[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: before the next beta comes out ..



On 2 Jan 2003, Dennis Gilmore wrote:

> On Wed, 2003-01-01 at 23:52, Robert P. J. Day wrote:
> <snip> 
> not necessarily dangerous from memory each release had its own bugzilla
> you should know what release you are running if you forget you can
> always run rpm -q redhat-release or cat /etc/redhat-release that will
> give you your version you are currently running.  i was assuming that
> posts to the mailing list would assume that the latest beta is being run
> if its not the user should check first to see if the bug has
> subsequently been fixed in a newer release. hence the user should be
> running the latest beta. 
> 
> Dennis

at the risk of inflaming passions, this is asking a *lot* of users
who are already volunteering their time to help locate bugs.

beta releases are, naturally, a two-way street.  users get an
advance look at upcoming software, while red hat gets your
basic thousands of eyeballs to pore over their stuff, and help
an understaffed QA team.  win-win for everyone.

IMHO, beta testers are already contributing a fair amount of
time and effort in downloading and testing the software, putting
up with possible system crashes, verifying that what they've
found is *really* a bug, making sure it's repeatable (if that's
the case), documenting it carefully, and submitting it to 
bugzilla.  i think it's safe to say that beta testers really
do put in some time doing this.

and having done all of *that*, i think it's inappropriate
to give a newer beta the same code name as an earlier beta,
and having it share the same mailing list.  this just causes
more unnecessary work for the beta testers who, when they find
a bug, now not only have to check if it's already been
reported in the *newer* beta, but should for completeness,
see if it was in any *previous*, identically-named beta,
and if it was, whether it was addressed or fixed.  is it
a new bug?  is it just an old bug that's still there?  if
it was reported as fixed, does that mean the fix didn't take?
should i report it again?  when you've got this population
of competent, useful beta testers, it's *really* a bad idea
to make even more work for them.

like i said, i would like to see every beta release constitute
a whole new round of testing and reporting.  and naming.  don't
forget naming. :-)

rday





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]