Skip to navigation – Site map

HomeThemed issues24Mind-walking through Instagram: e...

Mind-walking through Instagram: extracting the essences of Jakarta’s street vending

Anak Agung Ayu Suci Warakanyaka


Despite its economic and social significance, street vending is still seen as a culturally backwards and faulty development unbecoming for Jakarta. This has led to the development of urban strategies that attempt to eliminate its presence from the city’s streets. In response to the issue, through this paper, I posit street vending as a form of interior urbanism; a creative micro-scale intervention born from precarious situations that temporarily transform the street into a habitable space. And thus, its assimilation into the city’s formal urban framework is integral in creating a more liveable and socially equitable urban environment. Hence, a thorough understanding of street vending spatial practices is needed. To compensate for a severe lack of official data and direct access to the field, this paper proposed using mind-walking to reveal the ingredients of Jakarta’s street vending interior urbanism through Instagram. Inductive coding and drawing, interpretative tools that are less intrusive, are utilised within the analytical process to provide a comprehensive reading and extract the necessary components from the Instagram datasets of three different areas in Jakarta. This experiment successfully recognises crucial ingredients that characterise Jakarta’s street vending activities, including six types of street vending practice; the typology of interior urbanism. Despite its inherent downside of only covering the spatial features and yet touching on the socio-political aspects, mind-walking can establish and set out the ingredients that make up the social phenomenon, bringing the research a step forward to formulating a comprehensive and Jakarta-centric urban street framework.

Top of page

Full text

Introduction: street vending in Jakarta

1Like many other cities, street vending also plays a significant role in Jakarta. It provides job opportunities amid high unemployment, low education, and non-existent adequate social services (Wilson, 2010; Simone, 2015; Khairunnisa, 2016). Street vendors also provide cheaper alternatives to daily necessities like clothes and food, especially for those living in poorly conditioned housing that does not offer appropriate cooking spaces (Arciniegas, 2021)—this includes almost 50% of the citizens, which is roughly around 5 million people, who live in the city’s many informal settlements or kampung (Alzamil, 2018). However, unlike the popular belief, street vending is not only essential for the poor but also for the rest of the citizens—especially the young ones where it facilitates an alternative option to enjoy the city’s many outdoor public places as opposed to the malls, which for last forty years become Jakarta’s default pseudo-public spaces, especially for the middle and upper classes. In addition to these, street vending also has been proven to have economic significance in view of the fact that it made Indonesia stay afloat during two major financial crises that hit the country in 1998 and 2008 (Pattiradjawane, Schnepf-Orth and Stoetzer, 2013; Gibbings, 2016), most likely by providing employment, as well as generating revenue from fines and taxes.

2However, despite everything, not only is it not getting well-deserved support, street vending is not yet legal in the country. The 2004 Republic of Indonesia Acts No. 38 stated that the street should only be used as a space for movements; any objects or activities that would disturb the movements in any street component—for example, street vending—should be illegal (Figure 1). Parallel with this act, city-wise in Jakarta, there are at least three laws that condemn the existence of street vendors, with 2007 Jakarta bylaw No. 8, per my personal observation, as the most popular to be referred to—by the press or by the authorities through the printed banners they hung as a warning in the various corner of crowded streets across the city. The law stated that it is illegal to sell goods in the streets, to become a street vendor, and to buy something from a street vendor, except in areas designated by the governor. The law was established during Sutiyoso’s reign, Jakarta’s governor from 1997 to 2007, with a vision to transform Jakarta into a modern city (Wilson, 2010). To do that, he seemingly adopted the binary view of formal-informal, a view shared by the likes of De Soto (2000), that sees street vending as culturally backward and faulty development that is not befitting the idea of a modern city. This view was also shared by some of his successors, especially during the leadership of Basuki Cahaya Purnama or Ahok, in which he made it very clear that street vending has no place in the city by conducting many surprise raids and forceful evictions to scare the vendors away (Sulistywati, 2016).

Figure . Street Diagram issued along with 2004 Republic of Indonesia Acts No. 38 shows the street and its components that should be free from any objects and activities aside from traffic and pedestrian movements. Redraw by Author.

Figure . Street Diagram issued along with 2004 Republic of Indonesia Acts No. 38 shows the street and its components that should be free from any objects and activities aside from traffic and pedestrian movements. Redraw by Author.

3Raid and eviction were not the only solutions the provincial government came up with to eliminate street vendors. In 2010, the government established a new strategy, which I call localisation­­. According to the 2010 Governor Regulation No. 33, localisation would group street vendors based on specific criteria and place them in specially-built locations equipped with proper vending infrastructure. Another strategy was also attempted in 2013, which I call relocation, where the provincial government would relocate street vendors within several established markets across the city (Taylor and Song, 2016). I noticed that there is a shared motive between these two strategies. Both aimed to transform street vendors into proper entrepreneurs, changing the nature of their trade to fit the formal sector. The places where they would be localised or relocated to were carefully selected in the hope that by inhabiting the same spaces with other formal traders, street vendors could be persuaded to assume a new identity and act like what was expected. Simply put, these two strategies can be seen as elimination attempts, somewhat subtler than raids.

4As might be expected, these strategies never really succeeded. Despite the numerous attempts of eviction, localisation and relocation, the number of vendors working on the streets is just kept growing throughout the years—in 2012, it was estimated that there were around 300.000 street vendors in the city, while in 2019, it was suspected that the number was increased to about 600.000 (Wilson, 2010; Rachmawan, Fathy and Luthfi, 2021). Furthermore, it was also often reported in the media that, usually just weeks or months after being localised or relocated to new locations, many street vendors returned to their usual spot in the street (Herman, 2013; Ng, 2022). According to a study conducted by Taylor and Song (2016), there are several reasons why vendors are reluctant to comply with these strategies. The most apparent reasons were the high rent prices and the fact that these government-approved places were often far removed from the crowd. The latter significantly lowered the chances of the vendors gaining customers and thus affected their daily earnings. Hence, it was not that surprising that currently—at least per my field investigation in 2022—some of these government-approved vending locations, especially those specifically built ones, were left unoccupied.

