Conference PaperPDF Available

From Black Box to Generative System

Authors:
  • Carnegie Mellon University / University of Arkansas
Conference Paper

From Black Box to Generative System

Abstract and Figures

Under the umbrella of the digital turn, novel computational workflows and distinct aesthetical principles are becoming an integral part of architectural education. Nonetheless, in current educational settings, there is not much scope for a deep understanding nor the development of custom computational design methods beyond standard toolkits. To fill this gap, we outline an educational framework for the development of new generative systems. The proposed framework combines canonical techniques for generative systems from different fields with recent advancements in Artificial Intelligence. It comprises eight schemas: unstructured constructive, structured constructive, variational, improvement, discrete simulation, continuous simulation, generative learning and behavioral learning. Each schema consists of a different formulation of design space and navigation, providing a knowledge base and a common language for design. Their adoption in design education can potentially expand the boundaries of design both within the agendas of the authorial design, nurtured in the studios, or even expand the boundaries of the profession to address future demands from society.
Content may be subject to copyright.
BLACK BOX: Ar cula ng Architecture’s Core in the Post-Digital Era 1


 
     


     
 



     




 
             

            

 
Reyner Banham consistently cri cized architectural design
for its mysterious mode of opera on and super cial incor-
pora on of new technologies. In early works, he claimed that
the avant-garde only adopted an aesthe c of the industrial
revolu on, while preserving the academic composi on as
an implicit design method1. Later, he depicted the modus
operandi of architecture as a black box2, “recognized by its
output, though unknown in its content”3. In this polemic anal-
ogy, architecture has nothing to do with the quality of the
built environment; it is an exercise of an arcane, privileged
and unspoken aesthe c code, inculcated in the studios and
glorifi ed in the dra smanship of architectural drawing.
In this black box hypothesis, computer-aided design and the
binary logic of the computers would represent a “probably
fatal blow”4 to the mys que of Architecture. Coincidentally,
its original publica on in 1990 was followed by a digital turn
in architecture5. This turn invoked a paradigm shi in design
educa on and prac ce, marked by the emergence of compu-
ta onal design models and a new conceptual vocabulary6. In
the 1990s, architects started to manipulate digital geometry
with the new CAD, modeling and anima on so ware, chal-
lenging the limits of tradi onal architectural representa on.
In the following decade, programming was rediscovered by
designers with scrip ng languages and the rise of graph-
based parametric editors integrated in CAD systems. The use
of algorithms in design required the externaliza on of the
instruc ons for design genera on, moving the designer away
from the autographic domain of architectural drawing and
fracturing the black box.
However, the incorpora on of computa onal techniques in
digital prac ces did not promote the type of ra onality in
design and educa on aimed for by Banham. As in the past,
the claims of technological shi were saturated by aesthe -
cal disputes. Associated with digital fabrica on and new
advancements in architectural geometry, the computa onal
methods supported a new repertoire of non-standard forms
that were developed in research pavilions and, eventually,
applied to the design of the building surfaces, components,
and envelope. Prac cal architects intertwined design meth-
ods and aesthe c for the digital architecture.7 Even specifi c
technologies, such as anima on or parametric modeling,
were assimilated as a blend of method and design content.8
In this sense, the dra smanship associated with architectural
drawing was replaced by digital varia ons.
This recurrence of techno-ideological feuds in architecture
is related to the socializa on of the aesthe cal code and for-
ma on of the professional behavior, in par cular with the
prominence of the “design crit in the architectural school
studio”9. The studio culture is reminiscent of the Beaux Arts,
where students were educated by the guidance of one or
more experienced architects, who cri que their solu ons 10.
In this se ng, design success is measured by sa sfying the
expecta ons of the circle of educated “patrons”. Despite
the aesthe c and technical diff erences, the fi rst digital edu-
ca onal experiences, such as the paperless studios, also
revolved around the explora on of an aesthe c agenda of
experienced architects11.
Notwithstanding, the studio, being a well-established and
tested educa onal se ng, in face of new technologies estab-
lishes a problema c rela on between design method and the
expecta ons of the cri c. For instance, it is usual to adopt
pre-defi ned computa onal workfl ows and plug-ins as a plat-
form to explore a certain design agenda. This se ng comes
with the cost of immediacy and bias, as there is no  me to
inves gate the poten al of computa on for design beyond
the agenda and toolbox provided by the studio.12
 
