From: Andrew Gabb (agabb@tpgi.com.au)
Date: Mon Jul 26 2004 - 18:56:41 EST
I'm not a professor, but regardless ...
The appropriate "current requirements engineering practices" should work anyway. I think that it's important to note that 'agents' are really nothing new - it is really the environments and the uses of agents which are changing. And the names of course.
In some ways the agent oriented approach is an extension of systems engineering in determining the components of a system, their requirements and their behaviour. Modern definitions of architecture, for example, also include the behaviour of the components. Similarly, optimisation of a system requires a system view, rather than optimisation of the individual components (which for agents may also be competing for resources).
One of the thing that concerns me about 'new' development models like 'agent oriented' is that they are usually developer-centric, ie they tend to either ignore requirements or assume that the developers know and understand the requirements well. This was particularly obvious in OO models and earlier SwE texts. One classic example was the SW-CMM which virtually ignored requirements. This crime of omission is perpetuated by both practitioners and academics, so it's hard to fix. I see agent-oriented going the same way.
So I do see this as a potentially useful area of research, but it needs people who are experienced in the types of systems envisaged, experienced in the needs for these systems, and close to other research regarding agent oriented approaches.
Andrew
"Jose Antonio Macias Trejo" wrote:
> Dear Professors,
>
> I'm currently studying for my master's degree and am in the process to subm=
> it a proposal for my final dissertation. Due to my great interest in Requir=
> ements Engineering, I would like to prepare a work that would help me to go=
> on for a PhD on next year in this field of study.
> I would like to ask for your wise advice about an idea which I=92ve come up=
> with recently. Would it be possible that the current requirements engineer=
> ing practices are not suitable for developing agent-oriented software?
> I=92m posing this question based on the fact that the Agent-Oriented paradi=
> gm proposes a different way to conceptualize the development of this kind o=
> f software.
> I=92d be glad if I could have your comments and suggestions about this issu=
> e. Should you let me know your different points of view I=92d be very pleas=
> ed. Additionally, ideas about cutting-edge fields of research in Requiremen=
> ts Engineering are welcomed.
-- Andrew Gabb email: agabb@tpgi.com.au Adelaide, South Australia phone: +61 8 8342-1021, fax: +61 8 8269-3280 ----- ------------------------------------------------------------------------------- To send a message to this mailing list send it to re-online@it.uts.edu.au. To unsubscribe from this mailing list, email majordomo@it.uts.edu.au with the message `unsubscribe re-online' in the BODY of the mail.
This archive was generated by hypermail 2.1.6 : Mon Aug 02 2004 - 09:00:20 EST