5The failure of these strategies seems to generate from the fact that they were made solely to maintain some sense of order without caring for the intended benefit. Based on how the vendors respond to the strategies, it is also clear that they were formulated without a thorough study, although, admittedly, the lack of research is not that startling. The country is known for its lack of reliable data when it comes to street vending and any other informal activities (Baker and World Bank, 2012), which might have something to do with the non-existence of regular surveys and the fact that street vendor as a type of profession has yet to be included in Indonesia’s labour force statistics (Wilson, 2010). In turn, this reluctance on the government’s part to acknowledge their presence and significance caused street vendors to develop mistrust in revealing their identities. This shows especially in the past few years where under Anies Baswedan, Jakarta’s governor from 2017 to 2022, the provincial government have been more lenient about street vending activities and opened a self-registration channel through their official website ( However, when I accessed the database in 2022, only 303 street vendors were willing to register on this platform—0,05% of the total street vendors in the city.

6This exploration shows the contradictory natures of street vending activity in Jakarta; it serves as a crucial economic sector that impacts the livelihood of many while being criminalised by the governments. This duality posits street vending as both a much-needed survival tactic and a direct challenge to formal urban governance. Therefore, exploring Jakarta's street vending can be seen as a microcosm of the broader debate surrounding urban informality and governance in many growing cities worldwide, especially those within the global south where the issue is most persistent (Mörtenböck and Mooshammer, 2015; Simone, 2015; Graaff and Ha, 2020). Ultimately, this paradoxical nature of Jakarta's street vending raises two questions that will be addressed in this paper. How can we formulate more nuanced policy approaches that balance regulatory measures with the vendors' livelihoods? And how do we do that without formal or official data?

Street vending as interior urbanism

7To be able to answer the first question, I argue that, first, we need to reject the binary view that places the informal beneath the formal. This is not a ground-breaking proposition, for many experts, including Anaya Roy (2005, 2009) and Ann Varley (2013), have similar arguments. The rationale is that by placing these two ideas on the same level, we acknowledge the critical role of informality as part of urbanism just as much as its formal counterpart. At the same time, we also need to see urbanism as an inclusive concept—as the many practices, visions and interventions that make up urban life (Cupers, 2020). We need to shift the idea that urbanism is not a process that exclusively belongs to professionals and formal institutions but is also formed by the contribution made by amateurs, the ordinary citizens. Only then can we begin to see street vending and other informal activities not as things that fall outside of what is normal but as part of the norms that shape Jakarta as a city.

8Secondly, in addition to becoming a part of urbanism, I propose we also need to see street vending as part of interior urbanism. As an emerging concept, interior urbanism has been defined and redefined constantly for the past two decades and is often used simultaneously with urban interior (Attiwill, 2011; Hinkel, 2011; Rice, 2016). Regardless, I found two common themes that bind most of the concept’s various iterations. One, it defines practices that allow inhabitation in existing urban contexts, and two, informality—the reliance on tactic rather than strategy (Certeau, 1984)—is understood as part of the practices’ many approaches. Hence, I see interior urbanism as a micro-scale practice responding to the existing urban environment and temporarily transforming it into a habitable space.

9I believe that street vending fits into this description of interior urbanism; it can briefly transform the street from an infrastructure that only allowed movement into a space that could facilitate various domestic or indoor situations such as cooking, eating, socialising and trading. For that, street vending provides inhabitation and adds the feeling of intimacy and familiarity in the usually unsympathetic setting. Not only that, based on the interview conducted by Kompas in 2019 with Jakarta’s pedestrians, street vendors are also known to create a vibrant environment that generates a sense of safety to walk on the streets at night (Rosalina, 2019)—a quality fits Jane Jacobs’ eye upon the street idea for a thriving city neighbourhood (Jacobs, 1992). Therefore, situating street vending as part of interior urbanism provides a new emphasis on the phenomenon. Street vending is not only an amateur contribution to the city but also a creative intervention born from a precarious situation that presence is integral in creating a liveable city. Hence, its assimilation as it is—meaning in its current form—into the formal urban framework, I believe, would produce not only more socially equitable urban strategies for the vendors and the poor but also produce habitable urban environments for all Jakarta’s inhabitants.

10I have yet to find an example of the use of interior urbanism framing as the cornerstone in formulating an urban street vending framework; however, several cities are already acknowledging the crucial role of the activity, which subsequently leads to its legalisation and integration into their urban governance that a lot similar with the interior urbanism view towards street vending. Even though the activity is still branded as illegal in many parts of the world, for the past decade, more and more cities have jumped on the bandwagon to address the need to legalise street vending. This might have something to do with the growing international sentiment towards the activity, triggered by the current economic climate that has affected not only undeveloped and developing countries of the Global South but also those where advanced capitalism is practised (Graaff and Ha, 2020). Especially in the aftermath of the COVID-19 pandemic and the threat of recessions currently looming over us, I believe that now, more than ever, is the perfect time to acknowledge street vending as the alternative mode of urbanism that allows income generation during the period of uncertainty and insecurity.

11As a start, I propose we look at the existing urban frameworks established by several cities worldwide where street vending is legal, and displacement is not involved as part of the strategies. The cities are New York, Los Angeles, and Kolkata. If we look at the matrix I made based on the cities’ publicly accessible official documents, we can see that each city has distinct regulations that are based on the careful understanding of the local vendors’ activities. None of these regulations proposed significant changes that would alter the informal characteristics of the local vendors—instead, they embrace the transient and fleeting aspects of street vending activities and create rules to ensure that these aspects could meet the health and safety standards that would protect both the vendors, the customers and the rest of the streets’ users. Here we can see the example of attempts that make the authorities and vendors meet in the middle—to compromise rather than enforce.

Table . The urban street vending regulations matrix in New York, Los Angeles and Kolkata. Source: (The Kolkata Gazette Extraordinary, 2018; Bureau of Street Services, no date; the City of New York, no date; Kolkata Municipal Corporation, no date; The Street Vendor Project, no date)

Table . The urban street vending regulations matrix in New York, Los Angeles and Kolkata. Source: (The Kolkata Gazette Extraordinary, 2018; Bureau of Street Services, no date; the City of New York, no date; Kolkata Municipal Corporation, no date; The Street Vendor Project, no date)

12The deep understanding of who, why and how street vendors do what they do that are clearly represented in the matrix is what Jakarta lacks. Without this information, the city cannot formulate appropriate street vending strategies. If I were to compare the development of a framework to a cooking process, I would say that the city does not even know the ingredients they have or need, let alone be ready to develop a recipe. Nevertheless, the matrix provided a clue about the key components that all three cities shared, including vending categories, units, zones and spatial rules. However, the matrix also shows that the ingredients that make up every component are different in every city. This suggests that every city has its own interior urbanism characteristics when it comes to street vending. The ingredients were picked and compiled based on an extensive reading of these unique characteristics through a bottom-up approach instead of a top-down approach. Only then can we develop approachable and sympathetic urban street vending frameworks that befit a particular city.