PEDRO VELOSO, DR. RAMESH KRISHNAMURTI
Carnegie Mellon University
BLACK BOX: Ar cula ng Architecture ’s Core in the Post-Digital Era2 3
Table 1: Computa onal schemas and techniques for GS and space
planning.
Table 2: Recent concepts associated with GS and computa onal design.
This rela on is even more cri cal in face of design automa-
on and Ar cial Intelligence (AI), which have long been
encroaching on the territory of architecture. AI acquired new
momentum with the recent wave of deep neural networks,
which succeeded in automa ng ac vi es such as pain ng
style transfer, playing go, medical imaging, speech recogni-
on, face recogni on and synthesis. Not surprisingly, Mario
Carpo’s response to this momentum was the announcement
of a new digital turn for architecture 13. This  me, the turn is
based on the computa onal logic of search and the availability
of big data14, resul ng in a movement from the non-standard
forms and smooth surfaces to a focus on voxeliza on and the
genera on of form with search and simula on.
The complex rela on between the terms computa on and dig-
ital is crucial for the future of design educa on. Computa on
strictly refers to the use of algorithms or models to perform
opera ons on symbolic representa ons. For example, com-
puta onal design methods use these algorithms and models
to represent, analyze and synthesize design alterna ves. In
contrast, the term digital has been used fl uidly to describe
a certain cultural condi on or state of being related to the
advent of diff erent informa on technologies. Design benefi ts
from historical and philosophical interpreta ons of the infl u-
ence of these technologies on our society and environment.
However, pursuing legi mate expressions of the digital with
restricted computa onal schemas prevents the access to a
more general knowledge on computa onal logic and struc-
tural changes in design prac ce.

While computa on logic can address diff erent aspects of
architecture, in this paper we focus on the synthesis of forms
and spa al pa erns. A proper knowledge base of compu-
ta onal synthesis can be developed based on genera ve
systems (GS) − systems that can generate design alterna ves
automa cally for a certain problem. While current digital
studios and programming textbooks focus on parametric,
NURBS and mesh modeling, GS comprehends a wider vari-
ety of techniques from di erent elds and domains. Design
educa on would benefi t from a systema c and historical
understanding of the diff erent computa onal schemas avail-
able for GS.
The defi ni on and systema za on of GS have been devel-
oped in computa onal design books15, courses, ar cles
and research conferences in Computer-aided Architectural
Design16 (CAAD). A brief review is given below.
Christopher Alexander’s understanding of “genera ng
system” is as a system composed of a kit of parts and com-
binatory rules that can generate many varia ons17. In his
structuralist view, all natural and ar cial phenomena are
themselves systems generated by a specifi c set of interact-
ing forces or rules.18 However, while natural systems are
adaptable to the interac on of forces, the built environment
requires ar cial methods to capture the exis ng forces and
promote a global behavior based on their equilibrium. In
opposi on to conven onal design, which addresses this task
by intui on, Alexander categorized three methods to gen-
erate form based on forces
19
: (1) numerical methods, which
use linear op miza on of design variables to look for the
best form; (2) analog methods, which use a physical model
to represent the forces of the system and look for a stable
confi gura on; (3) rela onal methods, which use diagrams to
represent the diff erent forces of a system and fuse them to
generate the proper form.
William Mitchell wrote one of the rst overviews of GS in the
eld of CAAD20. His understanding of GS is as systems that
can produce a variety of poten al solu ons for a problem.
He proposed three general categories for GS that are simi-
lar to Alexander’s methods: (1) analogue GS are composed
of analogue elements that enable mechanical opera ons to
change the state of the system; (2) iconic GS are systems that
use movies, models, drawings, and geometric opera ons to
generate solu ons, (3) symbolic GS use symbols and compu-
ta onal data-structures to represent a solu on and rely on
arithme c and logical opera ons to change it.
While Alexander’s work focused on GS based on rela onal/
iconic GS21, Mitchell’s overview focused on symbolic GS to
produce spa al solu ons that meets certain specifi ed criteria
automa cally – i.e., space planning. In contrast to Alexander’s
structuralist approach, Mitchell interprets symbolic GS with
classical AI concepts. Each GS operates with discrete steps.
Related designs that are visited in the computa onal process
are codifi ed in a directed graph called state-ac on graph,
which structures the space of all possible designs and avail-
able opera ons for naviga on.22 The goal of GS is to nav igate
the set of solu ons in this state-ac on graph looking for a
subset of solu ons that sa sfy the design goals.
Mitchell23 and other researchers, for example Henrion24 and
Ligge 25, classifi ed and described the solu on procedures of
symbolic GS for space planning. Most of their categories are
inscribed in two computa onal schemas: search and op mi-
z a  o n 26, which were the main topics in Simon’s famous text
on a science of design27. See Table 1.28 29 30
Since the 1990s, not only metaheuris cs but also other com-
puta onal concepts, such as cellular automata and swarm
algorithms became part of architectural experimenta on,
which were explored in computa onal design books31.
Addi onally, architects also incorporated anima on, NURBS,
mesh and parametric modelers in their design toolbox. While
most experimenta ons departed from classical AI, eventu-
ally, techniques, such as search and dissec on, are revisited32.
In Table 2, we organize authors who try to capture the
diff erent computa onal and mathema cal concepts associ-
ated with design. We combine (1) the pedagogical categories
for the teaching of GS proposed by Fischer and Herr33, (2) the
concepts iden ed by Kolarevic to characterize the digital
morphogenesis34, (3) the digital models of design beyond
CAD systems described by Oxman35 and (4) the mathema -
cal concepts iden ed by Burry and Burry36 in contemporary
design. We organize them according to the following sche-
mas: complex geometry and topology, packing and ling,
diagramming and data visualiza on, anima on, op miza on
and performance, parametrics, rule-based systems and com-
plexity. 37 38
Recent classifi ca ons have moved away from the computa-
onal logic to focus on their source or the applica on domain,
indica ng the incorpora on of GS as a common prac ce for
architectural design. Oxman and Oxman39 provide six general
models of form genera on: (1) mathema cal: which exploits
mathema cal formulae for genera ve procedures; (2) tec-
tonic, which employs tectonic pa erns for form genera on;
(3) material: which uses tectonic and assembly pa erns,
BLACK BOX: Ar cula ng Architecture ’s Core in the Post-Digital Era4 5
state of the system is specifi ed, naviga on is defi ned by the
selec on and applica on of rules or procedures, manually or
automa cally, to change the state of the system in discrete
steps. The design space is neither explicit nor defi nitely fi nite.
See example in Figure 1, top.

