with OOo-dev 3.4" from Josef Latt. To simply fob off testing between the
two (particularly when I've pointed out that both report the same in
Help|About). So how are any 'evaluators' able to tell if the 'report'
from user x (your's for example) is from someone that is testing x or y?
I certainly hope that further "tests" on AOO/OOo do not take a similar
shotgun approach.

I think it also *very* important to point out to user/testers that
downloading and installing one or the other will wipe out/overwrite
existing installations of earlier versions. Granted that not many
standard users will have had earlier ooo-dev or ooo 3.4.x versions
installed before testing. *But* if they had it is *important* to advise
them that installing these new untested versions will replace their
existing installs. And yes, I did have previous versions of both
installed (linux and Windows) and the r1293550 versions overwrote the
existing installs. For me that is not an issue, but for a standard user
it may very well be a *significant* issue.

I fail to understand why it is so difficult for you to: 1) reply to my
questions on the user list (afterall you did post there asking 'users'
to test & you've still not replied on that list), and 2) just post a
followup to the original "Calling all volunteers: It is time to test"
and clarify that users should download and test x or y... and also
please report which version you've downloaded, installed, and tested.

Point being is that I think that AI should review, revise, and fix how
you ask & promote pre-release testing. I do not believe that there need
to be new rules as the process/rules were well established under OOo.
Follow those & perhaps AIOO can improve on them. I'd also pay particular
attention to the former OOo release mailing list whereby folks could
report 'showstoppers', release test issues, etc.

Gary Lee

...


Mime
View raw message