As a matter of fact, urban performances nowadays depend not only the city’s physical capital - the endowment of hard infrastructure but also on the intellectual and social capital, namely the availability and quality of knowledge communication and social infrastructure. It ha been shown (Berry e Glaeser 2005) that the most rapid urban growth rates has been achieved in cities where a high share educated labor force is available. Because not all cities are equal successful in investing in human capital, an educated labor force spatially clustering over time. This tendency for cities to diverge in terms of human capital has attracted the attention of research and policy makers. It turns out that some cities, which were in t past better endowed with a skilled labor force, have managed to attract more skilled labor, whereas competing cities failed to do The concept of smart city has been introduced as a strategic device to encompass modern urban production factors in a common framework and to highlight the growing importance of social and environment capital in profiling the competitiveness of cities. Hence, a city can be defined as "smart" when investments in human and social capital and traditional (transport) and modern (ICT) communication infrastructure fuel sustainable economic development and a high quality of life, with a wise management of natural resources, through partipatory action and engagement (Caragliu, Nijkamp e Del Bo 2009).

Following the above-mentioned criteria, smart cities can be identified and ranked along six main axes or dimensions, which conn< with traditional regional and neoclassical theories of urban growth, and development. In particular, the axes are based on theories of regional competitiveness, transport and ICT economics, natural : sources, human and social capital, quality of life, and participati of citizens in the governance of cities (Giffmger, et al. 2007). What differentiate the smart city approach from the past are the willingm and the ability to capture in a single framework many aspects of the city that until now have been addressed separately. The city is thought of as a set of interconnected networks, such as the transportation network the power grid, the buildings network, the network of social relations and so on. The integration of such networks in a coordinated design is the one that makes possible new services that were impossible in the past decade and opens the possibility of progressive transformation of the city.

2.2 A DIFFERENT WAY TO LOOK AT THE CITY

Among the factors that shed new lights on how cities are being interpreted, evaluated, criticized and also planned, there are a series of theories, some well established, other gaining ground from the early 1980s. These theories are, in some ways, questioning some aspect of the current approach to urban and even architectural design. In his recent book "The new science of cities" Michael Batty pointed out how, throughout recorded history, the urban environment has been studied as a place whose form and structure can be represented as models, maps and pictures of locations. The author highlighted how this physical representation has framed our understanding of cities in terms of how we might manipulate urban activities. He claimed in fact, that location encapsulates urban activities, but does not reveal the relations and interactions between populations, which represent the essence of a city. If it is true that by the end of this century most people will be living in cities, is time we change our focus from location to interaction.

If we approach cities by dealing merely with patterns of location, we might miss the point of why cities exist in the firs place. As Michael Batty suggest, we need a new methodology for understanding cities, that is

"[...] Built on concepts of why people come together to trade and exchange commodities and ideas, to realize social contacts, and to procreate; in short, to relate. Interactions, hence networks, are therefore considerably more important to our understanding and planning of cities than are locations." (Batty, The new science of cities 2013, 15).

As Batty suggests, this does not mean disregarding locations, for these are intimately related to interactions and networks. But if we agree that the patterns characterizing urban growth and form emerge as consequences of interactions, flows of energy, and information, then ideas about the shape of cities, about urban sprawl, about the integrity of neighborhood and so on, which have been used in the past in illustrating density and accessibility, follow naturally from this approach.

As a matter of fact, the author reports Chadwick (1971) observation that early in the twentieth centuries, the approach and the ideas about urban planning moved toward more systematic social science approaches, which favored the collision between city studies and systems approach. Systems are defined as organized entities that are composed of elements and their interactions. It follows that hierarchical organization from the bottom up is essential for systems that evolves, and is the way nature and society develop robust and resilient structures. As soon as this notion of hierarchy became fashionable in thinking about cities and their neighborhoods, it was relaxed to reflect the notion that such strict subdivision could only be a simplification. Indeed, a famous paper entitled "A City is Not a Tree" (Alexander 1965) argued that the kind of variety and diversity that was the essence of cities, as articulated by Jacobs (1961), was being destroyed by the implementation of city plans that imposed such a rigid hierarchy through, for example, zoning. This way of thinking did not consider any sense in which systems might change, because when systems theory was first applied to cities, it was widely assumed that articulating the city system in terms of some long-term equilibrium was the appropriate response. Nevertheless, Batty emphasizes, cities might appear in equilibrium because of the huge lag between the speed at which the built environment changes and the way human behaviors change. He observed that the problem of a systems theory of cities is that it tends to underestimate external inputs. In the last fifty years in fact, has been realized that the notion of systems freely adjusting to changed conditions was never valid. If we agree to analyze the urban environment starting from the conception of an open system, existing in a volatile environment, then the idea of a top-down order is no longer appropriate. As the concept of openness become significant and the idea that systems in equilibrium are the exception rather than the rule, the idea that systems grew from the bottom-up gain ground. As Batty states:

