Mailing-List: contact ooo-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: ooo-dev@incubator.apache.org
Received-SPF: pass (athena.apache.org: domain of frank.schoenheit@gmx.de
 designates 213.165.64.22 as permitted sender)
Message-ID: <4E779DFB.4030607@frank.schoenheits-abc.de>
Date: Mon, 19 Sep 2011 21:54:35 +0200
From: =?ISO-8859-1?Q?Frank_Sch=F6nheit?= <frank.schoenheit@gmx.de>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
 rv:1.7) Gecko/20040620 Mnenhy/0.7
MIME-Version: 1.0
To: ooo-dev@incubator.apache.org
Subject: Re: VCL TestTool
References: <4E760F80.3090401@gmx.ch> <4E762547.1000603@gmx.net>
 <CAP-ksojigQc_6j4N15VyjMhtUmz_qfwQZSCK4u7_jy7r_5r5Cw@mail.gmail.com>
 <20110918204144.4fe8b38e@beryllium> <4E766F1F.1080101@gmx.net>
 <4E76EEB9.2040406@frank.schoenheits-abc.de> <4E770E1F.30403@gmx.ch>
In-Reply-To: <4E770E1F.30403@gmx.ch>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

Hi Raphael,

> All good Ideas, but we should not investegate blind too much time in the 
> TT. We realy need to evaluate, what's the benefit, and how much time it 
> cost. It's a difference if we find 10 Regression or 500 in one year with 
> the TT. So the first thing is to see how much time TT save for QA.

I'm absolutely with you on this.

> In general, I'm for automated testing, because this is a good method to 
> find regressions. Skilled QA people are rare, so they should not waste 
> time with manual testing to avoind regression.

The reason why I invested time into a Java TT client was exactly this
"Skilled people should not waste time with ..."-thingie. Writing,
reading, and maintaining those Basic scripts is an art of its own. In
the rare moments I needed to deal with them as developer ('cause there
was some TT-reported bug which could not be reproduced manually, but was
nonetheless considered worth fixing), I bit into my desk multiple times,
simply 'cause working with TT is as ... unpleasant as anything.

So, *if* we say that the idea of the TT has some value - and I tend to
agree to Mathias that it indeed has -, then we should evaluate its
biggest obstacles, and work on them. To me, the Basic TT IDE is one of
those most important facets to improve.

Of course, one might think about whether the current automation approach
- in terms of how it is technically done, with all the problems as false
alarms, no complete mimic of real world user behaviour, and the like -
is, well, sufficient for the 21st century. I never thought about this in
deep, but I wouldn't be surprised if people smarter than me would come
to the conclusion that implementing this from scratch might be a better
solution. I, myself, have neither the expertise nor the time to embark
on this, so I started with were my expertise lies - which was the Java
client :)

Ciao
Frank