This schema addresses problems structured as a graph or
a tree, which are composed of states (nodes) and available
ac ons (edges). The states are the nodes and the possible
construc ons or ac ons are the edges. Naviga on combi-
natorically explores valid states in the design space, which
contains “the set of all states reachable from the ini al state
by any given sequence of ac ons”43. In a search, the solu-
on is a sequence of ac ons from an ini al state at the root
to a desired state. In a traversal, the solu on is a systema c
way to access some or all the states. In a sampling, a solu on
is generated by following a single path defi ned by a certain
probability or policy. Diff erent algorithms organize the search
in diff erent ways by using strategies to order the explora-
on, to evaluate the costs of the decisions or to check the
consistency of the diff erent constraints of the problem. See
example in Figure 1, middle and bo om.

This schema defi nes geometric en es and rela ons using
such as folding, braiding, kni ng and weaving, to generate
form; (4) natural or neo-biological, which employs biologi-
cal principles to generate form; (5) fabrica onal, which uses
exis ng pa erns of fabrica on for design genera on; and (6)
performa ve, which models physical data of the context as
the input for a genera ve process that sa sfy certain objec-
ves. This type of categoriza on refl ects the ubiquity of GS
in diff erent architectural approaches, which is also reinforced
by categoriza on of solu on procedures in specifi c domains,
such as digital fabrica on with parametric modeling40.
  
The challenge in categorizing GS is that their underlying tech-
niques originate in diff erent fi elds, they overlap, or they apply
to mul ple domains of design. While pioneer classifi ca ons
focused on the computa onal logic of the GS, recent ones
address speci c technologies, design inspira on or applica-
on domain in architecture. These recent approaches are
very produc ve for educa on and bring design to the center
of a en on. However, they limit the scope of GS to the status
quo and design instan a on – i.e., to a set of solu ons with
its own tested work ows and tools.
In the opposite direc on, our framework recovers the idea
of computa on not as a tool for design, but as an alterna ve
logic of design. It does not focus on the representa on of the
design elements but on the high-level concepts that mediate
design and computa onal genera on: design spaces and nav-
iga onal strategies. Broadly speaking, design space refers to
the space of possible alterna ves that can be generated given
a certain formula on of the problem. Design naviga on refers
to the opera ons and control strategies that are available to
navigate between these alterna ves. In tradi onal prac ces,
designers use heuris cs to formulate and reformulate the
problem, building expressive design spaces and naviga onal
strategies to explore solu on candidates.41 In GS, the design
spaces and naviga on strategies are a consequence of its for-
mula on with specifi c algorithms and models.
By focusing on computa onal logic with more abstract
categories, this framework comprehends both the exis ng
solu on procedures and the poten al incorpora on of recent
advancements in AI. The eight schemas of our proposed
framework are in Table 3.42
 
