Re: Best practices for updates from trunk

[Jacques Le Roux]

Fortunately we have now Selenium and Junit. We should be able to do better, but you are right, it's not a simple task and moreover 
not very appealing for most of ordinary mortals

Jacques
From: "BJ Freeman" <bjf...@free-man.net>
ah yes I do remember win9x, I was on the Microsoft Campus during the
time working on a different project.
there was a room of about 200 machines of different configurations
(speed, memory, os settings) that ran the application under test.
There were no tester but Test engineers(5-20) that built test scenarios
in Microsoft own Test software. Their software tested from the UI down
to the core subroutines.
if something was broke is was given a priority as blocking to trivial
like jira.
this just set what was worked on first. however all were resolved.

let me say I am not saying that ofbiz should be at this level, only that
it should be accurate about what is the reality of ofbiz state.

and yes since 2002 ofbiz has come a long way and most of the newbies
don't know what they missed (cough cough).




=========================
BJ Freeman  <http://bjfreeman.elance.com>
Strategic Power Office with Supplier Automation  
<http://www.businessesnetwork.com/automation/viewforum.php?f=52>
Specialtymarket.com  <http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist

Chat  Y! messenger: bjfr33man


Jacques Le Roux sent the following on 8/31/2010 10:22 AM:
From: "BJ Freeman" <bjf...@free-man.net>
I am Glad for this discussion.
My years on Design teams had a lot different criteria.

so to Summarize if something is broke and no one wants to work on it,
it is not considered a bug.

I don't think so, it's considered a bug but not a blocking nor a
critical one.

As far as buildbot it is only as good as the tests written that it
uses and as was pointed out on the dev list ofbiz UI and minilanq is
not tested, as in the math error.

So the fact it "passed" a build only says the test so far, don't fail,
not that ofbiz is bug free.

I was told one day (15 years ago, I guess), that, at this time, for Word
(remember Window 95, tadaaa...), Microsoft had a team of 10 to 20 devs
and at all between 120 to 150 persons Still the testers/dev ratio was
less than 10 (much administrative people too make the glue). At the NASA
(not sure the period), for embedded software the ratio was 100! But
don't take me wrong, I don't mean that we need more bureaucracy ;o)

reminds me of a used car salesman.

I never get there, else I buy a new one (I did once), or rather I buy it
from owner to owner (I know mechanic).
Anyway, it's still far better than what he had a couple years ago (not
my car)... Hopefully it will be far better in 2 years (still not my car)...

Jacques


=========================
BJ Freeman <http://bjfreeman.elance.com>
Strategic Power Office with Supplier Automation
<http://www.businessesnetwork.com/automation/viewforum.php?f=52>
Specialtymarket.com <http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist

Chat Y! messenger: bjfr33man
Scott Gray sent the following on 8/31/2010 2:56 AM:

OFBIZ-3912 is a good example of an issue that is critical to an
individual rather than the community I think. A good indicator is
that it has been broken for 5 months before being reported, can't be
that critical then huh?

Regards
Scott

On 31/08/2010, at 9:45 PM, Jacques Le Roux wrote:

Scott Gray wrote:
A blocking bug IMO is nothing more than the community (rather than
an individual) deciding that a bug is important enough that a
release shouldn't occur until it is fixed. I don't think we need
any criteria other than that really.

Yes that's fine to me. And a critical (for instance at
https://issues.apache.org/jira/browse/OFBIZ-3912)?

Jacques
Regards
Scott
On 31/08/2010, at 9:09 PM, Jacques Le Roux wrote:
Hey Scott, that's right!
Happy to close this one. So eventually you were right, there are
ANY blocking bugs in OFBiz under our criteria ;o)
Though I guess we should better define our criteria for blocking.
Because if we allow to use blocking only for bugs blocking all
OFBiz there should be hardly any such bugs. I mean it would be
very, very quickly fixed and we would have hardly the time to
create a blocking bug. Notbaly because we have now BuildBot
running and all commiters are quickly aware of any errors they
woulds have made. Did you thought about it Scott, what is your
perception? Thanks
Jacques
Scott Gray wrote:
Hi Jacques,
I thought you fixed OFBIZ-3837 recently? Or was that some other
shipping estimate problem you were working on?
Regards
Scott
On 31/08/2010, at 7:52 PM, Jacques Le Roux wrote:
There are currently 275 UNRESOLVED bugs, 11 have patches
available, 14 are reopened.
This is againt all versions. There are only 78 for trunk but
this is not a reliable criteria (if any are)
https://issues.apache.org/jira/browse/OFBIZ-3837 is marked as
blocking. But I guess not under Scott's criteria (it's not
blocking OFBiz, just blocking a part of it, so it's critical
actually) 4 are critical and 173 major
No needs to say that any help to clean things would be really
appreciated...
HTH
Jacques
From: "BJ Freeman"<bjf...@free-man.net>
Quick scan of Jira shows 4 open bugs not counting mine.
=========================
BJ Freeman<http://bjfreeman.elance.com>
Strategic Power Office with Supplier
Automation<http://www.businessesnetwork.com/automation/viewforum.php?f=52>

Specialtymarket.com<http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist
Chat Y! messenger: bjfr33man
BJ Freeman sent the following on 8/30/2010 10:17 PM:
Actually I meant a response to it like there are no bugs that
fit this
condition.
=========================
BJ Freeman<http://bjfreeman.elance.com>
Strategic Power Office with Supplier Automation
<http://www.businessesnetwork.com/automation/viewforum.php?f=52>
Specialtymarket.com<http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist
Chat Y! messenger: bjfr33man
Scott Gray sent the following on 8/30/2010 10:12 PM:
You missed something by reading an email from a community
member and
interpreting it's contents as some sort of official policy.
Regards
Scott
On 31/08/2010, at 5:07 PM, BJ Freeman wrote:
I was reffering to this:
On 07/29/2010 01:53 AM, Ean Schuessler wrote:
I guess the basic questions might be: Did we have an
official "freeze"
process where we only accepted bug fixes against that
branch? Do we
have
any open bugs that are considered "release critical"?
I agree. At least *all* bugs should be looked at, and given
a target
release tag, etc. Time for some bug triage. I don't think
there has
been a concerted effort like that in like, forever.
I have not seen anything that shows this as taken care of.
Have I missed something?
=========================
BJ Freeman<http://bjfreeman.elance.com>
Strategic Power Office with Supplier
Automation<http://www.businessesnetwork.com/automation/viewforum.php?f=52>

Specialtymarket.com<http://www.specialtymarket.com/>
Systems Integrator-- Glad to Assist
Chat Y! messenger: bjfr33man
David E Jones sent the following on 8/30/2010 9:47 PM:
On Aug 30, 2010, at 10:38 PM, BJ Freeman wrote:
There are a number of bugs that have to be resolved before
release
remember this is a volunteer effort so there is no full
time person
doing the patches for the bugs.
What are you basing this on?
once they are completed and testing has been done then
there will
be a release.
right now we are about 4 months past the planned release
date.
What are you basing this on? Maybe there is some
confusion... the
date on a release is the date it was branched from the
trunk, not
the date that a binary release is done from the branch.
-David
Matt Warnock sent the following on 8/30/2010 9:27 PM:
Thanks. Is 10.04 nearing release? Is there an expected
release date?
Just curious.