Ann M. Graf – Simmons University, United States Domain Analysis of Graffiti Art Documentation A Methodological Approach Abstract: Details are presented of a recent research project undertaken to ascertain the documentary and descriptive practices associated with graffiti artwork from within the graffiti art community as evidenced by 241 graffiti websites. Domain analytic methodologies following a pragmatic approach to knowledge organization and using evidence obtained from within an artistic community are extremely useful ways to provide insight into what are the most important facets of information to capture for works not often documented from within libraries, archives, and museums. This paper will discuss various methods used to analyze community-driven graffiti art collection, organization, and description in the online environment, the results of which form a part of the basis upon which a faceted KOS can be built. 1.0 Introduction Knowledge organization systems (KOS) in use for the documentation and description of artworks have a respectable, if shorter, history compared to those used in libraries (Urban 2014). This is often understood to be due to the fact that museums are most often collecting and creating surrogate records for unique objects (Taylor 1999). Unlike the library, the art museum is representing objects that would not benefit from shared cataloging practice, though this is changing as images of artworks are increasingly available online and users desire to have access to them regardless of where they are physically located. For certain types of art that often fall outside the purview of the formal institution, such as graffiti art, documentation and organization of the resulting image records is carried out largely by the graffiti art community itself, including graffiti artists and enthusiasts. This is due to the extra-institutional nature of the artworks, the legal complications often surrounding their creation, and the inability to monetize the works when found “on the street” (Schacter 2014). Interest in the artworks continues to rise, despite these documentary challenges, evidenced by the large number of websites dedicated to preserving the images of the works around the world. While KOSs in popular use in the library, archives, and museum environment do not include granular terminology to address the many facets of graffiti art, at least one, the Getty Art and Architecture Thesaurus (AAT), has recently added a limited number of graffiti art descriptive terms. This paper introduces research conducted on the state of descriptive and organizational practice applied to collections of graffiti art images in websites from around the world. The data for analysis comes from a set of 241 graffiti websites. This stage of the research will report on the categories – or facets – used to organize the image galleries themselves. This research does not explore description as applied to individual images. Knowledge of organizational practice among those documenting the art form in image galleries online is foundational to understanding the implicit categories, vocabulary, and details associated with a particular artistic tradition and may inform further research on documentation and organization of outsider art. It is very hard to pin down a community, to draw lines around those within and those without. In the case of graffiti art, with its intersecting boundaries of what is legal and what is not, and what is defined by some as true graffiti, and by others as vandalism, 162 and yet others as the more sanitized term street art, it can be impossible to find consensus. Should the artists themselves be the ones to say what they are doing, how they are doing it, and how it is described, documented, and organized? Are those who simply love the often bright, complex, and (sometimes) publicly placed works allowed to have a say? Do those who actively look for the works and photograph them, sharing them online in large galleries, belong to this community? Because graffiti image galleries online are very often cooperative endeavors, whether knowingly or not, willingly or not, of images submitted by artists who created the works, photographers who stumbled upon them, webmasters or social media mavens who enjoy them, and any combination of these and more, the easiest way to begin this research was to use the evidence of the collections themselves. It is acknowledged from the start that this is a messy endeavor, trying to decide who is acting and what role they may