The schema is based on the existence of a discrete repre-
senta on, which can be geometrical or even alphanumerical,
that is sequen ally constructed by the applica on of rules
or procedures, which might require a certain shape or a cer-
tain rela on between shapes to be matched. Once the ini al
Table 3: A GS framework based on design space and naviga on
Figure 1: Construc ve schema. Top: unstructured construc ve schema
of the qGrowth grammar. Middle: qGrowth structured in a tree with a
greedy best-fi rst search algorithm that orders the fron er based on the
Euclidean dis tance to a target. Bo om: ex ample of qGrowth generated by
sampling. It follows the target behind a wall (ellipse).
BLACK BOX: Ar cula ng Architecture ’s Core in the Post-Digital Era6 7
space is defi ned by all possible confi gura ons of a simula on,
given its ini al se ngs and the agents’ policies
The next two categories employ Machine Learning – a mul -
disciplinary fi eld concerned with programs that automa cally
improve with experience – to search for the best hypothesis
about a given data, behavior or knowledge.44 They open the
possibility of learning GS from data.

To solve many learning tasks, researchers employ genera-
ve models, which a er having been exposed to a dataset,
“explicitly or implicitly model the distribu on of inputs as well
as outputs
45
. More than simply learning how to perform a
certain task, they model how the data has been generated.
Thus, a genera ve model enables the sampling of synthesized
data based on the data distribu on that it learned for the
task. The design space is the space and domain of the input
vector, which is translated to a result by the learned mapping.
See example in Figure 4.

This schema is based on learning the ac ons of agents to cus-
tomize a GS. Among other approaches, it includes adap ve
agents46 and reinforcement learning47. The system learns a
policy (the probability of choosing the available ac ons in a
state) by combining simula on with evolu onary algorithms
or by exploring and exploi ng ac ons to maximize a reward
in an environment.
In prac ce, the schemas presented above can be combined
to form hybrid GS. By combining the diff erent categories,
designers can develop a custom GS with a proper design
space and naviga on for their problems.

The post-digital should not be understood as an emerging
era, but as a cri cal a tude towards the fascina on or denial
of the digital. New formula ons of a digital zeitgeist48 or re-
mys ca ons of the architectural representa on49 are both
short-sighted acts. In contrast, if we accept the ubiquity of the
digital, we can use computa on to explore complex spa al
pa erns and interac ons, addressing previously ungraspable
aspects of society and environment.
parameters and constraints through explicit func ons
defi ned by the designer. The resul ng design space is explicit
in the parameter space. It comprehends all the geometrical
and numerical varia ons resul ng from all possible combina-
on of the parameter values, which can be done manually or
algorithmically, using stochas c or determinis c procedures.

This schema is based on the transforma on of a state by a
search for alterna ves that perform be er according to
a metric. The design space is the parameter space, where
parameter values describe all possible design solu ons. The
func on space contains the possible results of a single or
many evalua ve func ons applied to a solu on. The fi tness
space is a one-dimensional space that translates the results of
the func ons to a single measurement of success. The com-
bina on of these spaces results in a representa on called a
tness landscape that contains all the fi tness value for all the
solu ons in the parameter space. Improvement procedures
can be solved by the applica on of calculii, op miza on
strategies or metaheuris cs. See hybrid example in Figure 2.

This schema is based on a discrete representa on, such as a
grid or a graph. In combina on, the states of the local units
characterize the global state. Once an ini al global state is
defi ned, the rules or procedures aff ect some elements of the
collec on, based on local states and neighborhoods, whilst
preserving the global characteris cs of the representa on.
The design space comprehends all possible varia ons of the
global representa on, considering the ini al state and the set
of rules applied over  me. The examples of discrete simula-
ons are generally associated with mathema cal models of
urban or natural phenomena. See example in Figure 3.