13However, it is worth noting that incorporating street vending into formal urban governance in the cities explored above is not as smooth as the matrix shows. In NYC, for example, the government struggled to raise the cap—according to Devlin (2021), the cap has remained unchanged for 40 years since 1983, with only 3000 street vending permits issued to accommodate what is predicted to be 10.000 street vendors in the city. The political risk of offending real estate owners and BID made many local politicians avoid working with this issue altogether. There is also a case that some vendors' categories were privileged over others, creating unfair competition among the vendors (Duneier, 2001). In LA, the legislation that acknowledges street vending is still new and has not yet been evaluated. After a lengthy ban since 1930, LA was finally legalise street vending in 2020. According to Cupers (2020), the reason for the ban was that the vendors, who are primarily poor immigrants, were thought to not fit with the city's images, which are highly motorised and have exclusive suburban images. Clearly, sociopolitical and economic aspects become the other ingredients that must be addressed in formulating an urban street vending framework. It requires the framework to be continuously contested, both spatial and socially, to ensure it would provide for the street vendors' right to inhabit the street. Thus, the first step to developing a good recipe for Jakarta-centric urban street vending frameworks is to identify these ingredients. The question is how to collect this data to extract the necessary ingredients without official support? And how do we ensure that the ingredients were extracted based on a thorough understanding through means that cannot be tampered with by assumptions and biases?

Mind-walking: Instagram, clues, inductive coding and drawing

14I will begin this section by stating that establishing appropriate street vending frameworks is perfectly doable with the right instruments and proper infrastructure. However, in this case other than the problems already said, which include the vendors’ reluctance to reveal their identity, my own study about Jakarta’s street vendors began in 2020, at the height of the COVID-19 pandemic lockdown. At the time, I was living in London, and because of the flight restriction and the fact that Jakarta was on a very strict quarantine measure, it was impossible to conduct the research in person. Interestingly, during this period, I saw several news reports about how Jakarta’s street vendors seemed to find a way to go around the restriction rules to work as usual on the street, albeit with slight adjustments to avoid the authorities (Fachriansyah, 2020; Leo, 2020). In addition, the press also reported how some people across the country were forced to try their luck as street vendors when the pandemic impacted their employment (Ardanis, 2022). These situations, which show how resilient and significant the role of street vending in Jakarta, prompted me to carry on with the research by putting together an approach that allows me to collect and analyse data about Jakarta’s street vending activities remotely.

15I used mind-walking, an approach introduced by anthropologist Tim Ingold that can be perceived as the remote version of walking—a situated practice used to understand the world from the grounds (Ingold, 2010b, 2010a). Walking requires the researcher to engage with the physical environment, which then the engagement would trigger follow-up actions. The actions taken are crucial because they reflect the researcher’s understanding of the environment. It is one of those learning-by-doing approaches that make us understand things through experience. And as the remote version of this approach, mind-walking engages with a medium that triggers active imagination, like text, pictures, and art objects, instead of dealing with the physical environment. We interact with the medium by observing, reading, and touching—and then we act out by writing, drawing, performing and composing our initial comprehension of the phenomenon. Just like its embodied counterpart, mind-walking relies heavily on the bi-sequential process of engagement and action—which, in its production of knowledge, makes it quite indifferent to the researcher’s pre-conceived belief or established theories. In theory, it is the ideal approach to remotely gathering and processing appropriate data about Jakarta’s street vending activities from the ground, especially in its capacity as a form of interior urbanism; to capture the intimate urban aspects that domesticate the existing urban environment.

16As for the medium to be mind-walked, I chose to work with Instagram because, one, it is the second most popular social media platform in Indonesia after Facebook (Nurhayati-Wolff, 2022), and two, while both platforms have privacy settings, Instagram provides more publicly available data to be extracted and analysed. The other reason is that Instagram offers glimpses of ordinary citizens' everyday lives through visual and written manifestations created, curated and posted by the citizens themselves. The pre-dominantly visual content is essential to assess the spatiality of Jakarta's street vending, thus providing a more focused dataset for this research. The experiment focused on personal profiles rather than institutional or companies and only chose those without the verified badge to avoid using data heavily tampered with by social, political or commercial agendas. In mining the data while at the same time establishing the data contexts, I utilised Instagram’s geotag feature and time stamps; only those that were taken or posted in Tanah Abang, Sudirman and Kota Tua for the last two years (2019-2021) were considered. These three locations were chosen based on my quantitative reading of the news published by nine Indonesian media companies for the last five years, up to 2021, showing Tanah Abang, Sudirman and Kota Tua as the hotspots of Jakarta’s street vending activities. The three locations have different urban characteristics: Tanah Abang is a commercial district where the biggest textile market complex in Southeast Asia is located; Sudirman is the city's business district that houses many skyscrapers owned by multinational companies; and Kota Tua, or the Old Town, is the city's historic centre, where it houses the remains of Dutch occupation in the city. Choosing these three different contexts was also done deliberately to see if the data in a specific area would reveal distinct interior urbanism characteristics. The temporal aspect of the data, however, was chosen simply because I needed the most up-to-date information about street vending activities to extract the right ingredients and formulate an appropriate recipe.

17In utilising geotag and timestamp features to collect the relevant Instagram posts, I did the opposite of the big data approaches like text-mining and image processing (Huang et al., 2021); instead, I utilised a small data approach to ensure the data collected have all the necessary information. The decision was mainly caused by the fact that it was pretty challenging to determine a set of parameters that would locate and target relevant Instagram posts through computer-aided tools if street vending activities in Jakarta have irregular and arbitrary physical forms that I was not yet aware of. Moreover, relying on hashtags is also out of the question—based on my experiment, it was not unusual for people to put random hashtags unrelated to the activities visually presented on the posts. Hence, I opted to go with the analogue sorting process to avoid overlooking crucial information. Here I am utilising the idea of clues which refer to hints used to identify an atypical or unfamiliar situation in a particular context (McCullough, 1945; Emans and Fisher, 1967). The atypicality we are trying to recognise here is street vending activity, which, according to the Indonesian government’s dystopian concept of a street (Figure 1), is considered abnormal. So in the context of mining the correct data, the Instagram posts should contain clues that indicate any atypical situation that suggests the existence of informal vending activity, for example, the misuse of the street for purposes other than movement and the presence of atypical tools and objects on the street, to be considered appropriate.