"The complexity sciences, which change the focus from highly organized, top-down conceptions to much more organically structured, bottom-up types of systems, grew from these concerns and now provide a framework that is considerably richer and more appropriate for the city systems" (Batty, The new science of cities 2013, 61).

In human systems, decisions about organization and behavior are made at all scales, from the local and routine to the global and strategic, but they tend to be made by individuals and in this sense

are highly decentralized. Articulating systems as structures that grow from the bottom up puts dynamic into account and leads to the notion that patterns and structure at the most local scale, where individuals operate, "emerge" from these actions. Emergence, which involves novelty and innovation, is the watchword of this new view o: systems. It captures the notion that a system such as a city cannot be planned from the top-down but emerge organically.

2.3 The needs and challenges of analytical methods

From the point of view of urban planners and architects the cooccurrence of the above mentioned theories and criteria, along with the increasing requirements of energetic efficiency, economic viability and level of comfort, among other requirements, have raised the bar in terms of the overall quality of their design proposals. In order to match the client’s expectations, being a private, a group of stakeholders or a municipality, the designers increasingly need models and means of analyze their projects on a quantitative level to reduce the risk of failure and test their proposals against the required criteria. As highlighted by Karimi (2012), the challenge of using analytical methods in urban design begin with questions such as what type of analysis should be used, or how they should be applied. There are various analytical tools and models, such as transport models and planning models that have not been developed specifically for urban design, but have been used in the disciplines that are associated with urban design. Recently, with the advancement of computer programs, new techniques of rendering and 3D modeling have emerged that are mainly used in representation of design, but sometimes are also used to analyze specific aspect of the design. Among the most technical development in this field, perhaps the invention of Geographical Information Systems (GIS) has had the most direct influence on analytical approaches in urban planning. The capability of overlaying layers upon layers of geo-referenced data and the ability to analyze these layers quantitatively has turned GIS into a powerful tool in urban planning. The primary problem with the implementation of these analytical techniques into the design process is the lack of an urban theory that could link physical aspects of the urban system with its functional, social and behavioral aspects.

This theoretical shortfall, firstly highlighted by Hillier (1996), ere atesa gap between the analysis of things and how their manipulation in design could impact people.

Karimi also cite Bryan Lawson’s opinion (2005), regarding the act of design, that see it inherently a process. A process, Lawson explains which is normally considered as a continuous action, operation o] series of changes that take place in a continuous manner, seems to be very relevant to any design activity. Furthermore, if we look at design as a process it seems natural that it starts with an initiation phase - a project brief, a request or some sort of undefined needs and ends up with an outcome, as a plan or an object. In other words design is a purposeful process that starts with some sort of objectives, well-defined or ill-defined, and ends up with an outcome that responds to them. From this point of view, it is clear that the design process involves some degree of problem solving or solution making if we seek to respond to some objectives to produce a result through a series of actions, we have to think of different ways of achieving the results and responding to the challenges involved in each approach. It follows that design cannot be an entirely logical or discursive process, and that some form of intuition, creativity and novelty, which are not entirely governed by logical or scientific discourses, can be identified in some parts of the design process. During this process, designer are frequently involved in a ‘conjecture-test operation, which is predominantly based on a cycle of creating design concepts and testing them against certain criteria. Accepting that the design is not an entirely logical process give rise to the question whether the design is an entirely intuitive process, where intuition is considered as a form of knowledge created by instinctive feelings as opposed to deductive knowledge, or it can be informed at any stages by non-intuitive actions, such as reasoning on analysis. It is in fact very difficult to argue that logical thinking cannot play a role in any part of the design process. Accepting the design as a purposeful process of problem solving inevitably leads to conceding that some degree of rational thinking and reasoning has to be applied throughout the process. As Lawson (2005) suggests, a design process needs to reflect on itself and assess whether the output of each stage respond adequately to the objectives of design, even if this reflection appears as an implicit form of reasoning.