This schema is based on a collec on of agents that sense, act
and interact. Each agent interweaves local evalua on of its
goals and ac ons on the environment − the space, fi xed ele-
ments and the other agents – by the applica on of local rules
or procedures. Agents can have diff ering levels of autonomy.
While simpler systems use basic refl ex agents, more com-
plex systems have a program to evaluate and act. The design
Figure 2: Hybr id (2 + 4). Op miza on of fl oorplans with Gene c algorithm
an d K DTr ee .50 Figure 3: Discrete simula on. Rules and example of Beady Ring.51
BLACK BOX: Ar cula ng Architecture ’s Core in the Post-Digital Era8 9

We would like to express our gra tude to the Brazilian
Na onal Council for Scien c and Technological Development
(CNPq) for gran ng Pedro Veloso a PhD scholarship (grant
201374/2014-5)

1 Reyner Banha m, Theory and Desig n in the First Machine A ge (The MIT
Press, 1980).
2 A black box is a s ealed device origin ally depicted in an ele ctrical engi-
neer proble m. The observer must de duce the behavior of a sea led box by applying
diff eren t distur ban ces to the in put te rminal s and by o bserving th e result s.
3 Reyner Banh am, “A Black Box: The Secret Pr ofession of Architec ture,”
in A Cri c Wr ites: Sel ected Ess ays by Reyn er Banha m (Berkel ey: Unive rsi ty of
Californi a Press, 1999), 293.
4 Ibid., 298 .
5 Mario Carp o, The Digital Turn in Archit ecture 1992-2012 (Chichester:
John Wile y & Sons, 201 3).
6 Rivka Oxman, “ Theory and Desi gn in the First Digital A ge,” Design
Studies 27, no. 3 (2006): 229 –265; Rivka Oxman, “Digital A rchitecture as a
Challeng e for Design Pedagogy: Th eory, Knowledge, Model s and Medium,”
Design Stud ies 29, no. 2 (2008): 99–120.
7 Joseph R osa, Nex t Genera  on Ar chitect ure: Fold s, Blobs , and Boxe s
(New York: Ri zzoli, 20 03); Pete r Zellner, Hybri d Space: G ener a ve For m and
Digital Arc hitecture (New York: Rizzoli , 1999); Branko Kolarevic, “ Digital
Morphogenesis,” in Architecture in the Digital Age: Design and Manufacturing
(London: Spo on Press, 2003), 12–28.
8 Greg Lynn, Anima te Form (New York: Princeton Arch itectural Press,
1999); Patrik Schumacher, “Parametricism as Style - Parametricist Manifesto,”
2008, h ps://www.patrikschumacher.com/Texts/Parametricism%20as%20Style.
htm.
Our own approach in this paper situates the logic and his-
tory of computa on as a way of designing and not as a tool
to express a digital condi on. In the same way that an algo-
rithm course in computer science might refer to canonical
algorithms as a base for new developments – without claims
for an algorithmic era –, canonical computa onal schemas
can provide a common background to support research on
GS. By exploring and hybridizing the diff erent schemas, the
designer can navigate in the complex territory of computa-
on to look for novel design logics.
We introduced this framework in a mini course in our
ins tu on, addressing the fi rst ve schemas. In the end of
the course, the students had to choose a problem in their
domain (game, building, landscape, urban design, etc.) and
develop a GS to produce alterna ves of solu ons. We s ll
plan to extend it to a full course where we can discuss all
the schemas, incorpora ng the recent advancements in AI.
For a future objec ve, we intend to refi ne and formalize the
categories and their rela ons into a book.
Figure 4 Genera ve learning: Deepcloud. Top: architec ture of auto -
encoder tr ained on a dataset of chairs. Bo om: chair series generated by
naviga on on the resul ng feature space.52
9 Banham, “A Bla ck Box: The Secret Profes sion of Architecture,” 294 .
1 0 Joan Drap er, “The Ec ole D es Beaux- Arts an d the Arch itectural
Prof ession i n the Un ited States: Th e Case of Joh n Galen Ho ward ,” in Archi tect:
Cha pter s in the His tor y of the Prof essi on, ed. Sp iro Kosto f (Ber keley: Un iver sity of
Californi a Press, 2000), 211; John F. Hab erson, The Study of Arch itectural Design
(New York: Norto n, 2008), 182.
11 Stan Al len, “T he Pap erless Stud ios in Cont ext,” in Wh en Is th e Digita l
in Architec ture? (Montreal: CCA and Sternb erg Press, 2013), 383–4 04; Bernard
Tschu mi, “The M aking of a Gene ra on: Ho w the Paper less Stu dio s Came About,”