18To normalise the data further, every relevant Instagram post was screenshotted and saved as an image in png format. The reason for this is to have control over the data, to move them from the relatively dynamic environment of Instagram, where the users can still delete and edit the posts, to a more static and secure domain. And to ensure all essential components are included, the screenshot captures all parts of the visual and textual aspects of the posts, including the written descriptions, the geotag information, timestamps and comments. If there were multiple Instagram posts about the same situations or people, both street vendors and customers, only one was chosen to be screenshotted to avoid unnecessary redundancy. Throughout this sorting and normalising process, 1.089 Instagram posts are collected and ready to be read.

19Since the primary approach is mind-walking, in which the generation of knowledge should be based on the information gathered on the 'ground' rather than pre-conceived knowledge, even after the tedious sorting process, I still had to ensure that my personal and professional biases would not tamper my engagement with the data. It is easy for me, a former Greater Jakarta citizen of 26 years from a middle-class background, to have a negative preconception about street vending activity. Not to mention that within Indonesia's architecture and urban discourse, the discipline I have been involved professionally for the last 15 years, street vending is still often seen as a disruptive urban element to be eliminated (Dimas, 2008; Yatmo, 2008). Therefore, I used inductive coding to help me interact with the data. It is an approach developed within the discipline of grounded theory (Corbin and Strauss, 1990; Gioia, Corley and Hamilton, 2013) that would help me produce codes solely based on what I found in the data rather than assign pre-conceived meanings to them. Using inductive coding as a tool for semantic analysis of Instagram posts is not revolutionary. It has been used, for example, by Kylie Budge in constructing the meaning of museums across Australia from the visitors’ digital traces on Instagram (Budge and Burness, 2018; Budge, 2020). Learning from Budge’s research, inductive coding could keep the study relatively objective since the understanding and meaning of the phenomenon are purely constructed by the data, the evidence at hand.

20On some level, inductive coding seems to check all the mind-walking boxes where knowledge can be produced purely based on the engagement and actions taken towards the data. However, as I experimented with this approach at the initial research stage, I realised that it only provided a surface-level understanding of the street vending activities—it extracts the ingredients from the data but only touches upon the most apparent quality of the data. Thus, to compensate for this, I proposed the use of two techniques of drawing, tracing and diagram, to be used in conjunction with inductive coding. As an interpretative tool that is less intrusive, drawing would complement inductive coding, particularly in the geometrical understanding of the phenomenon, without jeopardising the need to be objective. Together, these tools would provide a comprehensive reading of street vending activities and extract the necessary components to formulate urban frameworks from Instagram. To achieve that, I would divide the mind-walking process into three steps: open coding + tracing, axial coding + diagram, and lastly, selective coding.

Open coding + tracing

21Like grounded theory’ inductive coding, mine also began with open coding. This is a stage where the data, the Instagram post, was going to be broken down into bits to identify the clues within it (Corbin and Strauss, 1990, 2007). Since the clues had been roughly identified during the data sorting process, this stage focuses on placing labels into the clues using words or short sentences that best reflect the clues’ disposition. By labelling them, the clues are transformed into codes. Furthermore, to avoid redundancy in assigning the labels, constant comparison between one data and the other is a must—and based on my experience, this was quite difficult to be done digitally. Thus, to make the process easier and allow me to engage more efficiently and carefully, I printed out all of the screenshots of the Instagram post and placed them on the walls. Moreover, the printed screenshots were organised according to their location to allow local engagement and comparison with the data. In assigning labels, I used removable post-it notes that allowed me to physically manifested the codes through writing and placed them directly on the printed screenshots (Figure 2).

Figure . Pinned-up printed screenshots of Instagram posts on the walls with post-it notes that label the clues

Figure . Pinned-up printed screenshots of Instagram posts on the walls with post-it notes that label the clues

22I could identify around 6-8 codes per data through this process. However, the codes generated were quite generic—they only represented the most identifiable aspects of the data. I believe this is due to the ability of the naked eye observation that can only do so much without the help of any other tool. However, this does not mean the approach is a futile one. On the contrary, It produced data samplings, which narrowed down more than a thousand raw data to only 105 with a unique set of codes to be investigated further.

23Thus, to remedy the lack of details in open coding processes, tracing—a copying technique that draws over the original sources through transparent layers—was utilised in the next step. Although tracing is rarely seen as an inquiry tool within the built environment discipline, it can unpack the source beyond what it initially represents by drawing things over (Lucas, 2017). Furthermore, it can pick up the oddities from the source and point these out for further investigation—it is both inscriptive and iterative. I found that tracing is a much-needed examining tool to look at every singularity and idiosyncrasies of street vending activities that might escape my observation in the initial open coding stage—a means to make sure that no important details were left unaddressed.

24The data sampling done by open coding helped to commence tracing as the follow-up step. Tracing would be tedious and lengthy without the samplings, involving drawing over a thousand Instagram posts. Since mind-walking does not rely on the quantitative reading of the data to construct semantic meaning of the phenomenon, using data samplings would not, in any case, affect the approach’s production of knowledge. However, contrary to the previous process, tracing is all done digitally using Adobe Illustrator (AI). Even so, this tool is chosen because it can mimic the analogue tracing process, not because it can automatically generate tracing results of the data. To maintain consistency, I had to ensure that every data was traced similarly. Thus the process always began with tracing the outline of all the visual components using continuous lines, followed by highlighting the newfound clues using dashed lines with arrows, and finished with labelling these clues using text boxes. As expected, I could generate more codes utilising this approach, especially those that were hard to catch by unaided eyes (Figure 3).

Figure . Comparison between the codes generated by open coding (left) and tracing (right)

Figure . Comparison between the codes generated by open coding (left) and tracing (right)

25Ultimately, these two processes produced 663 codes—176 found in Tanah Abang, 208 in Sudirman, and 279 in Kota Tua, each representing a characteristic of street vending activities. Some codes could be found in all locations, for example, “push-car”, “pavement”, and “food”,—implying the three urban settings shared some of the street vending features. Additionally, as we can see in Figure 4, each location also seemed to have its own unique codes; for example, “Car Free Day (CFD)” can only be identified in Sudirman, while “fortune telling” and “tattooing” could only be found in Kota Tua. This distinct variety of codes suggests that my initial hypothesis on how each part of Jakarta might have unique interior urbanism characteristics might be accurate. It is also supported by the fact that each area seems to generate a different quantity of codes, suggesting that perhaps the street activities within an area that developed more codes, such as Sudirman and Kota Tua, are more diverse than Tanah Abang. However, this was just speculation at this point since open coding and tracing are tools to expose the essential components of the data—not yet making something out of them. These ideas still need to be contested in the next stage of the mind-walking process. And to be able to do that, the investigation should continue the theme of engaging with the data based on their geographical information.

