Path: news.daimi.aau.dk!glad From: Albertina Lourenci Newsgroups: comp.lang.beta Subject: polymorphism Date: 27 Dec 1995 20:44:47 GMT Organization: DAIMI, Computer Science Dept. at Aarhus University Lines: 103 Approved: mailtonews@daimi.aau.dk Distribution: world Message-ID: <4bsb7v$c0i@krone.daimi.aau.dk> NNTP-Posting-Host: daimi.daimi.aau.dk Dear "all", Erik Ernst and Wolfgang Ziller: Of course I was aware that BETA supports polymorphism. It seems it is not stated directly with this word. I studied the book carefully in August and September 1993. After this I had to share my time in understanding what environment meant till I read the book Object oriented environment: the MJolner Approach in November and december 1994. Simultaneously I had to deal with the heavy topic Geotechnical mapping (it is heavy even for geologists) and sustainability, design and planning processes, the urbanization of the megacity Sao Paulo, thermodynamics, philosophy and so on. All together. Suddenly I have to dedicate entirely to the implementation of my ecodesign model. And curiously rereading a previous paper presented to the Third International Conference on Systems Integration 1994 held in Sao Paulo, entitled " A design model reasons on the urban ecosystem as an autopoietic entity" I realized I had understood well the notions of the pattern construct and the concurrency level in BETA. However I had not exploited polymorphism, aggregation and so on. Then I have been trying to look for papers with good examples of applications of polymorphism such as Polymorphism and type checking in object-oriented languages Peter Grogono and Anne Bennet. Sigplan notices, vol.24, no.11 On understanding types, data abstraction and polymorphism Luca Cardelli and Peter Wegner ACM Computing Surveys Vol. 17, No.4 December 1985 The Cartesian product algorithm Simple and precise type inference of parametric polymorphism Ole Agesen (of the BETA team) ECOOP'95 - in LNCS The paper after Ole's paper is about Polytoil a new OO language highlight polymorphism. Then I say all this makes me very tired. And if I have patience and love to truth, characteristic of a true scientist for me it is evident computer scientists make very little endeavour to coordinate efforts or to act synergetically. IF Polytoil is wonderful to implement polymorphism like the paper by GRogono and Bennet, which mimics the syntax of Pascal| it orbits around the concept of class which does not present the generalization of the concept of pattern. Anyhow I got an insight and suddenly realized how my model of primary waves in architectural design generation would become simple and elegant if I could use polymorphism. Each subeco-system is the figure of my architectural language. There are as many subeco-system as one might wish, the main ones are activities, environmental comfort (acoustics, lighting and thermal comfort), structural systems, building systems. Each subeco-system with hyphen is characterized by the hypothesis homeostasis, continuity, differentiation and repeatability. Homeostasis would deal with the "comfort zones" for thermal comfort, hearing and seeing (the level of lighting) and so on. In the case of structural system the homeostasis would deal with the bearing capacity of the terrain, the activities would deal with the minimum and maximum measures required for an activity ergonomics, anthropometry and so on. Basically the homeostasis hypothesis will deal with tables, continuity will deal with description of climate and so on and differentiation deals with geometric design in strong interaction with the submodel of the Architectonic sign (or the geometric modeling of the ecodesign model). All these are concurrent processes (in the head of a designer all this occurs simultaneously, Roger Penrose explains through quantum mechanics how this is possible). Then it is very important if I have a superpattern called primary waves, then I could call it as first class variable. My ecodesign works as an autopoietic entity. Hence theoretically I can build from housing groups (a single house in a site is not sustainable and my model does not worry about this case, I would consider this as poor design - the geometric modeling deals with simplicial complexes, discrete groups of the plane, including crystallographic groups, tilings and even fractals, so one may design with well defined separation between private and communitarian spaces) to a sustainable neighborhood, a city, a bioregion, .-..the whole planet. Have you grasped the importance of keeping the ecodesign model intact like a map? Of course suddenly I got lost with the complexity of the task. I am sure to build the model was much more difficult, it was like creating things from void and yet to implement it poses a challenge. I feel I need to grasp the concept of polymorphism in-depth in order to implement it. When I have handled the problem, I will have still many doubts so I would be happy with input from the computer community. A Happy Green New Year to all| Albertina Lourenci Ph.D. student at Faculty of Architecture and Urbanism, University of Sao Paulo Brazil E-mail: Al at vmcisc.cisc.sc.usp.br