Generative Design Digital Manual (in progress)
Celestino Soddu and Enrica Colabella

This Digital Manual is built for the first time for supporting the managing, step by step, of the Generative Design process.
Following the schedule of lessons, we will be published it in the journal GASATHJ for a more wide use in Generative Design teaching activities.
It will be implemented during the teaching process with different material: examples, theoretical and practical references, dedicated papers, lesson movies etc. used in each lesson and in each teaching exercise.

Why generative Design?
The main aims of a Generative Design process are to identify a possible vision and to construct our performing way for developing ideas in design. There is not a question to find out unexpected and fascinating forms, but to find the possibility to communicate at the best our design ideas.
Operatively, Generative Design help us to construct and develop an idea following a scientific process and not by representing it with only a singular result, with a "solution" as simplification of the complex possibilities in act. Generative Design is how we can plan a transforming process, by following a poetic logic able to bring us from a first sketches to several possible variations as final results. All together these variations are able to communicate our subjective idea, our peculiar vision as designers, open to a possible identifying style.


1st step: identification and definition of aims as design characters.
We cannot run progressive paths if we haven't clear aims. These aims are not only practical aims, configured only for identifying functions.
The main generative aim is to reach possible results able to represent our subjective vision for gaining complexity.
For this a good tool is to define some characters of our performing aims as characters that we love to rediscover in our design results.
Operatively we can identify these characters with three adjectives.
Number three is very useful for controlling them in easy way in our first steps. In fact only one is too much axiomatic, two are many times in opposition or in alternatives and more than three are too much for controlling them in our first operative procedure.
One example: smart, baroc, exciting…..
Our chosen adjectives might be also in their significance, in contrast one each other’s.
They stay together in a performing way for defining a not linear system, open to complexity