Figure . Codes were generated through open coding and tracing in three locations: yellow indicates Tanah Abang, green for Sudirman and red for Kota Tua

Figure . Codes were generated through open coding and tracing in three locations: yellow indicates Tanah Abang, green for Sudirman and red for Kota Tua

Axial coding and diagrams

26After the codes were all generated, I needed a tool to help me read and construct meaning from these codes. If we refer to grounded theory’s version of inductive coding, this refers to axial coding—a process in which codes are analysed and organised into categories, and later themes, based on their shared similarities (Corbin and Strauss, 1990). The logic behind the method is that categorisation would reveal the codes’ conceptual meanings and their roles within a phenomenon we are trying to unpack. Using this understanding, I began to organise the codes based on their shared qualities; for example, “push-car”, “bike-car”, and “shopping-cart” were grouped under the ‘analogue (mobile)’ category while “car” and “motorbike-cart” were grouped into ‘motorised (mobile)’. Furthermore, there is a similarity between these two categories—both represent the different features of units used by street vendors. Hence, I grouped these two categories into the ‘vending unit’ theme.

27Throughout this process, I managed to generate five themes that, surprisingly, exist in all three areas of Jakarta. However, the categories that comprised the themes varied in every location (Table 2). So if I were to explain this using a food analogy, what is shared by the three locations is the nutrients-- carbohydrates, protein and fat. Whereas the categories are the ingredients, the place where these nutrients are found. And supposedly, different combinations of ingredients would produce different types of food, not combinations of nutrients. Hence, this finding reaffirms the speculation on how each area in Jakarta has their own local street vending characteristics. However, we cannot combine the ingredients if we are unaware of their relationship with each other, the reaction caused if we mix them together, what must be cooked first to achieve the expected result, and so on. Axial coding might generate the ingredients, but it did not yet reveal how they might work and their role in assisting street vending activities.

Table . The result of the axial coding process. The three locations shared five themes with different variants of categories

Table . The result of the axial coding process. The three locations shared five themes with different variants of categories

28To be able to do that, we need to look beyond the apparent qualities of each code to link them into categories and themes that might reveal the different types of interior urbanism operations. However, unlike axial coding, this process might require interpretation. Thus, I needed to utilise a tool to draw the interpretation entirely from the codes to continue the theme of not tampering with the process with my personal assumptions. Thus, I employed diagrams, a drawing technique that is commonly used to represent arrangements and relations. Using diagrams would not only ensure less intrusive interpretations of the relationship between codes but also determine the role of each ingredient in the greater scheme of street vending operation.

29The diagramming process began by projecting the 105 tracing results into top-view drawings resembling site or floor plans. By doing this, I could start to spatially interpret the position and function of the codes in every street vending situation. The plan drawings also allowed me to imagine the interior space produced by the activities and draw the connection between the space and the existing urban contexts. However, while plan drawings do wonder for the codes that represent noticeable physical entities, it has limitation in exploring the immaterial ones. Therefore, a different type of drawing—a notation that explores activities, interactions and movements—is used to complement the first drawing. But, again, these drawings had yet to address the codes representing small material entities—the small objects that might not be captured by unaided eyes. Hence, I did another drawing in which I placed the visual representations of each code in correlation with the notation drawing. The aim is to explore the codes’ spatial arrangement and reveal their relationship with the rest of the components involved. Throughout the diagramming process, I constantly check the raw data, the original Instagram posts to avoid misinterpreting each code’s role and position within the whole interior urbanism situation.

30Instead of layering them on top of each other, I place them in individual frames to offer a detailed reading of each interpretative drawing. However, to show that every drawing is a part of a set that explains a unique street vending interior urbanism situation, I arranged them sequentially, side by side, beginning with the tracing result, the plan, the notation, and ending with the small objects drawing (Figure 5). For this arrangement, specifically, I drew inspiration from Bernard Tschumi’s Manhattan Transcripts, where to ease the analytical process, he successfully emphasised the significance of each interpretative drawing while simultaneously revealing its part within a grander scheme of events (Tschumi, 1994). The apparent difference is that while Tschumi focused on exploring the relationship between space and human activities, my version of the diagram also incorporates the engagement of all spatial components, including objects and immaterial entities. In some way, my diagrams are the interior translation of Tschumi’s—exemplifying the relationship between space, body, object, function, (mis)used, movement and intimacy (Taylor, 2010)—the whole interior packages in an urban setting.

Figure . Examples of the four-frame diagram as the continuation of the axial coding process

Figure . Examples of the four-frame diagram as the continuation of the axial coding process

31The result of this diagram exploration is a set of categories that organise the codes based on their spatial characteristics and their roles or function within street vending interior urbanism operations. For example, it is established that codes such as “push-cart” and “bike-cart” fall into the ‘analogue (mobile)’ category based on their similarities. However, based on how they were explored with the diagram, it is revealed that both units require the very minimum appropriation of the existing setting to facilitate street vending activities. Hence, on top of being part of the ‘analogue (mobile)’ group, they are also part of the ‘limited contact with the existing surface’ category (Table 3). Another example is the placement of codes such as “plastic stool”, “chair”, “table”, and “mat” that, through axial coding, were thrown together within ‘stationary activity’; through the diagram, it is revealed that they are also part of the ‘loose tools’ category—which is not only touched upon how these codes interact with the customers but also their nature in assisting the activity which includes their flexibility and attachment to vending units.

Table . The categories and themes produced through diagrams

Table . The categories and themes produced through diagrams

32While making diagrams was quite a tedious process compared to axial coding, organising the categories into themes was similar and as straightforward. For example, categories like ‘loose tool’, ‘built-in tools’ and ‘ephemeral tool’ can be grouped within ‘tool’, whereas ‘anticipative strategies’ and ‘responsive strategies’ fall into the ‘temporal condition responses’ theme. What makes it different, however, is that, unlike axial coding, the categories and themes generated through the diagram for all locations are similar. Nevertheless, it might be too hasty to dismiss the speculation that each location has its own interior urbanism, yet, since these ingredients we uncovered, the categories, are still pretty much raw. We might disclose their roles, but we have yet to find how they fit together and what kind of combinations are at play in each location. Since it is possible to produce different meals with the same set of ingredients, I believe it is also possible to formulate the different types of interior urbanism practices from the same set of categories. Knowing how the vendor utilises these ingredients would be a step forward to understanding the street vending phenomenon in each area of Jakarta.