in Whe n Is the Dig ital in Arc hitect ure? (Mont real : CCA a nd Sternb erg Pres s, 201 3),
405 –19.
12 For example, parametric design can be associated with NURBS and
mesh modeli ng to explore complex g eometry. If perform ance is part of the
agenda, plug-ins for op miza on or physics simula on are added. Whenever
some addi  onal cus tomiza on is nece ssary, th e student s try to “ha ck” the avai l-
abl e tool box to sa sfi c e the desig n task.
13 Mario Carpo, The Second Digital Turn: Design Beyond Intelligence
(Cambridge: T he MIT Press, 2017).
14 Carpo’s percep on of technological change is legi mate, but his defi ni-
on of a digital zeitgeist bas ed on the duality of “don’t sort; search” (Ibid., 23.) is
fragile. In hi s interpr eta on , the di screte l ogi c of comput ers and it s capaci ty to
operate with b ig data – which he refers to as p osthuman complexi ty (Ibid., 48.) –
is opposed to th e formulaic and compr essive logic of classi cal science. However,
many o f the techn iques th at Carpo us es to repre sent this logi c of compu ta on in
design (heuris c search, op miza on, cellular automata and simula on associ-
ated with FE M) are not di rec tly re lated to bi g data. Addi onally, c omputa on is
not op posed to c omp ress ion. The d iffi cult y of curren t (or even fu ture) comp uters
to dea l with larg e and con nuous st ate sp aces jus  fy th e use of heur is cs and
func on approximators, such as deep neural net works, instead of brute-force
search or tabu lar methods. Carpo a lso opposes Calculu s to the discrete logic of
computa on and its capacit y to search (Ibid., 65–70.). However, not only Calculu s
can b e disc re zed for geo metry proc essi ng (see dis crete diff eren a l geometr y),
but it is als o the base fo r backpr opaga on, a main tech nique to tr ain neura l
networks for big data analysis.
15 Gabri ela Celani and P edro Velo so, “The Inter sec o n of Theor y and
Technology: Computa onal Concepts Applied to Architectural Design since Late
198 0s - a Literatur e Review,” i n Fron  ers of Sci ence and Tec hnolog y: Water
Availability, Automa on and Sensor Technologies, Digital Fabrica on (7th
Brazilian-German Conference, Campinas, 2017).
16 Gabrie la Celani and Pedro Veloso, “C AAD Conferences: A Bri ef History,”
Elec tronic Proc eedings o f 16th Inter na ona l Confer ence CAAD Fut ures 2 015. Sao
Paulo, July 8 -10, 2015., CAA D Futures, 2015.
17 Christopher Alexander, “Systems Genera ng Systems,” Architectural
Design 38 , no. 12 (1968): 605–10.
18 Georg Vrach lio s, “ How Form C ame abou t from Soc iety: Ch ristop her
Alexander o r About Architectu re as a Form of Culture and Str ucture,” in
Stru ctural ism Relo aded: Rul e-Bas ed Design i n Archit ecture a nd Urb anism
(Stu  gar t: Axel Me nges , 2011), 61–6 8.
1 9 Christo pher Alexander, “From a Set of Fo rces to a Form,” in The Man-
Mad e Obj ect , ed. Gy orgy Kepe s (New York: Ge orge Bra ziller, 19 66), 9 6–107.
2 0 Will iam J. Mitchel l, Compu ter-Aid ed Archi tectura l Design (New Yor k:
Mason Char ter Pub, 1977).
21 Alex and er’s own wor k emphas ized the use o f rela o nal meth ods fo r the
synt hesis of fo rm, from the fus ion of fun c ona l diagram s to the deve lopmen t of
a language of s emi-autonomous r ule sets for design. Fo r a general understand -
ing of this trans i on, see: C hristop her Alex ander, Notes on th e Synthes is of Fo rm
(Harvard University Press, 1964); Christopher Alexander, A Pa ern Language:
Towns, Buildings, Construc on (Oxford University Press, 1977).
22 Mitchell, Computer-Aided Architectural Design, 46–48.
23 Ibid., 425–74.
24 Max He nrion, “Au toma c S pace- Planning: A Pos tmorte m?,” in
Conf erence Proceedi ngs (Ar cial Int elligen ce and Pa ern Reco gni on in
Computer Aided Design, New York: North-Holland, 1978), 175–91.
25 Robin S. Lig ge , “Autom ated Fac ili es La yout: Pas t, Presen t and
Future,” Automa on in Cons truc on 9, no. 2 (2000): 197–215.
26 Whil e op miza on i s itself a form of sear ch, it is di ere n ated fro m
cla ssical se arch met hods bec ause it doe s not expl ore a syste ma c pat h to a
solu  on, but sp ecifi c no des/stat es, based on the ir fi tnes s. This cr eates som e
ambigui  es. For mo re detail s about cl assica l search a nd op miz a on, see ch ap-
ters 3 and 4 of Russ el Stu art J. and Pete r Nor vig, Ar c ial Intel ligenc e: A Modern