2nd step: constructing own reference imaginary world
Imaginary references: as just said, interesting examples of design, of objects, of music pieces, of poetry, of architectures, of cities, of movies, of faces, of animals, and so on.
When we chose a reference, soon after the choice we must associate to the reference one (or more) of the characters that we identified. That is: we must have a clear design vision able to drive us when considering them.
Our 3 adjectives identify in a deep way our first peculiar points of view as attractors of the real world. The character expressed with our 3 adjectives are a very good key for interpreting the world of images, poems, music, movies, drawings, postcards etc. that we choose for subjectively discovering in them our defined characters.
In this way we are performing a selected imaginary world as a representation of our references for gaining the world of our incoming generative design results.
So we start by performing aims as characters of an imaginary world of references.
This is a selection made for association in resonance between attribution/word ( character ) and images, sounds, word etc ( reference world ).
A double moment in a discontinuous process.
Suggestion: collect several references for each character, paying attention to each reference difference from all others, covering a peculiar aspect of our own impressions. The process is in discovering the character significance of our attributions.
The reason why we identify several references for each adjective is in the plurality of possible poetic logic for each adjective/attribution. By performing this associative process, we are collecting in our mind a plurality aspect of the same significance. So we become able to focus a plural useful formalization of the same adjective for performing our generative process.
With this process, we are delineating in deep our aims.
Two ways to approach these references:
1 - Considering the whole reference
2 - Considering only a part of it.
Whole reference / macro level: we consider the global geometry, the whole structure and organization, performing a complex system that we can focus by the interpreting our reference.
1st example: looking at Milano we could consider its peculiar global geometry as a system able to connect everything in subsequent circles and, in the same time with transversal straight lines belonging to different main locations. This is not only a peculiar static form, this is a "Generative Geometry", able to perform a great open numbers of connections, all belonging in congruous way to our generative Milano system. (see C.Soddu, paper at 17th GA conference)
2nd example: looking at the Bach fugues, we could consider the structure of a global system created by subsequent repetitions, following well identified rules, where each repetition is a variation of the same theme. This structure of organization is well identified as a "Generative Geometry" too.
Details - Fragments of the references / micro level: Operatively we identify one (or more) peculiar detail(s) of our reference as a particular discovered aspect of our characters at micro scale, by itself or in its connections.
For doing that, we must identify the role of each detail. If it belongs to "how" this event ends, "how" it fold itself, "how" it is divided in several parts, and so on, by performing geometric rules of connections. In practice, we can associate each reference to:
a) one of the characters that we identified,
b) a peculiar aspect that belongs to a particular "design act" that we can interpret as performed in the design process that constructed this results,
c) a peculiar "Generative Geometry" able to represent our interpretation of the geometric structure of our reference in reasonable way.
1st example: we choose as reference the Chrysler tower in NY. The detail is the cup of the tower. The "design act" is " how it ends". The adjective is "exciting”. Now we made a logical interpretation of this detail and we "discover" that the tower ends by repeating the final geometry (in this case an arc) several times progressively reducing the dimension and putting the reduced arc over the previous one. This logic performs all the sides of the tower. Following this peculiar (and subjective) approach we can identify a possible rule able to perform the "how to end" aspect, as a rule applicable to other and different occasions. We can do that also if the "arc" is configured as a different geometric event or also if the event is not geometric but a word, a note, a decoration, and so on. This rule is similar to a fractal geometry, where each event is repeated reducing it in a scalar way and generating it in a definite process until it will be so little at minimum scale, but always following the same geometric logic, as in nature. This rule is a rule of Generative Geometry. We can associate this rule to our character (exciting) and to a peculiar "design act": how ending one of the event of our projects (ending a communication, a dress, an architecture, a pen, and so on).
We haven't found a form but a rule able to manage possible progressive transformations fitting a peculiar character defined as our aim.
It's clear that; the logical interpretation, expressed as a poetic logic is a subjective and plausible hypothesis related to a possible design path and it is not the objective reconstruction of the real design process, that can happens also following other rules. Each design result of good quality is performed in a complex way, where the possibility of discovering hidden rules is always open. Our site of interest is our logical interpretation because we are looking for the possibility to construct our own design process related to our own design vision. Performing these expressions of poetic logic we are learning how we can gain the complexity of our time, starting from our singular vision.
These "design acts" are the results of decision moments that we norm ally perform during our design activity: how to end an object, a communication, a piece of music, a dress, how to divide it in two or several parts, how to fold it or to change the tonality, or to move from one material to another, how to make a skin, how to open a hole, and so on. Each one can identify these possible "design acts" following own identity as designer. It involved the micro scale too of our reference with the same logic of Generative Geometry.

3rd step. Constructing our Generative Tools.
We are constructing our Generative Tools when we identify the "Generative Geometry" discovered by the geometric logics (rules, geometric progressive transformations and exceptions) that we like to read in our references. These Generative Tools are able to involve the whole project process.
We can call them "Logics of Transformation" as our logical hypothesis of a progressive transformation (folding, dividing and so on) able to perform a possible form characterised by one of our adjectives.
"Generative Geometries" and "Logics of transformation" are like operative algorithms applicable to the generative design process. We can write real algorithms too, if we are able to represent with algorithms each transforming process. In any case by drawings or by algorithms, they are our generative design tools that we can use in all our design activity.

4th step. Choosing (or accepting) a design occasion.
The step of starting with a new design project is a very important moment.
In our teaching experience, each student can identify the theme of his/her design experiment. This decision is not important because it belong to an activity that is not proper of designers, but proper of clients. After this occasion, soon we will have the theme to develop. We are working as designer.
The only thing that we must follow when we choose the design theme is that we need to find out in the theme a possible complex structure with many unexpressed questions to answer. This because each designer needs to effort questions: each question can generate a design answer. More are the occasions where we have to gain an answer with our design act more the quality of results can increase in complexity.

