On Wednesday and Thursday, March 10-11, 2010, I witnessed my first software launch. After 8 years of development and already substantial use, TAO, an open source computer-based assessment (CBA) platform was officially launched. The development of this online assessment tool started as a partnership of the Research Unit EMACS of the University of Luxembourg and the CITI department of the Centre de Recherche Public Henri Tudor in Luxembourg in 2002.
The two days were filled with presentations about CBA in general, the partnership, the history of the development of TAO, and its use in international assessments such as PISA and PIAAC. Although TAO was not officially released as open source software until now, it has already been in use for some time now testing it thoroughly and improving it constantly.
Besides presentations, there were user and developer sessions dealing with all aspects of getting to know TAO. The workshops ranged from setting up TAO on a server to creating test items and running the tests. Thus, we could gain first-hand experience of the software.
TAO is being developed as a versatile and generic solution for CBA instead of being geared towards a very specific test. The aim was and is to create a system that can accommodate as many CBA tests as possible allowing also for the exchange of tests among the TAO users.
The launch did not only kick off version 1.0 of TAO, but also marked the beginning of the user and developer community for this open source system. The developers know that this official start is not the end of their work but the beginning of a more public development to improve the software.
In the user workshops which I attended we learned how to set up users, test takers, tests, test items, and the delivery of the test. I was very impressed by the possibilities of TAO, but also saw that there is a lot of work to be done in regard to usability. Being used to a great number of easy-to-use social media, some of the steps I had to take in the TAO system seemed not necessary and sometimes even annoying. However, I also learned that the usability guys and gals had not taken up their work until a few months ago and are working hard to make the software more user friendly.
During the final round of presentations, William Thorn, senior analyst with the OECD, brought the metaphor of haute couture vs. prêt-à-porter into the discussion. This metaphor was taken up by Marilyn Binkley and Romain Martin (both University of Luxembourg). Haute Couture was seen as where creativity goes on, where new developments are brought about to further CBA. At the same time it also signifies developments that have not yet been thoroughly tested and that cannot be applied everywhere, e.g. for technical reasons. On the other hand, prêt-à-porter assessment is well-tested, easier to administer, immediately usable. However, that does not necessarily mean paper-and-pencil tests, but can also be CBA. In contrast to haute couture CBA, it would only have little added value to paper-and-pencil tests. Just having multiple choice items on a computer screen instead of on paper is not very innovative. Like with elearning activities, the real added value must be something else, e.g. the use of multimedia, new forms of test items etc.
Altogether between 80 and 100 people attended this two-day event at the Abbaye de Neumünster in Luxembourg City. There were plenty of opportunities for discussions in the workshops and the breaks where developers, users and potential users mingled. The TAO Days were very well organized by the TAO team and also included a Walking Dinner for a selected number of guests (photos).
The coming months will show how quickly the word can spread about TAO and hopefully get the community to a good start.
