Difference between computational design and procedural design

I agree, sort of. Because it’s not black and white. If found out, that the more experience you have with that technology, the less likely it is, that an algorithm can surprise you.

The more you understand how they operate, the more obvious it is how the design is going to end. With that understanding, you are likely to skip automating the form-finding process, because it’s easier to sketch the idea and then execute it with a direct modeling.

Another point is the complexity. With Grasshopper, I have worked most of the time in automotive design. There are so many constraints to a design, that the effort in automating the form-finding is astronomically. Anybody has a deadline in some way. And even if there is a clever way, it makes no sense at all. I have created parametric models of any sort of shapes. Then the design idea changed just slightly and the script was totally useless. There was no easy way in integrating the new constraints and parameters without starting from beginning on. In practice, direct modelling turned out to be the most effective way of working for me… By far!

So one consequence of this is, that if you want to work with Grasshopper efficiently, you only do these things which are worth to automate. That could be a complex generic pattern or some simple reusable parts. That means you design with the help of algorithmic tools, but it’s not designing exclusively algorithmic. That’s per definition closer to ‘automation’ than ‘design’.

1 Like

This conflict emerge historical when the word design was invented.
Originally the word is suppose to come from the latin word “designare” but that word in latin is just really the word “assign” in english, so nothing similar to the current meaning.
Then in the 1800’s germans and english people started using the word to represent an “aestethic concept” or something like that, like a collection of characteristics that make objects beautiful, very similar to the word “style”… but in the early 1920’s the architects stopped been consider a formal engineers and started been called a designer(mainly because a lot of them love to solve problems of beauty than real practical problems) making the word a little more attractive.
Suddenly a bunch of arts&crafts people emerge trying to profit on that shift and they loved to be called “designer” and “designing” to whatever their practice was. Today that is shifting again cause of hyper-functionalism. Design today is strongly dependent on function.

The design word originally was not meant as “process” but as a “state”, similar to “architecture”, architecture is not really a “process” is an “object”(state).

A “design” are just the results of any optimizational process, this include automation, evolution, learning, solving a problem by hand and thinking…etc, even drawing is an optimizational process. Optimizational processes are the ones that have a goal and you optimize towards it, like all error functions, cost functions, minimization, maximization, gradient descent…Every designer has a Goal, it might not be clearly define, but its still a Goal. Having a Goal is the difference between a Designer and an Artist(the modern concept of artist not the roman one). This Goal Orientation is really the origin of function. This is why when you use an algorithm and its goals are not clearly define, the result end-up been useless, they might be beautiful thou. In a sense they are not random but they are also not very functional.

If you ask people what the word design means they really cant define it. Again similar to “architecture”(at least 68 definitions) everyone has their own definition, so “design”(has like 24 definitions) is really a useless word.

When we say the word “designing” we are really applying any process we are capable for transforming and object into an desire result(goal). One of the best examples is nature, nature has designs but there is really no conscious thinking behind it, is just a process.
Concepts like “parametric design” or “procedural design” are really silly because every design is already parametric independent of a computer or not and every design is the result of a process. “Procedural” comes from “Process”.

I personally don’t use the word “design”, cause for me there are better substitutes like “plan”, “specification”,“building” or “product”. I only use it for clients thou :wink:

1 Like

Although in German there is a profession called ‘Designer’ and people generally know what the verb ‘designen’ means (a weird sounding anglicism), the German translation for ‘to design’ is ‘entwerfen’. Which rather is translated as ‘to create (something from imagination)’. Unfortunately ‘computational design’ is translated best as ‘rechnergestütztes Entwerfen’ which is exactly an old translation for drawing design sketches in CAD. A CAD program is nothing else as an application with a lot of procedures and computations…

I think the popularity of these terms is just part of the typical buzzword bingo found in many academic professions, to make a field of research, or the researcher himself, more revolutionary as it/he might is in reality.
Basically the same why the German language dilutes with anglicism, because people want to express their cosmopolitan mindset…

1 Like

Language is several orders of magnitude less complex than the reality it is trying to describe, it simply cannot achieve that much. Even mathematics has been shown to be incomplete with Godel’s theorem. To claim that everything can be expressed with a definition is simply not possible, and the appearance that it can is due to our simplicity. So what one aspires to in such cases is to seek to differentiate it from its fellows and define it in that way, rather than by what it really is. The good thing is that doing so is just as useful, because it is not a matter of classifying but of communicating. So, although conventional design involves parameters and processes (like everything else in life for other side), these terms refer to (or differentiate themselves from the other types of design) to the subgroup that uses them explicitly.

