Nikolaos Argyros
“Toward a theory of architecture machines”

Toward a theory of architecture machines, the text is based on the visionary and ambitious thoughts of Nicholas Negroponte for the next generation of ar- chitecture machines that can work not only by doing but also by learning about what they are doing.

In other words he believes that the architecture ma- chines need to be enough intelligent to self learn and adapt in different tasks and solve different pro- blems.

Negroponte in his text specify five particular subas- semblies that would characterise an architecture machine: – 1)a heuristic mechanism that follows strategies which drastically limit the search for a so- lution – 2) a rote apparatus with the ability of rote learning and association of this knowledge with a response, in case when a situation is repeatedly encountered, a rote mechanism can retain the cir- cumstances for usage when similar situations are next encountered – 3) a conditioning device an en- forcement device that handles all the non exceptio- nal information (each robot would develop its own conditioned reflexes) – 4) a reward selector that se- lects from any action that which the creator inte- ract based on satisfaction or disappointment of the creator but also on evaluation and observation of the goals as well as results – 5) finally a forgetting convenience because unlearning is important as le- arning, obsolescence occurs over time or through irrelevance and because information can assume less significance ever time and eventually disappe- ar. These five items are just pieces, the entire body will be a group of mechanisms interconnected with the designer and a parent machine.

For the parent machine Negroponte imagine a ma- chine like a main server with powerful processors and extensive memory that is able to communicate with all the others architecture machines. Moreo- ver states three properties that are necessary for an architecture machine to give a so-called design solution: – 1) an event – 2) a manifestation – 3) a representation.

In such a manner, architecture machines could ac- quire information beyond that which they are given and therefore would have the potential to challenge and to question. Currently, a great deal of concern and reattach effort is placed on the machine-gene- ration of form from a given statement of criteria but

according to Negroponte for the eyes of an archi- tecture machine, the problem is the opposite; given a form, generate the criteria… learn from the criteria and someday generate new forms.

We need to recognise that the book “The archi- tecture machines” and the next one “Soft archi- tecture machines” was written buy Negroponte in 1969 and 1975. In our days we can see that lot of the ideas that Negroponte wrote in these books became reality, and this architecture machines our now our computers that everyday architects and not only use for design and communication. I totally agree with Negroponte that architecture machines and technology can improve tha quality not only in architecture but generally of life.

In conclusion i will like to share the first quote in the introduction of The Architecture Machines book:

“ … so much corn, so much cloth, so much everything, that things will be practically without pri- ce. There will be no poverty. All work will be done by living machines. Everybody will be free from worry and liberated from the degradation of labor. Every- body will live only to perfect himself.”

Karel Capek, Rossum’s Universal Robots