Selective coding

33The step after axial coding is called selective coding, an approach that develops a narrative that links the themes generated by the previous coding process to produce one unified theory (Corbin and Strauss, 1990). The narrative, in our case, is the link that would combine different ingredients to suggest an understanding of interior urbanism operation utilised by Jakarta’s street vendors. However, this process might generate more than a narrative based on how the mind-walking processes develop up until this point. It would also finish the search for the ingredients; to complete the set halfway uncovered by the axial coding and diagramming processes. To begin the search for ingredients, I referred to the traced data (the 105 data sampling) and matched their codes to the diagram-generated themes and categories (Table 3) to find every possible combination of ingredients in a particular location. Each combination generated would reflect the type of spatial operations used by the vendors to inhabit the streets of Jakarta; together, these types would form the typology of Jakarta’s street vending. And through this matchmaking process, four types are revealed in Tanah Abang, five in Sudirman, and Kota Tua generated six (Table 4). The six types of interior urbanism include: converting, assembling, laying, lingering, parking and fleeting.

Table . The result of the selective coding process generates the typology of interior urbanism operation utilised by Jakarta’s street vendors as part of the ingredients

Table . The result of the selective coding process generates the typology of interior urbanism operation utilised by Jakarta’s street vendors as part of the ingredients

34There are a lot of similarities between converting and assembling (Figure 6); they both anticipate unpredictable temporal conditions with having tools or structures ready at all times. They also utilise the combination of built-in and loose tools to support their activities. By employing loose tools, both types encourage active participation from the customers that allow the customer to loan and adjust the position of the tools to facilitate their needs. And because of that, these types usually have multiple vending areas where the vendor’s and customer’s spaces are clearly divided. The difference between the two types is that with converting, the vendors interiorised the street by making a permanent physical alteration to the setting that would leave traces when the street is not inhabited. For the assembling type, the interior set-up is done by responding to the existing condition of the setting rather than making an alteration. Hence the interior form of this particular street vending is adjusted every time there is a change in the existing environment. However, regardless of this difference, the mobility of both types is pretty low. Because of their strong connection with the settings, their anticipatory response to the weather and the facilitation of social engagement with the tools used, these types require a longer duration and longer pauses to set up and pack up the activities. These three aspects also make the interior territories of these types pretty rigid, meaning that there is a strict order and clear expectations within the interior space of what kind of activities are involved and where.

Figure . The examples of interior urbanism types of converting, assembling and laying

Figure . The examples of interior urbanism types of converting, assembling and laying

35The third type, laying, have many shared characteristics with assembling. One of the two differences is rather than having an anticipatory strategy to respond to unpredictable weather and temporal conditions, this type develops a very responsive attitude. This means there is no equipment ready to protect the vending activities all the time; instead, it only utilises the equipment when necessary. The other distinction is that although laying still uses loose and built-in tools as well as encourages active participation from the customers, the fact it does not have as many ties with the context means that laying has higher mobility than the previous two types. It also has more adaptable interior territories since not only does it has to respond to the physical condition of the street, but it also has to be responsive to the unpredictable. Because of these reasons, you can find that this type is usually smaller in size and more versatile compared to the previous two.

36Just like the first two types, lingering, parking and fleeting, are pretty similar to each other (Figure 7). Nevertheless, somewhat in contrast with the other types, in producing the interior space, all have limited engagement with the existing street condition. They exist as is, meaning they do not change their original form to fit the settings and do not require alteration of the setting to facilitate their needs. These types are also very responsive to the weather and unpredictable temporal conditions. The significant difference is the tools used; lingering still facilitates social engagements and encourages customers’ active participation, while parking and fleeting only allow customers to participate passively. The other distinction is fleeting only utilised embodied tools attached to the vendors’ bodies that make their interaction with the customers very brief—hence the name. Additionally, by having active engagements with customers, lingering could generate multi-vending areas during their operation in contrast with parking and fleeting, which only has one the entire time. Nevertheless, since all three almost have no tie with the setting and utilise responsive strategies, as well as having most of the vending tools either fastened in the vendors’ bodies or the units, these types produce ephemeral interior situations with short pauses and a high degree of mobility.

Figure . The examples of interior urbanism types of lingering, parking and fleeting

Figure . The examples of interior urbanism types of lingering, parking and fleeting

37These types are part of the key ingredients in understanding the street vending phenomenon in Jakarta. As I was formulating the types, I noticed parts of these ingredients might be tightly related to the other ingredients uncovered through axial coding processes. For example, the vending units—although the choice of units might have affected the type of interior urbanism tactics used, no vending unit is exclusively associated with a particular type. For example, bike-cart or push-cart might fit the description of units used in parking, but if the vendors decide to use them to stay in one place for a long time and construct a temporary structure to protect them from rain, their interior urbanism would be described as either converting or assembling. I am not saying that the information about the various forms of vending units is unimportant; on the contrary, it is very crucial to have, but ultimately when it comes to formulating spatial rules within an urban street vending framework, this information needs to be read together with how the vendors utilised the units to make the activities happened (Table 5). Since there is a clear distinction between axial coding’s ingredients and selective coding’s—the former expresses nouns and adjectives while the latter represents the spatial system, these ingredients need to be used simultaneously to formulate an appropriate street vending framework for the city.

Table . Matrix of ingredients formulated through a mind-walking process

Table . Matrix of ingredients formulated through a mind-walking process

38Furthermore, I found a recurring narrative that binds all street vending operations in Jakarta; these interior urbanisms all operate based on the principle of mobility and temporality. Even the most rigid types, like converting and assembling, require the vendor to pause rather than stop. In these rigid types, the connection that ties the street vending activities and the setting is not extensively binding; it can be severed at any time. It is just that for these types, breaking the bond requires more time and effort than the other, more flexible types. Moreover, for most types, the interior forms and territories change from time to time depending on the condition of the settings – whether physical, socially or weather-wise, suggesting a considerable degree of temporality. This narrative suggests that, in Jakarta, street vendors work in a dynamic situation; hence the government’s current street vending strategies of displacement and immobilisation would never work to regulate them.


39I conclude this paper by saying that, unlike popular beliefs, street vending is not a marginal economic sector but a crucial urbanisation mode responsible for creating domestic feelings that make the city habitable, especially for the urban poor. Therefore, there is a need for a shift of attitude, especially from the policymakers, in acknowledging the significance of street vendors in the city by assimilating their informal practices into more comprehensive urban strategies. Without official data and government exclusion of street vending as a recognised labour force in the country, I proposed the use of mind-walking in uncovering Jakarta’s street vending interior urbanism ingredients using the publicly accessible social media platform Instagram. Combining inductive coding and drawing tools as a way to objectively read and extract the essential information from Instagram, mind-walking offers an alternative to field research, a remote way to do the bottom-up investigation of a social phenomenon.