On the other hand, there are two main ways of understanding design, design as a creative process and design as a product. If we use the second one, it is clear what is parametric or procedural or what is not, because to consider it as such this product must be able to be modified by parameters or by algorithm, something that a chair does not allow you to do (or a 3d chair modelled in Rhino either because if you do that you get another product), so not all design is parametric or procedural.

1 Like

I accept the fact that the meaning of the words change over time, but not all words change, the definition of a column today is the same as it was in the times of the greeks(expansion of a definition is not the same as a substitution). Definitions change mainly because we don’t really understand the object we are trying to describe at the time.

Its true that language is limited(I’m not talking about mathematic formalization, cause mathematics cannot really be use to describe the world completely, I think programming languages have a better chance for doin this thou) but formalization is require for a correct teaching of the discipline. Architecture is the best example, if you try to teach yourself architecture you discover there is no one book you can use that gives a general view of the whole field, so students are giving a bunch of incomplete resources and get lost on the field without a clear thread that joins the whole discipline together, programing has the same problem(this is why a lot of people try to learn to program but then they quit). But if you take a book for example of biology, its clearly structured and define an by the end of the book you will have a clear(good enough) understanding(maybe not a complete) of the current state of the discipline.

If you take any book of biology, the first page is gonna have a clear definition of “Biology” as the study of life(even thou “life” is not well define, the definition is close, independent of source the definition is gonna be similar),this allows people to talk in the same level, but if you ask for a definition of “architecture” or “design” the answers are very poor even for people with years of practice.
This makes structuring the whole discipline very difficult and is one the reasons why learning it is so “hard” and starting your own practice is also difficult(where do you even start?). It also brings a more important problem. If the architects and the designers are not even able to define their own discipline how do we expect that other people(clients) understand why our disciplines are necessary. If you ask anyone in the street what a doctor does they will answer very quickly that a doctor cures people, but if you ask wat an architect does they cant give you a clear answer(this depends on the country, there are countries where there is more formalization but in countries like Germany or USA the answer is still lacking).

I really don’t think this is a problem cause by some type of logical impossibility inherent in the structure of the discipline. Part of it is that this fields use a lot of tools and tools are constantly changing(Very Dynamic), they also represent very complex processes and systems(Complexity) and also they are heavily dependent on the thinking processes of humans(Variation or Diversity) and its not till recently that AI has giving certain formalization to it that could be use to structure this disciplines.
Words like “computational-design”, “procedural-design”, “design” and “architecture”, they are useful in so far they are use like names, but that’s it, if you try to used them to learn about the discipline they might end-up giving you more headaches than necessary.

I really don’t think that if a group of people uses a word that immediately makes the word valid, that is why we live surrounded by buzzwords. For me a word is valid because it survives thru time, time has proven its usefulness. I can invent one of this terms in 5s right now,…maybe…Artificial-Potato-Design or ADP for short, Did I just invent a new discipline or a new movement of designers? Not really or at least I hope so. :smile:

About parametric design…The recalculation inside a computer of the parametric 3d model is equivalent to take a bunch material and make another chair with it. The fact that using a computer makes the recalculation faster or cheaper it doesn’t imply that a new whole system of design has been created. One of the best examples of this is the idea of the “production-line” invented by ford, Historically the production line has been around for thousands of years the fact that Ford made it more efficient or faster it does not mean he invented something new that requires another name. The system structure really tells you if something new was invented or not. Procedural design really goes into the boundaries of this problem(but it doesn’t cross it) because, procedural design results normally in very complex geometries that are almost impossible to design by hand, but the fact that they are very difficult to design by hand doesn’t mean they are impossible to design by hand. There are gothic churches all around europe fill with procedural design.

Sorry for this book lololo