Designers with the task to design a project, at architectural scale as at urban scale, need to identify a number of questions that are either given to them directly, or arise from their own understanding of the tasks. Then they need to develop design ideas that would in their parts, or entirely respond to those questions. The important issue in this process is that the solutions have to be somehow evaluated against a series of criteria that are introduced internally or externally. A part of this evaluation takes place during the idea-generation stage, which is a conjecture-test cycle, which leads to an initial option generation and option testing. In a design process, conjectures are normally tested intuitively and the designers come up with their own judgment of whether or not a design conjecture would work. However when the design ideas are shaped, a more rigorous evaluation is needed to determine whether or not the design idea could potentially become the right design solution for the project. The conjecture-test cycle does not treat design as a linear process, but present it instead as a cycle of design generation and design development, where these two main stages are distinct but feed into each other. The design cycle normally starts with an acknowledgement of what is intended to be achieved eventually, and at the end of the design development phase a design output appears, that needs to fulfill at least partially the requirements of the brief.

Following Karimi (2012), what it is important is to understand whether any form of analytical investigation could be applied to any part of the design process, and if so where it should be applied to make a meaningful contribution. The author points to Blakey’s (1850) definition of the term analysis, as the process of dividing a complex entity into its constituent components, study each component in detail and bring them back together to form a better understanding of it. On the other hand, design is inherently a complex issue comprised of different components and facets. In principle the design process can be divided into components to be investigated separately and then be synthesized within the general framework of the design. In this sense analysis is an advantageous method when there is a need to build more rigor in the study of design components and evaluating them against certain criteria: even before generating any ideas, the analysis can provide the designers with the information that they might not be able to obtain intuitively. Furthermore, during the conjecture-test process, the ‘test’ part of the process could be enhanced by an analysis of the conjecture. It is conceivable that only intuition could be applied at this stage, but human intuition is limited in many ways and a pure intuitive test could be inaccurate and biased. Also after the formation of design ideas through a conjecture-test cycle, the design ideas could be tested more systematically using the same analytical techniques that are applied at the beginning of the process. The analysis of the design ideas would determine whether they respond adequately to the objectives of the design and whether they work as intended by designers. By applying analytical methods, a more reliable evaluation of the design ideas is expected, as it is not the mind of one individual that determines whether or not the design ideas would work, but there is a method that could be repeated and applied by others to get the same results.

Following Karimi’s argument, two extra stages can be implemented before and after the idea-generation, or design development phases: in the beginning of the design process a set of analytical investigations, or a baseline analysis, is produced before the generation of any ideas or solutions. The baseline analysis aims to clarify the brief, the context limitations, particularities and other issues that are relevant to design. The design solutions are produced after the digestion of the analytical study, as well as the wider issues (social, economical, political) that exist and are relevant to the design. Once the design options are shaped, analytical tools are used to evaluate them. More than being just a rejection-approval filter, this phase could critically determine what aspects of the design options might not work. During the design development phase, where the design ideas are taken forward, analytical methods could still be used to assess specific aspects of the design. This could be achieved either by the analytical method that have been developed at the earlier stages or methods that are developed specifically to deal with certain aspects of the design.

When dealing with architectural or urban design process, Karimi argues that four important characteristics seem to be more relevant than other, and thus should be met by the analytical method in order to be implemented within the design process. First, the analytical approach to be used in design has to be a spatial one, because both urban and architectural design are about creating and shaping spaces, and therefore the analytical approach chosen should deal directly with this important aspect of the design. Second, the spatial analytical approach should be able to link directly space with people and users, because urban and architectural design are about shaping space for the people, and if we analyze the space in isolation from how it would influence the life of people we just produce an abstract representation of the space. Third, the analytical approach has to be able to deal with different scales, because both urban systems and architectural projects manifest themselves in many scales, each one with different characteristic, but that are in continuous interaction with each other. Finally, a spatial analytic model has to be able to investigate a system as a whole or in its parts, because the parts are explored and perceived differently from each other and the entire system, but on the other hand the whole is made of its parts and it is influenced by these when it grows of transform.

It is here argued that space syntax, a set of theories linking space and society and a set of techniques for analyzing spatial configuration, could provide such a means, along with two other analysis methods, Visibility Graph Analysis and a particular version of Agent Based Modeling. The latter two analysis, that were built upon space syntax theories, serve both to extend its applications, taking account respectively of the visual perception of the environment and the forecasting of pedestrian movement, and also to reinforce its theoretical premises. However, before getting to the central question about what space syntax is and how it might be applied within the design process, a little digression on flows and networks is required, since both themes serve as a background to the argument.