At 10:18 3-6-2002, Markus Fischer shared with all of us:

>    This is true, but ... there is absolutely _nothing_ we can do
>     about it. The QA Team is, well .. sometimes I get the
>     impression it's only one person!

Derick had a good thing going, when he emailed people who reported
problems and needed confirmation of a fix. That stopped.

Besides that - there's no involvement of QA when extensions get moved,
broken or added new functionality - just at RC time.
Yes - most of us, will read php-dev, but a heads-up when something major
has changed with an extension is not a bad thing.

This will allow the extension maintainer to fix things when it's still
fresh - not when there's a release x weeks/months down the road, and he's
devoted his time elsewhere and can't do anything about it anymore.

>  Until we prison some people
>     to do it, I think our QA or Release Management  will not
>     improve that much. Other hot topics in this area would be
>     automated test system (more then the current make test does).

The current (=4.3.0-dev) make test at least provides a consistent output 
and logging.
With a simple command all reports can be collected:
find . -type f \( -name "*.log" -a \! -name "config.log" \) -print | xargs 
tar -cpvzf ./failed.tgz

Add that to the test target in the makefile and instructions on where to upload
it, and you'll make test-reporting a whole lot easier.

Met vriendelijke groeten / With kind regards,

IDG.nl
Melvyn Sopacua
Webmaster


-- 
PHP Development Mailing List <http://www.php.net/>
To unsubscribe, visit: http://www.php.net/unsub.php

Reply via email to