play in the organization of graffiti art online. While few authors have addressed the challenges of documenting graffiti art specifically (see Masilamani 2008 and Gottlieb 2008 for two of the more robust examples), the process is actively taking place in a very broadly distributed fashion, each participant seemingly acting independently of the others. Despite their autonomy, those doing the documentation in this research represent a community or a domain in that they share “an ontological base that reveals an underlying teleology, a set of common hypotheses, epistemological consensus of methodological approaches, and social semantics” (Smiraglia 2012, 114). This work fills a gap in the research by illuminating the facets for organization of graffiti art images as used by those working to share large collections of the works online. The value of domain analysis as a research tool is well documented in the knowledge organization literature (Hjørland and Albrechtsen 1995, Hjørland 2002, Smiraglia 2015, Albrechtsen 2015), as well as facet analysis (Hjørland 2013, Cho et al. 2018, Campbell 2004). There are no domain analyses that examine modern graffiti art image documentation and the facets used as attributes to organize the images. Research reported by Graf (2016) revealed the lack of graffiti art-related terminology available in the AAT. Only three out of the twenty most often used graffiti terms from her analysis of graffiti zines appeared in the AAT. Interestingly, within two years, eleven more of the same twenty terms were added to the AAT, bringing the original percentage from 15% to 70%. This indicates the influence of the graffiti art community and their practices, and the reporting of research on those practices, on widely used professional tools for the documentation and organization of artworks. Further granularity can be found in this current research for those desiring to extend the available terminological offerings for graffiti art documentation. 2.0 Methodology As preliminary research in this area, the first step was to decide exactly what would be examined. There are many websites devoted to the documentation of graffiti and street art. One of the very first and most well known of these is Art Crimes (www.graffiti.org). The About page on their website states that “Art Crimes was the first graffiti site on the net, and we're still one of the biggest …” (Art Crimes 2020). As an early and large graffiti art website, Art Crimes has gathered links to numerous other graffiti and street art websites around the world. At the time of the research, Art Crimes included a 163 list of 709 links to other sites. This list was used as the basis for the eventual set of 241 websites evaluated. Each of the 709 links on Art Crimes was visited and a judgement was made on whether or not to include the site in the study based on several criteria. 318 of the links were either dead, empty, or presented a notification that the site had moved without providing forwarding information. 64 sites were fully in languages other than English and therefore eliminated from the study, though some sites that were kept employed other languages but kept navigation labels for the site in English. 57 of the sites were professional artists’ sites, not specifically galleries of graffiti or street art images. 20 of the sites were not relevant because they were focused on music, advertising, or other products or services. Eight of the sites were links to social media galleries, such as Flickr or Instagram. These were not included in this study at this time because of the organizational confines of social media platforms. Each social media platform includes specific ways that uploaded images can be labeled, organized, and grouped. There exist thousands of graffiti and street art image galleries on social media platforms, and they are ripe for further investigation, but were considered outside the purview of this research. One site among the 709 was not an independent website, but rather a sub-page of the Art Crimes website itself, and therefore eliminated, though Art Crimes itself remained in the study. After all of the sites were evaluated in this way, 241 live sites remained. Each site was evaluated for structural elements of pages and sub-pages, indicated by navigation labels and hotlinked text. Examples of navigation labels can be easily seen across the top of the webpage banner in Figure 1 for the site 50mm Los Angeles (http://www.50mmlosangeles.com/). These navigation labels include: Gallery, Articles, Events, L.A. Legends, Blackbook, Links, Forum, About Us, and Submit an Event. There is also hotlinked text, “login | register”, that leads to other sub-pages of the website. All of these labels were entered into a QDA Miner database for each of the 241 websites. Figure 1. 