• Unknown User
    Unknown User, 01/08/2014 00:11

     REVIEW 1 

    PAPER: 6 TITLE: On the “Local-to-Global” Issue in Self-Organisation: Chemical Reactions with Custom Kinetic Rates AUTHORS: Stefano Mariani

    OVERALL EVALUATION: 1 (weak accept) REVIEWER'S CONFIDENCE: 4 (high) Relevance: 4 (good) Technical Quality: 4 (good) Originality: 4 (good)


     REVIEW 

    I thought that this was an interesting attempt to approach the local-to-global problem from a fresh perspective. The author uses a process algebra with custom rates to allow the expression of local phenomena and builds up a library of patterns which others can re-use. To the best of my knowledge I think that this is novel. I found the paper well-written and relatively easy to follow.


     REVIEW 2 

    PAPER: 6 TITLE: On the “Local-to-Global” Issue in Self-Organisation: Chemical Reactions with Custom Kinetic Rates AUTHORS: Stefano Mariani

    OVERALL EVALUATION: 2 (accept) REVIEWER'S CONFIDENCE: 4 (high) Relevance: 5 (excellent) Technical Quality: 4 (good) Originality: 3 (fair)


     REVIEW 

    The paper is mostly well written and tackles a very interesting issue in collective adaptive systems, i.e. the "local-to-global" issue. The approach taken by the author is to use proper tools for modeling, validate and check self-organizing primitives, so to exploit and describe the simplest "building blocks" of a generic Self-* system.

    These primitives are modelled through a very interesting paradigm such as artificial chemistry and then expressed through the BioPEPA simulator. I found this approach very interesting and I agree with the author when he says that the tools used with this approach are indeed valuable instruments for the designer of self-* systems.

    However, it is not clear from the presented paper how exactly being able to plot such primitives can help us understand a global behaviour starting from the point of view of a "local" component of the system: as far as I have understood we can model "custom kinetic rates" according to our specific problem, so it is basically a different approach to parameter tuning. Specifying what is "local" and what is "global" in this paper would definitely provide more help in understanding how exactly the "local to global" issue has been addressed.

    Moreover, 100 Gillespie runs have been performed for each plot: what exactly changed from one run to the other?

    To conclude, I think that this paper presents very important issues that definitely deserve a proper in depth discussion at the workshop, therefore I  recommend acceptance. Clarifying the points that I've stressed in this review will help in presenting a stronger message to the audience.

    Also, some format errors should be fixed:

    Unspecified references (marked with ??) in the introduction. Paper length.

2011 © aliCE Research Group @ DEIS, Alma Mater Studiorum-Università di Bologna
1.1