Last Friday I gave a guest lecture to third-year Informatics students on the Software Design and Modelling course.
Professor Stevens, who leads the course, asked me to repeat the presentation that I gave last month to an audience of University staff. She thought that many of the issues I covered would be relevant to the course, and the topic of improving the online student experience was clearly one that the students could relate to. It's a long time since I did my own degree but unless times have changed markedly, I suspect that students don't often get to see the issues around integrating many pre-existing systems, rather than building small systems in the lab.
I enjoyed the session. I don't often get to meet students, so this was a refreshing experience, and we had a good discussion following the presentation. As expected, they confirmed that the online experience currently provided by University systems is "all over the place".
One of the questions was about whether we made much use of UML or other modelling languages in Information Systems. I explained that we only use UML informally, for example to draw class diagrams or activity diagrams when discussing a design. We don't use modelling tools as such. A few years ago, our business analysts did try using BPML to model business processes but that initiative ran out of steam. I think the problem was that keeping the documentation up to date was too clumsy and time consuming a task.
As I noted in an earlier post, I am looking at Archimate for architecture modelling. We'll have to see how well this works in practice; currently I am still only experimenting with its use. I think architecture may be a more suitable level at which to use modelling languages, partly because architecture doesn't change quite so quickly as software design, and partly because it is not easy to see the information being presented via other channels (compared with a class structure, for example, which you can view in a developer's programming tool).
In preparation for this talk, I did modify my presentation slightly. I added a couple of slides that showed "as-is" architectures for a couple of areas, to make concrete the issues that make the delivery of an integrated online experience somewhat less than straightforward. I talked about the problems of integrating third-party systems, especially where these are not implemented in the most modern technologies. This gave some context to the rest of the talk which the previous audience would have already understood.
As this audience was of future software designers, I put my discussion of business processes first, where it belongs. In my earlier presentation, I left business processes to the end, as I didn't want to challenge that audience out of the blocks and maybe put them off listening to the rest of the talk. In practical terms, of course, we need to start with business processes or we are doomed to fail.
All in all, this was a very enjoyable experience. If other ISG staff get the opportunity to give a talk to an audience of students, I would definitely recommend it.
Professor Stevens, who leads the course, asked me to repeat the presentation that I gave last month to an audience of University staff. She thought that many of the issues I covered would be relevant to the course, and the topic of improving the online student experience was clearly one that the students could relate to. It's a long time since I did my own degree but unless times have changed markedly, I suspect that students don't often get to see the issues around integrating many pre-existing systems, rather than building small systems in the lab.
I enjoyed the session. I don't often get to meet students, so this was a refreshing experience, and we had a good discussion following the presentation. As expected, they confirmed that the online experience currently provided by University systems is "all over the place".
One of the questions was about whether we made much use of UML or other modelling languages in Information Systems. I explained that we only use UML informally, for example to draw class diagrams or activity diagrams when discussing a design. We don't use modelling tools as such. A few years ago, our business analysts did try using BPML to model business processes but that initiative ran out of steam. I think the problem was that keeping the documentation up to date was too clumsy and time consuming a task.
As I noted in an earlier post, I am looking at Archimate for architecture modelling. We'll have to see how well this works in practice; currently I am still only experimenting with its use. I think architecture may be a more suitable level at which to use modelling languages, partly because architecture doesn't change quite so quickly as software design, and partly because it is not easy to see the information being presented via other channels (compared with a class structure, for example, which you can view in a developer's programming tool).
In preparation for this talk, I did modify my presentation slightly. I added a couple of slides that showed "as-is" architectures for a couple of areas, to make concrete the issues that make the delivery of an integrated online experience somewhat less than straightforward. I talked about the problems of integrating third-party systems, especially where these are not implemented in the most modern technologies. This gave some context to the rest of the talk which the previous audience would have already understood.
As this audience was of future software designers, I put my discussion of business processes first, where it belongs. In my earlier presentation, I left business processes to the end, as I didn't want to challenge that audience out of the blocks and maybe put them off listening to the rest of the talk. In practical terms, of course, we need to start with business processes or we are doomed to fail.
All in all, this was a very enjoyable experience. If other ISG staff get the opportunity to give a talk to an audience of students, I would definitely recommend it.
Comments