User-centered design and end-user development (May. 10)

Participatory design has been a part of ICT development for many decades, as we have seen earlier. But there are two specific strands that have become increasingly important and may well redefine the whole information systems or informatics discipline: user centered design and end-user development. User centered-design (UCD), according to Mao et al. (2005) is “a multi-disciplinary design approach based on the active involvement of users to improve the understanding of user and task requirements, and the iteration of design and evaluation”. By now, this shouldn’t sound surprising or revolutionary, most modern (agile) software development methods emphasize the active involvement of customers or users in an iterative fashion. However, the clear emphasis here is on user-centeredness and on multidisciplinarity. This means that beyond being participatory, the design effort is also focused completely on user needs, rather than the traditional system-centered design which may devote more attention to non-functional requirements, management input or algorithms. Multidisciplinary participation is also brought forth as a natural element in UCD, partly because users come from different disciplines, but more importantly, because different disciplines are required to capture their needs, coordinate their participation and translate requirements into technical specifications together with usage guidelines, manuals, training, change management, and economic and cultural viability assessments and transitions with an eye on sustainable usage and benefit.

Mao and others carried out a survey in which they found that the application of UCD means an improvement in usefulness and usability as well as time saving in the long run (mostly by reducing reworking deployed applications, which is significant given that between 70 to 90% of all software costs occur during “maintenance”). These arguments easily map to the pragmatic proposition behind participatory design (discussed in the previous entry), but it is also clear that the moral proposition is a key issue and that participatory, qualitative methods are required to carry it out in practice, including field studies, iterative design, focus groups, interviews, and task analysis. In an era where consumers are prosumers (well aware of technical trends and capabilities)and where alternative ICT tools may be just a click away, catering responsibly and effectively to users is no longer just desirable but mandatory. If we fail to do so, someone else will do it, including the users themselves…

End-user development (EUD) expresses precisely this behavior when the users themselves end up customizing (personalizing), configuring or even downright programming the software themselves, as discussed in Fischer et al. (2004) . As programming languages become lighter, more high level, easier to use, to access, to install and to get training for, this is becoming increasingly the case. For a user to develop his or her own webpage, not much more than a simple text editing software is required. A few hours browsing through tutorials on CSS, Javascript, PHP, XHTM or other similar tools or languages may be all that is required for more advanced functionality, not to mention the fact that actually getting it online can be completely free. More generally, data processing tools, statistical software, office suites, open source software have all become easier to use and full of customization possibilities, Wizards, or drag-and-drop features that make development at least seem less complicated and which has resulted in large amounts of amateurs developing their own systems. This is not new, there has always been the clerk or secretary that manages to create a very functional tools from simply tailoring a spreadsheet, but the scope and accessibility of languages and development tools is much wider now, enabling users to create ever more sophisticated applications by themselves. It is obvious that this is not without risks (heterogeneity, security, integration, quality, scalability, transferability are all potential issues), but it is also to be expected that the trend will continue. On a strategic level, IT professionals really need to look into this, as it may reshape their role and even their very livelihood. Today big software companies may end competing with basement teenagers that develop mobile applications in a matter of hours generating mass consumption software with much higher returns on investments. The gaming and entertainment industry is also a rich setting for end-used development , which is no small feat considering that games have for sometime been the highest grossing software applications on the market. But from a more operative point of view, the key is for IT professionals to rethink their role not as expert designers, but rather as facilitators of a collaborative process where techbnology os not the end but the means.

For example, in a business organization in order for EUD to be successful, Fischer et al. argue that it should include sustained motivation from the developer (the user), it should be supported by the right tools and it should have management support. Furthermore, they argue for “meta-design” as a way to (under) develop systems such that the result is a socio-technical environment that empowers users to become active and continuously engaged in the development and evolution of the systems which they use. This obviously goes beyond the technical and as such is connected with the socio-technical tradition that has guided all our discussions throughout the course.  Indeed, meta-design can also be linked to the work of Maturana and Varela (which you may recall were deep inspiration for the Winograd and Flores book with which we started the course). For them, technology is not the issue, the issue is rather what kind of humans we wish to become and what kind of culture we should strive for. Rationality, for them, is not the answer, the answer is emotional – there is a whole cognitive, systemic, autpoietic theory to back this up, but it goes beyond this entry. We should aim at becoming homo sapiens amans. As such, we should start by empathizing, indeed loving, the users for which we are designing solutions, and getting off our high horses or our isolationist, detached expert positions (all too common in IT professionals). It should be a complex, challenging and evolutionary path, but certainly exciting.