50mm Los Angeles website home page with navigation labels. Each of the sub-pages accessed through the navigation labels and hotlinked text were visited and each was evaluated for evidence of further navigation labels, or sub-divisions of organization. Some websites had very shallow organization with only a couple levels, while others were deeper structures with several levels. An example of a website with deeper structure is Fatcap (fatcap.com). The home page of this website includes a navigation label for “pictures.” Hovering the mouse over this label gives the user several sub-levels from which to choose: all pictures, worldwide graffiti, artists, crews, types, supports, and styles. Clicking on “worldwide graffiti” takes the user to a new sub-page 164 that includes links to 117 deeper sub-pages, arranged by larger geographic regions: Africa, Asia, North America, South America, Europe, and Oceania. Clicking further, on “United States” for example, takes the user to a list of 40 states and the District of Columbia, each with from 1 to 69 individual cities linked as sub-pages to go even deeper. Other websites only had a couple levels, but then divided a single level into hundreds of sub-pages. An example of a sub-page in a shallower organizational structure with numerous organizational divisions, again from the website 50mm Los Angeles, is shown in Figure 2. Only part of the screen is visible in this image, but the organization of the image gallery is divided into an alphabetical, hotlinked list of navigation labels, some representing artist names (pseudonyms), locations, and styles, etc. Clicking on any of the hotlinked labels in this expansive list will take the user to a gallery of graffiti and street art images with works that have something to do with the label. Each of these subpages used as organization for image galleries was also entered into the QDA Miner database for further evaluation, as will be explained in detail below. Figure 2. Gallery page of 50mm Los Angeles website with links to individual galleries. Once all 241 websites were visited and all navigation labels were entered into QDA Miner for all levels of organization, each individual label was coded to indicate what type of organization was indicated. The coding developed as the analysis proceeded. Six broad categories of codes, or facets, evolved during the research, two of which focused 165 on the websites themselves, and four of which focus on the artwork images on the websites. The two categories of codes that apply to the websites themselves include Sites and Other Media. The Sites category includes navigation labels that refer to aspects of the websites, how users can interact with the websites, and other information related to shopping, subscribing, and other graffiti and street art-related information accessed within the websites. They do not concern description of graffiti and street art images associated with the image galleries on the websites. The Sites codes include: About, Contact, ContributeFlix, Disclaimer, FAQ, Forum, Glossary, Guestbook, History, HowTo, Interviews, Map, MyAccount, Poll, Shop, Subscribe, and Videos. This category of codes relates to the structure, navigation, and use of the website in general. The second of the two categories not concerned with description of graffiti and street art images is the Other Media codes. This category of code was applied to navigation labels that linked to a blog or social media account associated with a website, such as an Instagram, Facebook, or Flickr account, or to a list of links to other graffiti or street art sites, or other associated media located outside the websites studied. The remaining four categories of codes are the focus of the research reported herein. These categories were used to describe graffiti and street art images themselves and include General, Types, Supports, and Locations. Each of these four works-based categories will be described in greater detail and will provide insight into the documentation, description, and organization practices of the graffiti art community, which includes artists, photographers, and various enthusiasts as described earlier. 