5th step. Identifying a catalyst for interpreting it as a possible topological structure
Design is transforming process with a progressive increasing of complexity. The first act of a design process is not a transformation because there in not anything existing. So we need to make complex our empty sheet.
We can do that using a sketch that we will consider as a catalyst or using an image as catalyst.
How can we choose or make an appropriate catalyst?
It must not represent a "formal" idea, because forms are interchangeable and they cannot be identified in this too early moment. They might gain only a simplified solution of our generative process, as an easy result. This becomes very dangerous, stopping our process. In this way is very hard to find out another solution.
Better, the catalyst could be a reference for the organic structure of our project; it could help us to identify a first topological structure.
So we will use this catalyst for transforming it into a paradigm, able to make a first identification of the parts of our project and their relations. No forms must be identified because if a form will early emerge, it will be too dangerous for the progressive design process. This early choice can kill the creative process.
example: Our theme is a poster. If we like organic structures, we can use, as catalyst, an image of an animal, i.e. an elephant. This is only a catalyst and, after using it for starting our project, we have to forget it. We don't use the elephant as a form! Our logical interpretation of this catalyst could be: there are 4 parts in my poster, as the legs of the elephant but there is a free event (like the proboscis) focused on the "last news" that we intend to communicate with the poster.
I can define that he four parts are "baroc", the free event is "smart". A final event, where I will communicate a peculiar detail, will be "exciting" and will be at the end of my poster, it will be little like the code of the elephant. Now I will forget the elephant and I will have a first "topological" structure of my project: the organization paradigm.

6th step. Toward the paradigm
In few steps, it is possible to organize our moving from a catalyst to the paradigm.
Firstly, we must make a logical interpretation of our catalyst, identifying some events (that we interpret as belonging to some parts of the project as a performing connection) and their direct or indirect relationships.
Secondly, we can associate to the entire object and to each event and each relationship a peculiar character, choosing it from the three characters that we previously identified.
Representing together events, their characters and their relationships we will have a first paradigm of organization that we can directly use for going ahead in our design process.
Obviously this is only a "first" paradigm because the object that we are going to design needs an increasing of complexity that we can obtain repeating the same process (catalyst and characterized paradigm) for each event inside the first paradigm. And so on.

7th step. Performing first formalization at macro and micro scale.
Looking at the just defined paradigm, we can make a logical interpretation of this structure and we can apply to this topological structure a geometrical structure. Practically we will apply to this paradigm one of the "Generative Geometries" that we identified in one of our references belonging to the characters. We will apply it at macro scale, involving the entire project. If we like to apply more than one character, we can apply more Generative Geometries. This work starts in transforming the previous structure, so we can apply these geometries one after the other, transforming the previous transformation.
Looking at each event, we will identify the "design acts" necessary to perform the event itself, their functions and characters. Following that, we will apply the "transforming rules" that we identified through our logical interpretation of our references at micro scale.
At the end, we will have a first formalization of our project. More we can gain a sequence of variations applying different "Generative Geometries" to the whole project and different "Logics of Transformations" to each event.

8th step. Increasing Complexity
Looking at our project, we will have the need to develop it toward the complexity, in a way that the results will satisfy the complex needs of our times.
We can do that increasing the complexity of the paradigm by developing new paradigms for each event and subsequent formalization of these new events.

9th step. Changing paradigm
However, its' possible that the adopted paradigm doesn't work well: it is not able to manage the increasing complexity, the increasing functions and requests of the project. For this reason, we can change the paradigm, but we are not losing our already made work.
Being a topological structure, it is possible to run in changing the paradigm, not transforming it, but jumping from a paradigm to a new one (starting from catalyst and so on).
The interesting matter is that we don't lose our previous work. Each event can be generated again using the same transforming rules and the same characters that we used in the previous paradigm

10th step. Final results
The final result is not an unique result. Each possible result is one of the representation of our idea, but cannot fully represent our idea, never our vision. It's necessary to generate a set of variations, each one different but all together representing the same idea.
The variations are like different individuals of the same species. Different olive trees, with completely different form but all recognizable as olive threes. All together, they communicate the character of the species, as our variations will communicate the uniqueness of our design vision.


download images:
http://www.generativism.com/img/GDManual_schema.jpg(external link)
http://www.generativism.com/img/grafo_xweb2.jpg(external link)

Online Users

8 online users


Google Search