Site Sponsor
Go WordPress
Site Sponsor
Quality of ATL model transformations (help needed)
Marcel van Amstel and Mark van den Brand are working on measuring the quality of ATL model transformations.
The basic idea is to find the correlation between several metrics on ATL transformations (number of rules, number of bindings, number of variables,… see their paper “Quality Assessment of ATL Model Transformations using Metrics” on the MtATL 2010 workshop ) and several quality properties (like complexity or readability)
To do this, they need to know the perceived quality of a transformation according to the opinion of ATL experts (yes, this is you!). They have prepared an experiment to gather empirical data on this topic. The strength of the results of this experiment will benefit from the amount of model transformations that are evaluated so please participate!. Completing it will take approximately 10 to 15 minutes of your time.
If you enjoyed this post you can subscribe to this Software Modeling blog and/or follow me on twitter or through the portal’s mailing list . AND if you really liked it help me pass it ON TO others by bookmarking AND sharing the post USING the links below:
Are they planning to do double-barreled correlation, as per El Emam et al’s 2001 paper [1], TO see how much OF the effects they measure can be explained simply by NUMBER OF lines OF code? El Emam et al FOUND that *ALL* OF the predictive power OF various OO metrics (e.g., Chidamber AND Kemerer’s metrics) were due to that.
[1] Khaled El Emam, Saida Benlarbi, Nishith Goel, and Shesh N. Rai: “The Confounding Effect of Class Size on the Validity of Object-Oriented Metrics.” IEEE Transasctions on Software Engineering, 27(7), July 2001.
- Greg
Greg, thanks for your comment. It would be interesting to see if this is the same for model transformations. Let´s see what they have to say about that
The paper you advised looks very interesting. I have not heard about it before.
In an earlier study we conducted in which we related metrics to quality attributes for ASF+SDF model transformations [1] we indeed found a relation between the size of a transformation, for example in terms of number of transformation functions, and certain quality attributes like understandability and modifiability. However, a relation between understandability and size metrics is a relation that in my opinion is to be expected. Therefore, I am not sure whether controlling for size is a good idea for all quality attributes. Nevertheless, I will look into this in more detail. Thank you for the suggestion.
[1] M.F. van Amstel, C.F.J. Lange, M.G.J. van den Brand: “Using Metrics for Assessing the Quality of ASF+SDF Model Transformations.” In Theory and Practice of Model Transformations, Proceedings of the Second International Conference on Model Transformation (ICMT’09), volume 5563 OF Lecture Notes IN Computer Science, pages 239-248, Zürich, Switzerland, June 2009.
–Marcel