W dniu 10.12.2011 r. 18:24, Andrzej Jarzabek pisze:
> On 10/12/2011 16:36, A.L. wrote:
>> On Sat, 10 Dec 2011 10:20:37 +0100, Edek<e...@g...com>
>> wrote:
>>
>> Agile: Code first think later. Rzeczywiscie, potrzeba jest matka
>> wynalazku.
>>
>> Wyzszy stopien Agile (Agiel advanced): Code first
>
> Lepiej najpierw pomyśleć, potem jeszcze pomyśleć, potem wyprodukować
> mnóstwo papierków, w końcu napisać kod, którego połowa nie działa w
> ogóle, a druga połowa robi nie to, co trzeba.
"Empirical studies of agile software development: A systematic review",
T. Dyba, T. Dings?yr

Abstract:
The search strategy identified 1996 studies, of which 36 were identified
as empirical studies.

"4.7.3. Product quality
Several aspects of product quality were examined by the
studies in this review. For example, comparing the results
for a new release of a project to those for an old release,
Layman et al. (S14) found a 65% improvement in pre-
release quality and a 35% improvement in post-release
quality. Ilieva et al. (S10) found 13% fewer defects reported
by the customer or by the quality assurance team in an XP
project than in a non-XP project.
In Wellington et al.'s (S32) study, the XP team's code
scored consistently better on the quality metrics used than
the traditional team. In addition, the quality of the code
delivered by the XP team was significantly greater than that
delivered by the traditional team. However, both teams
agreed that the traditional team had developed a better
and much more consistent user interface.
Macias et al. (S15) measured the internal and external
quality of the products developed by 10 XP teams and 10
traditional teams. However, in contrast to Layman et al.
and Wellington et al., they found no difference in either
internal or external quality between the XP teams and
the traditional teams.
With respect to product size, the XP model teams in
Dalcher et al.'s (S7) study delivered 3.5 times more lines
of code than the V-model teams. This is in sharp contrast
to Wellington et al.'s (S32) results, which showed that the
traditional team delivered 78% more lines of code than
the XP team. However, in contrast to both Dalcher et al.
and Wellington et al., Macias et al. (S15) found no differ-
ence in product size between the XP teams and the tradi-
tional teams."

Hmmm....