[Sigia-l] Usability Testing comments from Giga

[Pittas Marios]

> It depends. If testing was used for verifying hypothesis, new testing
> is required for new hypothesis.

Sure "if testing was used for verifying a hypothesis then new testing is
required for new hypothesis...", which is true in any and every
situation/application area etc. However many tests (including some usability
tests) are not done for verifying hypothesis (although research work is most
"often" carried this way)... they are done to "measure the usability of a
system", find usability defects, etc... just like a building engineer
measures the "structural soundness of a building", which brings us to...

> Genuine IA is about structures. Thus, a natural way to test is
> subjecting the structural specification to various scenarios to see
> how the structures will stand up.

which I would not dispute.. for me closure of this threat is: "could you
provide me with any references how one goes on about "subjecting the
structural specification to various scenarios to see how the structures will
stand up"..,  who should write these scenarios, when should the tests be
conducted, should alternative structures be given, who should evaluate the
results of the test cases, case studies etc. i.e. how do we fit IA into the
rest of the software engineering development processes..

Marios

-----Original Message-----
From: sigia-l-admin at asis.org [mailto:sigia-l-admin at asis.org]On Behalf Of
Boniface Lau
Sent: 02 April 2003 08:47
To: sigia-l at asis.org
Subject: RE: [Sigia-l] Usability Testing comments from Giga


> From: sigia-l-admin at asis.org [mailto:sigia-l-admin at asis.org]On
> Behalf Of Pittas Marios
>
> Boniface wrote:
>
> > More accurately, you *were* sure of them.
>
> > Thus, you are not sure anymore.
>
> But if the users change (through time/experience/geo-location) so
> could the results of the Usability Testing right (have a look at
> Mayhew's excellent book on Usability Engineering: "The Usability
> Engineering Lifecycle" with regards to building and using User
> Profiles and how they lend to System Development and Design)?

It depends. If testing was used for verifying hypothesis, new testing
is required for new hypothesis.

But if testing was used as a catch-all, typically by people who don't
know what they are doing, re-testing is required whenever the usage
conditions changed. There are many such people around. All they know
is that such and such worked as seen in testing. But they don't know
why it worked. Whenever usage conditions changed, they feel uncertain
as to whether such and such will continue to work. So, they need to
test again.


>
> A simple parallel would be the way that developers would test their
> apps should the Operating System change?

Not really. Users come and go but an operating system remains the same
until the next OS change. Since there is no such thing as an "average
user", every user is unique in some ways.


>
> > Testing is typically done when its cost is much lower than the
> > potential cost of not testing. It is a part of risk management.
>
> Sure, nobody said otherwise,

Some did - by suggesting that people test everything.


Boniface
------------
When replying, please *trim your post* as much as possible.
*Plain text, please; NO Attachments

Searchable list archive:   http://www.info-arch.org/lists/sigia-l/
________________________________________
Sigia-l mailing list -- post to: Sigia-l at asis.org
Changes to subscription: http://mail.asis.org/mailman/listinfo/sigia-l
---
[This E-mail was scanned for viruses.]




---
[This E-mail was scanned for viruses.]