3.0 Findings Each of the code categories is divided into sub-categories, which reflect aspects of description for graffiti works. The first of these is the General category, which is divided into 17 codes, or facets, as shown in Table 1. In each of the code tables, the name of the code is given first, followed by how many times that code was applied over all 241 websites. The third column indicates the percentage of all codes applied. The fourth column indicates how many of the 241 sites earned that code at least once, followed in the last column by the percentage of all sites that used that code at least once. Some of the websites earned the same coding in multiple places on the site, which accounts for the sometimes very large number of individual codes, like Artist. Whenever the name of an artist was used as a way to organize a gallery of images, that label of the artist’s name was coded as Artist. As evident in Figure 2, some websites included hundreds of individual artist’s names. Each of the tables lists codes in order of the percentage of sites making use of the code at least once, providing a type of ranking for the popularity of an aspect of organization across all sites. This also avoids the skewing effect of using each code instance as a popularity measure instead. While the month of a work was applied 35 individual times, putting it fifth in terms of instances, it was seen on only 2.1 percent of all sites, or 5 sites, which indicates it was 13th out of 17 in popularity. Some of the General codes reflect common aspects of traditional art documentation, such as the use of an artist’s name or the year. Others reflect affordances of an online gallery, such as New, Color, Featured, RatedHigh, and Old. A very interesting aspect of many of the General codes is their specific applicability to graffiti art. This is evident in 166 codes such as Gallery (used to indicate when a work was in a gallery and not in a traditional graffiti location), RIP (used for commemorative pieces in honor of a graffiti artist who has died), Legal, Outside, and Illegal. Most graffiti is assumed to be illegal, but there are also legal walls where graffiti is allowed. It makes sense to include organization for legal works, as they are created under very different circumstances than illegal ones. It doesn’t make as much sense to offer organization specifically for illegal works, and this code was applied on only 2 sites, compared with 7 sites that earned the Legal code. Table 1. General codes and their usage across all sites. General Codes Count % of Codes # of Sites % of Sites Artist 14439 71.2 50 20.7 Event 89 0.4 31 12.9 Gallery 49 0.2 29 12.0 Year 227 1.1 27 11.2 New 35 0.2 26 10.8 Old 35 0.2 26 10.8 Featured 27 0.1 20 8.3 Inside 11 0.1 10 4.1 RIP 75 0.4 10 4.1 RatedHigh 14 0.1 8 3.3 Legal 15 0.1 7 2.9 Outside 7 0.0 7 2.9 Month 35 0.2 5 2.1 Color 12 0.1 4 1.7 Day 5 0.0 4 1.7 Decade 8 0.0 4 1.7 Illegal 5 0.0 2 0.8 The next category of work-related codes are the Support codes. These codes were applied when organizing by the surface upon which the artwork was created or placed. One distinction in this group of codes is the Canvas code, applied here to works produced in a studio. One-third of all sites earned this code, reflecting the difference in perception of graffiti-style artworks committed on canvas as opposed to walls, trains, or other publicly accessible spaces. The use of the street is important to the notion of graffiti and street art (Austin 2010, Riggle 2010). Painting on canvas is often seen as a desire for profit, a safe way to make art in the comfort of a studio, or a type of selling out of the art form (Jacobson 2017). This conception is common enough that a relatively large number of sites used this type of organizational label to separate out works made in a studio from those made on the streets. Table 2. Support codes and their usage across all sites. Support Codes Count % of Codes # of Sites % of Sites Canvas 109 0.6 77 32 Walls 107 0.5 65 27 Trains 253 1.2 51 21.2 167 Blackbook 28 0.1 20 8.3 Freights 27 0.1 16 6.6 CarsTrucksVans 28 0.1 12 5.0 Subways 81 0.4 11 4.6 Billboards 10 0.0 5 2.1 Body 5 0.0 4 1.7 Clothing 11 0.1 4 1.7 Rooftops 4 0.0 4 1.7 Tunnels 5 0.0 4 1.7 Subway Cars 60 0.3 3 1.2 Buses 3 0.0 2 0.8 Highways 2 0.0 2 0.8 Signs 2 0.0 2 0.8 Skate Deck 2 0.0 2 0.8 Trash Bins 2 0.0 2 0.8 Shutters 2 0.0 2 0.8 One of the most interesting categories of codes is that devoted to types of art. This category is rich with terminology, much of it specific to graffiti art. It is also the largest of the code categories, with 31 individual codes. Many