Approa ch, 3r d ed. (Uppe r Saddle Rive r: Pre n c e Hal l, 2010 ).
27 “Th e Sci ence of De sign : Crea  ng the Ar cial,” in T he Scien ces of the
Ar cial, 3r d ed. (Camb ridge: Th e MIT Press, 199 6).
28 Word crea ted by Henrion to desc ribe cer tain for mula o ns of spac e plan-
ning that comb ine cons traint sa  sfac o n and goal op miza on .
2 9 Generate a nd test is a general search p aradigm based on the co mbina-
on o f a rand om gener ator and a te ster, wh ich fi lte rs the sa sfying s olu on s. It
can b e con side red a proto typical op miza o n where the teste r is an objec  ve
fun c o n that returns a B oolea n valu e or it can be s tructu red as a cla ssical s earc h,
if the g enerator rel ies on pre vio us gen era on s of solu ons. Mitchel l descri be it
as a problem -solving method for sp ace-planning a nd Simon use it as a general
strategy to dec ompose the design prob lem.
3 0 Mitchell de scribes shape gramm ars as a technique to formul ate the
problem (a synt ac c formul a on) and not as a sol u on proc edure. Th is dis n c on
is subtle, as the sh ape grammars can be d eveloped with shape inter preters or
with search strategies. However, shape grammars are canonical rule-based GS,
so we su bverted his or iginal clas sifi ca on and add ed them to th is table as a GS.
31 John Fraze r, An Evo lu ona ry Ar chitec ture, Arc hitect ural Ass oci a on
(Lon don, 1995); Paul Coat es and Rob ert T hum, Gen era ve Modelling (Un iver sity
of Eas t London , 1995) , h p://roar.ue l.ac.uk /948/; Pa ul Co ates , Programmin g
Architecture (London: Routledge, 2010); Kostas Terzidis, Alg orithmic Architecture
(Oxford: Th e Architectural Press , 2006).
32 Kosta s Terzi dis, Perm uta on Desi gn: Build ings, Tex ts, and Co ntexts
(New York: Routled ge, 2014).
33 Thoma s Fische r and C hris  ane M. Her r, “Teac hing Gen era ve D esign,”
in Pro ceedin gs of the 4th Conf eren ce on Gene ra ve Ar t (Genera  ve Art,
Politechnico di Milano University, 2001).
34 Kolarevic, “Digital Morphogenesis.”
35 Oxman, “T heory and Design i n the First Digital Age.”
36 Jane Bu rry and Mark Bu rry, The New Ma them a cs of Arc hitecture
(Th ames & Huds on Londr es, 2010).
37 Fischer and Herr combine diff erent techniques under the category
“Algorithmi c genera on a nd gro wth”. We opt ed to divid e them acc ording to our
general categories.
38 Kolarevic mixes diff erent algorithms (gene c algorithms and L-systems)
und er the categor y gene c s. We opted t o divide them ac cording to ou r general
categories .
3 9 “Fro m Comp osi on to Gen era on ,” in Th eories of the Di gital in
Architec ture (New York: Routledge, 2014), 55– 61.
4 0 Lisa Iwamoto, Digital Fabrica on: Architectural and Material
Techniques (New York: Pri nceton Architectur al Press, 2009); Nick D unn, Digital
Fabrica on in Architecture (Laurence King Publishing, 2012).
41 Peter G . Rowe, Design Thinking (C ambridge: The MIT Press , 1987).
42 T his is an in i al a emp t to defi ne a framewor k, so we mixe d general
meth ods, mod els and al gorith ms fro m diff ere nt areas in our ex amples .
43 Russe l and Norv ig, Nor vig Pe ter, Ar cial Intelli gence, 67.
44 Tom M. Mitchell, Machi ne Learning (Boston: McG raw-Hill, 1997), 2–15.
45 Chris topher M . Bishop, P a ern Reco gni on and Ma chine Le arning (N ew
York: Springer-Verlag, 20 06), 43.
46 John H. Ho lland, H idden Order : How Adap ta on Builds Comp lexity
(New York: Basic Books, 1995); John H. Holland, Signals and Boundaries: Building
Blo cks for Co mplex Ad ap ve Systems (Cam bridge: The MI T Press, 20 12).
47 Richard S . Su on a nd Andrew G . Barto, Rein forcem ent Le arning: An
Intr oduc on (Dra ), 2nd ed. (Cam bri dge: The MI T Press, 2018), h p://incomple -
teideas.net/book/.
48 Carpo, The Second Digital Turn; Mario Carpo, “The Post- Digital Will
Be Eve n Mor e Dig ital , Says Mar io Carpo,” Metr opolis , July 5, 201 8, h ps://ww w.
metropolismag.com/ideas/post-digital-will-be-more-digital/.
49 Sam Jacob, “A rchite ctu re Enters th e Age of Pos t-Di gital Dr awing,”
Metropolis, March 21, 2017, h ps://www.metropolismag.com/architecture/
architecture-enters- age-post-digital- drawing/.
5 0 Insp ired by Katj a Knecht and Rei nhard Kön ig, “Gen era ng Floor P lan