40With three steps, mind-walking generates ingredients crucial to understanding Jakarta’s street vending activities. It also provides a narrative that explains the nature of these ingredients. The narrative suggests that all street vending activities in the city operate on the principle of mobility and temporality—two themes that were entirely antithetical to the provincial government's current strategies that required street vendors to be immobilised and structured. It is no wonder that these strategies always failed to regulate street vending activities. Furthermore, mind-walking also revealed that every location being investigated had different ingredients or combinations of ingredients, which indicates that each street vending hotspot in Jakarta might have its own distinct interior urbanism characteristics. This locality aspect should be considered, especially in formulating an appropriate street vendor framework, since it might seem that homogeneous and inflexible regulation might not be the answer. Instead, a more open-ended approach to a framework might be more suitable for this type of situation. Furthermore, by organising the ingredients generated into typology, the result of this experiment becomes instrumental. The typology summarises and highlights the similarities and differences between street vending types. It can be used as a tool for further analysis of Jakarta's street vending activity.

41Ultimately, this experiment demonstrates the potential of publicly available textual data to formulate a more socially equitable urban framework. While the tools used are not revolutionary, this experiment introduced a twist on using established analytical tools to deal with digital data, showcasing the versatility and adaptability of social research methods such as mind-walking and inductive coding. However, while this experiment certainly generates exciting results, the methodology has an apparent downside that needs to be addressed. While Instagram contains valuable information, it can only provide partial views of the actual urban issues, those that already been selected and curated for the public consumption. Furthermore, using mind-walking as the primary approach in this experiment extracts more physical and spatial data, it has yet to retrieve street vending activities' social, political, and economic aspects, which are equally essential to form a deep understanding of the phenomenon, ultimately serving as the solid base for formulating an appropriate framework. Moreover, I am convinced there might be many other facets and dimensions of street vending that have yet to be touched beyond mind-walking capability. Ideally, they should all be incorporated in formulating the appropriate urban street vending toolkit. Thus, despite establishing and setting out the interior urbanism ingredients that make up the phenomenon, mind-walking still needs to be followed by fieldwork to calibrate its findings. Only then could we begin formulating a socially equitable and Jakarta-centric urban street framework.

Top of page


Alzamil, W.S. (2018) ‘Evaluating Urban Status of Informal Settlements in Indonesia: A Comparative Analysis of Three Case Studies in North Jakarta’, Journal of Sustainable Development, 11(4), p. 148. Available at:

Arciniegas, L. (2021) ‘The foodscape of the urban poor in Jakarta: street food affordances, sharing networks, and individual trajectories’, Journal of Urbanism: International Research on Placemaking and Urban Sustainability, 14(3), pp. 272–287. Available at:

Ardanis, Y. (2022) Pandemi Akibat Virus Corona dan Pendapatan Alternatif, Kumparan. Available at: (Accessed: 30 January 2023).

Attiwill, S. (2011) ‘Urban and Interior: techniques for an urban interiorist’, in R.U. Hinkel (ed.) Urban interior: informal explorations, interventions and occupations. Baunach: Spurbuchverlag, pp. 11–24.

Baker, J.L. and World Bank (eds) (2012) Climate change, disaster risk, and the urban poor: cities building resilience for a changing world. Washington, D.C: World Bank (Urban development).

Budge, K. (2020) ‘Visually Imagining Place: Museum Visitors, Instagram, and the City’, Journal of Urban Technology, 27(2), pp. 61–79. Available at:

Budge, K. and Burness, A. (2018) ‘Museum objects and Instagram: agency and communication in digital engagement’, Continuum, 32(2), pp. 137–150. Available at:

Bureau of Street Services (no date) Vending Program, Available at: (Accessed: 21 September 2022).

Certeau, M. de (1984) The practice of everyday life. Berkeley: University of California Press.

City of New York (no date) Street Vending in NYC,, The Official Websity of the City of New York. Available at: (Accessed: 21 September 2022).

Corbin, J.M. and Strauss, A. (1990) ‘Grounded Theory Research: Procedures, Canons and Evaluative Criteria’, Zeitschrift für Soziologie, (Jg. 19, Heft 6), pp. 418–427.

Corbin, J.M. and Strauss, A.L. (2007) Basics of qualitative research: techniques and procedures for developing grounded theory. 3rd ed. Los Angeles, Calif: Sage Publications, Inc.

Cupers, K. (2020) ‘The Urbanism of Los Angeles Street Vending’, in Graaff, Kristina and Ha, Noah (eds) Street Vending in the Neoliberal City: A global perspective on the practices and policies of a marginalised economy. New York: Berghahn Books, pp. 139–163.

Devlin, R.T. (2021) ‘Winning a Right to the Sidewalks: Street Vendors in New York’, WIEGO [Preprint].

Dimas, H. (2008) ‘“Street Vendors: Urban Problem and Economic Potential”’. Department of Economics, Padjadjaran University.

Duneier, M. (2001) Sidewalk. 1st paperback edition. New York: Farrar, Straus and Giroux.

Emans, R. and Fisher, G.M. (1967) ‘Teaching the Use of Context Clues’, Elementary English, 44(3), pp. 243–246.

Fachriansyah, R. (2020) Tanah Abang Market packed with traders ahead of Idul Fitri despite virus fears, The Jakarta Post. Available at: (Accessed: 30 January 2023).

Gibbings, S.L. (2016) ‘Street Vending as Ethical Citizenship in Urban Indonesia’, Anthropologica, 58(1), pp. 77–94.

Gioia, D.A., Corley, K.G. and Hamilton, A.L. (2013) ‘Seeking Qualitative Rigor in Inductive Research: Notes on the Gioia Methodology’, Organizational Research Methods, 16(1), pp. 15–31. Available at:

Graaff, K. and Ha, N. (eds) (2020) Street Vending in the Neoliberal City: A global perspective on the practices and policies of a marginalised economy. New York: Berghahn Books.

Herman (2013) Keluhan PKL Pasar Minggu Setelah Dipindahkan ke Lokbin, Available at: (Accessed: 27 September 2022).

Hinkel, R.U. (ed.) (2011) Urban interior: informal explorations, interventions and occupations. Baunach: Spurbuchverlag.