of the codes are familiar terms that could be associated with more traditional art forms, such as Sketches, Murals, Stencils, Posters, and Political. Many others have specific meaning within the graffiti art community, such as Tags, Pieces, Bombs, Throwups, Productions, TrainWholecars, TrainEtoEs (end-to-ends), TrainTtoBs (top-to-bottoms), Wheatpaste, and Wildstyle. Table 3. Type codes and their usage across all sites. Type Codes Count % of Codes # of Sites % of Sites Sketches 74 0.4 56 23.2 Graffiti 75 0.4 50 20.7 Other 184 1.1 43 18.3 CommercialDesign 63 0.3 43 17.8 StreetArt 38 0.2 35 14.5 Murals 39 0.2 32 13.3 Tags 22 0.1 17 7.1 3D 18 0.1 16 6.6 Characters 59 0.3 15 6.2 Pieces 37 0.2 15 6.2 Stencils 20 0.1 13 5.4 Bombs 13 0.1 12 5.0 Throwups 16 0.1 12 5.0 Letters 19 0.1 10 4.1 Productions 12 0.1 10 4.1 Stickers 14 0.1 10 4.1 Digital 8 0.0 8 3.3 TrainWholecars 13 0.1 8 3.3 168 Action 6 0.0 6 2.5 Posters 9 0.0 5 2.1 SprayPaint 4 0.0 4 1.7 Wheatpaste 4 0.0 4 1.7 Political 3 0.0 3 1.2 Projections 3 0.0 3 1.2 TrainEtoEs 5 0.0 3 1.2 Collaborations 3 0.0 3 1.2 TrainPanels 3 0.0 3 1.2 Silvers 2 0.0 2 0.8 TrainTtoBs 2 0.0 2 0.8 Wildstyle 3 0.0 2 0.8 Handstyle 2 0.0 2 0.8 The last category of work-related codes is the Location codes. Location can be considered a common attribute to document for most traditional artworks, but it holds special significance in the graffiti art community. Graffiti art styles are passed on from older, more established writers to younger ones, and graffiti writers will often “bite” or copy work they admire by others. Styles can be associated with geographic locations around the world as well as with individual artists. Having as precise a location as possible for an individual work is desirable for those wishing to see the work in person as well as for those who research the art style and its evolution across time and space. The value of location information is counterbalanced by the desire of artists acting illegally to remain anonymous and not leave a trail by which they can be tracked by law enforcement. This tension is evident in the very consistent lack of precise location information available across all 241 websites. The most commonly employed level of geographic location information was by city, followed closely by country. Only one site got close enough to mention a street address, while another once mentioned an intersection of streets. Thirteen sites referenced location via specific landmarks that might be recognizable to some familiar with the next level up in the location hierarchy, such as city name. Parts of cities were also used by thirteen sites. This would include mention of a specific borough of New York City, or a cardinal direction employed with a city name, such as East L.A. Many sites employed numerous levels of geographic faceting, starting by continents or countries, and working down through specific states and cities. A number of websites were geographically focused and indicated works from countries outside their focus with a gallery for World graffiti, a type of “other” code. Table 4. Location codes and their usage across all sites. Location Codes Count % of Codes # of Sites % of Sites Cities 1637 8.6 43 17.8 Countries 543 2.8 37 15.8 SpecificLandmarks 73 0.4 13 5.8 CityParts 94 0.5 13 5.4 World 22 0.1 12 5.0 169 Continents 42 0.2 11 4.6 States 117 0.6 6 2.5 CountryParts 10 0.0 5 2.1 Address 2 0.0 1 0.4 Intersection 27 0.1 1 0.4 Undisclosed 2 0.0 1 0.4 4.0 Conclusion and Further Research This research has analyzed the organizational facets employed across a large number of websites that share graffiti art images and how they are further broken down. The work is the first of its kind to describe the results of efforts by a distributed group of image collection managers to organize graffiti images online. It is valuable in that it provides insight into the facets of organization in use around the world to categorize artworks not often collected or documented by traditional art or cultural heritage institutions. The methodology employed has been fruitful for description of current organizational practice. Several factors intrinsic to the art form complicate the ability to effectively use traditional methods of documentation for works of art, including in large part the often illegal nature of creating graffiti and street art. Legal issues often contribute to the obfuscation of common aspects of art documentation, such as where works are, the identities of those who created them, and dates for creation, change, and