Layo uts with Kd Trees and Evo lu ona ry Al gorith ms,” in Proc eeding s of the 13th
Genera ve Art Co nfer ence (Gen era ve A rt, Mila n: Domus Ar genia Pub lisher,
2010), 238–253, h p://www.genera vear t.com/.
51 Bill Hillier and Julienne Hanson, The Social Logic of Space (Cambridge:
Cambrid ge Unive rsi ty Press, 1984), 55–6 4.
52 Ardavan Bidgoli and Pedro Veloso, “DeepCloud. The Applica on of
a Data -Driv en, Genera v e Model in D esign,” in Rec alibra  on: On Impr ecision
and Infi delit y. Proceedings of 38th ACAD IA Conference, ed. Phill ip Anzalone,
Marcella Del Signore, and Andrew J. Wit (ACADI A, Mexico Cit y: Universidad
Iberoamericana, 2018), 176, 180.
ResearchGate has not been able to resolve any citations for this publication.
Conference Paper
Full-text available
This paper analyzes the evolution and trends of international conferences addressing Computer-Aided Architectural Design over the past five decades. It starts with the Design Methods Movement conferences, when the possible contributions of computers to Architectural Design started being discussed. It shows how new interests along the history of CAAD development motivated the appearance of conferences on specific topics, and it ends identifying a recent interest towards architectural practice and the software development. With this work we expect to contribute toan "archaeology" of CAAD studies.
Article
This paper reviews the history of automated facility layout, focusing particularly on a set of techniques which optimize a single objective function. Applications of algorithms to a variety of space allocation problems are presented and evaluated. Guidelines for future implementations of commercial systems are suggested.
Article
The paper introduces and discusses current developments in architectural discourse, design theory, digital design models and techniques and their relations to design pedagogy. The evolution of design knowledge in architectural theory and praxis is explicated and its implications for required changes in design education are presented. The theoretical influence of architectural concepts is presented through historical references in digital architecture. This structure of design concepts is proposed as a medium of design education. An experimental design studio on: ‘Design as research: the exploration of digital architectural concepts’ is presented as a pedagogical framework for educating the digital architect and a series of research and design programs carried out in an experimental design studio demonstrates this framework.
Article
This paper examines methods in which digital media is employed not as a representational tool for visualization but as a generative tool for the derivation of form and its transformation - the digital morphogenesis. It explores the possibilities for the ?finding of form?, which the emergence of various digitally based generative techniques seem to bring about. It surveys the digital generative processes - the computational architectures - based on concepts such as topological space, isomorphic surfaces, kinematics and dynamics, keyshape animation, parametric design, and genetic algorithms.
Article
Digital design and its growing impact on design and production practices have resulted in the need for a re-examination of current design theories and methodologies in order to explain and guide future research and development. The present research postulates the requirements for a conceptual framework and theoretical basis of digital design; reviews the recent theoretical and historical background; and defines a generic schema of design characteristics through which the paradigmatic classes of digital design are formulated. The implication of this research for the formulation of ‘digital design thinking’ is presented and discussed.
Theory and Design in the First Machine Age
  • Reyner Banham
Reyner Banham, Theory and Design in the First Machine Age (The MIT Press, 1980).
The Digital Turn in Architecture
  • Mario Carpo
Mario Carpo, The Digital Turn in Architecture 1992-2012 (Chichester: John Wiley & Sons, 2013).
Parametricism as Style -Parametricist Manifesto
  • Patrik Schumacher
Patrik Schumacher, "Parametricism as Style -Parametricist Manifesto," 2008, htt ps://www.patrikschumacher.com/Texts/Parametricism%20as%20Style. htm. 15