Huang, J. et al. (2021) ‘The image of the City on social media: A comparative study using “Big Data” and “Small Data” methods in the Tri-City Region in Poland’, Landscape and Urban Planning, 206, p. 103977. Available at:

Ingold, T. (2010a) ‘Footprints through the weather-world: walking, breathing, knowing’, The Journal of the Royal Anthropological Institute, 16, pp. S121–S139.

Ingold, T. (2010b) ‘Ways of mind-walking: reading, writing, painting’, Visual Studies, 25(1), pp. 15–23. Available at:

Jacobs, J. (1992) The death and life of great American cities. Vintage Books ed. New York: Vintage Books.

Khairunnisa (2016) ‘Alternative for Urban Upgrading in the Central Business District Jakarta: A case study at Rasuna Said Avenue’. Available at: (Accessed: 19 July 2022).

Kolkata Municipal Coorperation (no date) KMC Trade License, Available at: (Accessed: 7 October 2022).

Leo, P.J. (2020) ‘Pedagang Kaki Lima di Stasiun Kota’, The Jakarta Post Image, 21 February. Available at: (Accessed: 2 May 2022).

Lucas, R. (2017) ‘The discipline of tracing in architectural drawing’, in C.M. Johannessen and T. van Leeuwen (eds) The Materiality of Writing: A Trace Making Perspective. Place of publication not identified: Routledge (Routledge Studies in Multimodality).

McCullough, C.M. (1945) ‘The Recognition of Context Clues in Reading’, The Elementary English Review, 22(1), pp. 1–38.

Mörtenböck, P. and Mooshammer, H. (eds) (2015) Informal market worlds atlas: the architecture of economic pressure. Amsterdam: NAI010 Publ.

Ng, S. (2022) Lokbin Kota Intan Ditolak PKL karena Sepi, PPUKM: Sekarang Sudah Beda, detiknews. Available at: (Accessed: 27 September 2022).

Nurhayati-Wolff, H. (2022) Indonesia: most used social media platforms 2022, Statista. Available at: (Accessed: 14 February 2024).

Pattiradjawane, H., Schnepf-Orth, M. and Stoetzer, S. (2013) Negotiating Informal Urban Space: Female Cake Vendors at the Pasar Kue Subuh Senen Night Market in Jakarta, Indonesia. Darmstadt: Technische Universität Darmstadt, p. 157. Available at:

Rachmawan, D., Fathy, R. and Luthfi, M. (2021) ‘Vulnerability and Survival Strategy by Food Street Vendors: Case Study in Triangle Gold of DKI Jakarta’, JSW (Jurnal Sosiologi Walisongo), 5(2), pp. 111–126. Available at:

Rice, C. (2016) Interior urbanism: architecture, John Portman and downtown America. London Oxford New York New Delhi Sydney: Bloomsbury Academic, an imprint of Bloomsbury Publishing Plc.

Rosalina, M.P. (2019) ‘Trotoar Multifungsi dalam Benak Warga’,, 13 October.

Simone, A. (2015) ‘The Urban Poor and Their Ambivalent Exceptionalities: Some Notes from Jakarta’, Current Anthropology, 56(S11), pp. S15–S23. Available at:

Sulistywati, R.L. (2016) Ahok to conduct massive evictions, Republika. Available at: (Accessed: 26 January 2023).

Taylor, J. and Song, L. (2016) ‘Return to the Streets’, Cityscape, 18(1), pp. 71–88.

Taylor, M. (2010) ‘Diagramming the Interior’, in M. Garcia (ed.) The diagrams of architecture. Chichester: Wiley (AD reader), pp. 134–141.

The Kolkata Gazette Extraordinary (2018) West Bengal Urban Street Vendors (Protection of Livelihood and Regulation of Street Vending) Rues. 2018.

The Street Vendor Project (no date) The Street Vendor Project. Available at: (Accessed: 21 September 2022).

Tschumi, B. (1994) ‘Manhattan Transcripts’, ANY: Architecture New York, March/April 1994(No. 5, Lightness (March/April 1994)), pp. 48–49.

Wilson, I.D. (2010) ‘“The streets belong to who?”: “Governance” and the Urban Informal Sector in Jakarta, Indonesia’, in The Elephant in the Room: Politics and the Development Problem. Perth: Asian Research Centre, Murdoch University, pp. 113–133.

Yatmo, Y.A. (2008) ‘Street Vendors as “Out of Place” Urban Elements’, Journal of Urban Design, 13(3), pp. 387–402. Available at:

Top of page

List of illustrations

Title Figure . Street Diagram issued along with 2004 Republic of Indonesia Acts No. 38 shows the street and its components that should be free from any objects and activities aside from traffic and pedestrian movements. Redraw by Author.
File image/jpeg, 233k
Title Table . The urban street vending regulations matrix in New York, Los Angeles and Kolkata. Source: (The Kolkata Gazette Extraordinary, 2018; Bureau of Street Services, no date; the City of New York, no date; Kolkata Municipal Corporation, no date; The Street Vendor Project, no date)
File image/jpeg, 992k
Title Figure . Pinned-up printed screenshots of Instagram posts on the walls with post-it notes that label the clues
File image/jpeg, 1.0M
Title Figure . Comparison between the codes generated by open coding (left) and tracing (right)
File image/jpeg, 600k
Title Figure . Codes were generated through open coding and tracing in three locations: yellow indicates Tanah Abang, green for Sudirman and red for Kota Tua
File image/jpeg, 993k
Title Table . The result of the axial coding process. The three locations shared five themes with different variants of categories
File image/jpeg, 312k
Title Figure . Examples of the four-frame diagram as the continuation of the axial coding process
File image/jpeg, 1.9M
Title Table . The categories and themes produced through diagrams
File image/jpeg, 217k
Title Table . The result of the selective coding process generates the typology of interior urbanism operation utilised by Jakarta’s street vendors as part of the ingredients
File image/jpeg, 377k
Title Figure . The examples of interior urbanism types of converting, assembling and laying
File image/jpeg, 196k
Title Figure . The examples of interior urbanism types of lingering, parking and fleeting
File image/jpeg, 222k
Title Table . Matrix of ingredients formulated through a mind-walking process
File image/jpeg, 183k
Top of page


Electronic reference

Anak Agung Ayu Suci Warakanyaka, “Mind-walking through Instagram: extracting the essences of Jakarta’s street vending”Articulo - Journal of Urban Research [Online], 24 | 2024, Online since 04 April 2024, connection on 20 June 2024. URL:; DOI:

Top of page

About the author

Anak Agung Ayu Suci Warakanyaka

PhD Candidate, Royal College of Art. Kensington Gore, South Kensington, London SW7 2EU.

Top of page



The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search