destruction of works. Commonalities in practice are easily seen from this research, but there is a lot more that can be gained by further study. Knowing how these works are organized right now is only half of the equation that could lead to development of systems to serve not only those who maintain these diverse collections, but the many users who approach them as well. Further research into the needs and desires of such collection users could fill in other missing facets that may be extremely useful. While conducting this research, it was found that a number of the websites studied were very well developed. Nineteen of the 241 sites had very large collections, employed notably consistent, very granular use of facets, were active and adding new images, and included clear and in-depth information about the sites themselves. These 19 sites have been noted for further study. Much more analysis was carried out on the websites and interviews of website curators were also conducted, adding still more valuable information about why certain facets and terminology are used. These additional details, in combination with study from the user perspective, would add to the growing amount of information that could be used to design systems for the documentation and organization of graffiti art and street art, as well as other types of found art and ephemera. References Albrechtsen, Hanne. 2015. “This is Not Domain Analysis.” Knowledge Organization 42: 557-61. Art Crimes. 2020. About Art Crimes: What We’re Doing and Why. https://www.graffiti.org/index/story.html. Austin, Joe. 2010. “More to See than a Canvas in a White Cube: For an Art in the Streets.” City 14, nos. 1/2: 33-47. Campbell, D. Grant. 2004. “A Queer Eye for the Faceted Guy: How a Universal Classification Principle Can Be Applied to a Distinct Subculture.” In Knowledge Organization and the Global Information Society: Proceedings of the Eighth International ISKO Conference 13-16 170 July 2004 London, UK, edited by Ia C. McIlwaine. Advances in knowledge organization 9. Würzburg: Ergon Verlag, 109-13. Cho, Hyerim, Thomas Disher, Wan-Chen Lee, Stephen A. Keating, and Jin Ha Lee. 2018. “Facet Analysis of Anime Games: The Challenges of Defining Genre Information for Popular Cultural Objects.” Knowledge Organization 45: 484-99. Gottlieb, Lisa. 2008. Graffiti Art Styles: A Classification System and Theoretical Analysis. Jefferson, NC: McFarland & Company, Inc. Graf, Ann M. 2016. “Describing an Outsider Art Movement from Within: The AAT and Graffiti Art.” In Knowledge Organization for a Sustainable World: Challenges and Perspectives for Cultural, Scientific, and Technological Sharing in a Connected Society. Proceedings of the Fourteenth International ISKO Conference 27-29 September 2016, Rio de Janeiro, Brazil, edited by José Augusto Chaves Guimarães, Suellen Oliveira Milani, and Vera Dodebei. Advances in knowledge organization 15. Würzburg: Ergon, 125-32. Hjørland, Birger. 2002. “Domain Analysis in Information Science: Eleven Approaches – Traditional as Well as Innovative.” Journal of Documentation 58: 422-62. Hjørland, Birger. 2013. “Facet Analysis: The Logical Approach to Knowledge Organization.” Information Processing and Management 49: 545-57. Hjørland, Birger and Hanne Albrechtsen. 1995. “Toward a New Horizon in Information Science: Domain-Analysis.” Journal of the American Society for Information Science 46: 400-25. Jacobson, Malcolm. 2017. “Marketing with Graffiti: Crime as Symbolic Capital.” Street Art & Urban Creativity Scientific Journal 3, no. 2: 102-11. Masilamani, Rachel. 2008. “Documenting Illegal Art: Collaborative Software, Online Environments and New York City’s 1970s and 1980s Graffiti Art Movement.” Art Documentation 27, no. 2: 4-14. Riggle, Nicholas Alden. 2010. “Street Art: The Transfiguration of the Commonplaces.” Journal of Aesthetics & Art Criticism 68: 243-57. Schacter, Rafael. 2014. “The Ugly Truth: Street Art, Graffiti and the Creative City.” Art & the Public Sphere 3, no 2: 161-76. Smiraglia, Richard P. 2012. “Epistemology of Domain Analysis.” In Cultural Frames of Knowledge, edited by Richard P. Smiraglia and Hur-Li Lee. Würzburg: Ergon Verlag, 111-24. Smiraglia, Richard P. 2015. Domain Analysis for Knowledge Organization: Tools for Ontology Extraction. Waltham, MA: Chandos. Taylor, Arlene G. 1999. “Museums and Art Galleries.” In The Organization of Information, Englewood, CO: Libraries Unlimited, 9-11. Urban, Richard J. 2014. “Library Influence on Museum Information Work.” Library Trends 62: 596-612.