28 May 2010

Architecture reconstruction

Each spring a project course called Industrial IT and Embedded Systems is given at the IT University. A description of the project from 2009 can be read here.
The students used trac to support their project and you can access the public project pages here.

This year I coached the students responsible for defining the overall architecture at the beginning of the project.
The project is finished with an official demo of the robot's capabilities, but some students approached me with an interest of doing a "post mortem" of the software as an additional learning experience. I thought this would be quite interesting, enough so to share any experiences on this blog.

These were the primary goals the students defined for the exercise:
  • Learning more about architecture
  • Only analysis, not update current project (code)
  • Reconstruct existing robot architecture
    • Automatic reconstruction using a tool
    • Create "logical view" from source code, showing dependencies on function level
    • Variable scope, global variables
    • Process view
  • Evaluation, compare against first architecture

Second priority goals:
  • What could have been done in other ways and what impact would those changes have made
  • Improvement Ideas

The students doing the reconstruction use the excellent tool Pirate Pad to capture notes and common ideas. Pirate Pad is a simple pre-runner of Google Wave where you don't have to register to create a pad, everybody just access the same web address and start collaborating.

26 May 2010

Yet another article

The blog is updated on a very irregular basis. the main reason is that I presently have 3 articles in the pipeline and this takes priority over anything else. The latest article which I need to update is to ECSA 2010:


Unfortunately your paper has not been accepted 'as it is' for inclusion in the program of the 4th European Conference on Software Architecture 2010 (ECSA 2010), which will be held on August 23-26 in Copenhagen, Denmark.

However, the program committee felt that the work reported in your paper has a number of interesting ideas, hence, I would like to invite you to prepare a shortened version of your paper for inclusion as "Emerging research" category Paper in the ECSA 2010 proceedings. As noted on the conference website, the "Emerging research" papers (8 pages of Springer formating) present promising achievements from work-in-progress and are intended to stimulate discussion related to ideas and experiences. For further details about the "Emerging research" papers, plesae visit ECSA 2010 website (http://www.ecsa2010.org). To accommodate publication schedules, final papers must be submitted by June 11, 2010.

... snip ...

4. The camera ready copy of your paper is due by June 11, 2010 (This is a hard deadline!).

... snip ...

Once again, congratulations on having your paper accepted as an "Emerging research" paper. We look forward to receiving your final camera ready paper, and to seeing you in Copenhagen, Denmark.

If you have any question or comment, please email me.



ECSA 2010 Program Chair.

4 May 2010

Time to plan a trip to Korea...

The paper Architecting Automotive Product Lines: Industrial Practice by Håkan Gustavsson from Scania and myself have been accepted to the 14th Software Product Line Conference.
"Dear authors,

Thank you for your submission to SPLC 2010.
We are pleased to inform you that your paper has been accepted for inclusion in the SPLC full paper program.
Due to strong competition, we could only accept 28 full papers out of 90 submissions, giving an acceptance rate of 31%.
Congratulations on being included in this select group.

... snip ...

Once again, congratulations on having your paper accepted.
We look forward to receiving your final revised paper, and to seeing you in September in Juju Island, South Korea.


Jan Bosch and Jaejoon Lee
PC Co-chairs
SPLC 2010"