Of Grasshoppers and Rhinos: A Visual Literacy Approach to Born-Digital Design Records

Archival discussions and professional literature on born-digital design records have, to date, largely focused on the technical digital preservation and access challenges of files such as computer-aided design (CAD) drawings, three-dimensional (3D) models, and building information models (BIM). From file complexity to reliance on proprietary software, and in some cases hardware, such challenges are numerous and well documented, and the archival community has engaged in several decades of research, experimentation, and collective action in an effort to meet these challenges.¹ Yet, these discussions largely have not engaged with the equally necessary visual literacy skills and domain expertise required for archivists to effectively conduct activities such as appraisal, description, and outreach and education that require engaging with the content of records in addition to their formats. Such skills will be necessary to empower archives to collect born-digital design records soon enough after their creation and active use to avoid significant social knowledge loss. However, nearly twenty-five years since the previous special issue of American Archivist on architectural records, few resources for developing such skills exist, and fewer still design archives are actively engaged in collecting and providing access to born-digital design records at significant scale.²

This article aims to serve as a professional resource to assist archivists in developing the requisite knowledge and skills to interact with the content of born-digital design records, while demonstrating the centrality of visual literacy to the task. “Appraisal of architectural records is often viewed as a difficult and puzzling problem for archivists,” and the complexities of CAD drawings, 3D models, and BIMs make this task even more challenging for born-digital design records than for paper drawings.³ In the absence of established best practices for and literature about applying visual literacy to born-digital design records, we aim to build on an existing approach to applying visual literacy techniques to analog architectural records developed at the University of South Australia and published in “Behind the Image: Assessing Architectural Drawings as Cultural Records.”⁴ To this framework, we apply knowledge about born-digital design records pulled from our own deep and varied experiences as archivists managing born-digital design records in collecting institutions and design firms and as researchers using them as primary source materials for scholarship.

This article is intended to lower the barrier to working with born-digital design records for any archivists who may be charged with appraising, arranging, describing, preserving, and providing access to them, not only those who already specialize in records of architecture and design. From otherwise-specialized archivists who suddenly find themselves in charge of the records of their own building to archivists in historical societies that will soon or have already received CAD drawings of their local built environment, a wide number of archivists may find themselves encountering such records at some point in their careers. In the interest of being accessible to a wide range of archivists, this article intentionally includes some basic knowledge about design records in general that might already be familiar to design archivists, in addition to discussing the unique aspects of born-digital records.

In many cases, significant overlap may exist between the visual literacy skill sets and domain knowledge required by archivists and researchers to engage with born-digital design records. Properly assessing the significance and cultural dimensions of design records requires an understanding of the types of research questions users bring to them and how to read a drawing or model to answer such questions. Building such an understanding will enable archivists to better perform functions such as appraisal, description, and outreach and education.⁵ Where possible, this article uses examples of born-digital design records from archival collections to illustrate some of the questions and methods that researchers will bring to these records and how they might inform archival practice.

Although the terms “architectural records” or “design records” are employed in the context of this article, the records being discussed can include drawings and models produced not only by architects but also by those in the associated fields of building, planning, engineering (e.g., structural, mechanical), graphic design, industrial design, urban design, landscape architecture, and interior design. Neither is reliance on computer modeling limited to the architectural field. As modeling permeates many other areas of human endeavor from materials science, chemistry, and medicine, to aerospace, fashion, and gaming, born-digital records produced in other fields may be approached in similar ways to architectural drawings and models.

Much of the work undertaken in the fields of architecture and engineering over the last thirty years has been designed and documented through two- and three-dimensional CAD and BIM models that are increasingly linked to a network of dynamic online libraries of materials and components. Born-digital architectural records continue the lineage of architectural drawing, having been based on the precedents set by paper-based drawings.⁶ These tangible drawings have been, and still are in some cases, used to communicate a three-dimensional object (the building) through scaled orthographic projections, that is, plans, sections, and elevations (see Figure 1). Each line on a drawing represents an edge of an object, with properties of material, color, and texture inherent in that object needing to be conveyed either by properties of the line work or by annotations. A plan is a horizontal view looking down through a building showing, at the site level, landscaping, utilities (e.g., sewage, water, gas), and topography; and at a building level, wall locations, rooms, wall thicknesses, furniture, and service penetrations. An elevation is a view, usually exterior, and perpendicular (90°) to the building, illustrating wall surfaces, window and door shapes and sizes, and roof shape, while a section is a vertical slice through the building showing floors, walls, ceilings, and roof planes. Interior elevations and sections showing cabinetry, windows, doors, and service openings are also used. Construction detail drawings (see Figure 2) are produced to show close-up views of particular parts of the building, often joints or typical sections through walls, floors, roofs, windows, and doors, so that materials and fastenings can be put together by those responsible for fabricating the building, be they in a factory or on-site.

Colored or rendered perspective or axonometric views of a building exterior or interior have been traditionally used to communicate with clients or for marketing a building rather than to instruct those constructing the building. These were the territory of the architectural illustrator who, in the past, worked in watercolor, colored pencils, or markers. In the digital environment, a 3D render fulfills this role and is useful for providing information about materiality, style, height, and footprint. A render does not, however, necessarily convey information about how a building functioned. Because they sometimes lack contextual details, such as signage, neighboring buildings, streetscape, landscaping, traffic, people, or street furniture, it is important to view such records in the light that they were created. Often drawn and rendered to promote or explain the design to the client or interested parties before it was fully documented, these illustrations present only an approximate representation of the final design (see Figures 3 and 4).

As almost all drawings are created prior to a building reaching completion, remembering that they do not necessarily represent the constructed reality is necessary. Sets of drawings marked “as built” or measured drawings, while not always available, are valuable because they most closely approximate the physical building's instruction manual and are produced mainly for facilities management purposes once the building has been commissioned. Site visits to extant buildings remain among the best ways to understand their physical materiality, albeit with the proviso that alterations and additions have occurred to the building's fabric throughout its life. Once a building has been demolished, photographs, measured drawings, videos or films, and three-dimensional scans provide the best record of it. Textual representations of a building such as media reports, historical accounts, memoirs, and building reviews can also help create a picture. A useful record of a building's use is the postoccupancy evaluation, which asks the clients or inhabitants of a building about their experiences of using the building, although these are not common, especially in smaller and domestic structures.

Design record literacy involves understanding how a two-dimensional representation of a building drawing or model conveys shape, texture, three-dimensional form, volume, and space—between, inside, and around the building. It also involves an understanding of construction techniques represented in the drawing—not only materials, but structure, mechanical services, electrical, plumbing, acoustics, thermal performance, water penetration, access, and egress, with statutory building codes involved in all of these. To interpret such information from a drawing and translate drawn lines into an understanding of the building, the architectural graphic standards and conventions used by the architect, draftsperson, or renderer are useful to know. These include being able to identify the north arrow; scale bar; standard representations of walls, doors, stairs, windows, wet areas, and section cut-through lines; and the use of any color coding or hatching (see Figure 5).

In traditional paper-based architectural drawings, proportional scales varied depending upon the information required to be conveyed on a particular drawing sheet, which was limited by its physical size. These varied depending on the country in which the drawings were produced and its use of imperial or metric measurement units. For example, a site plan might be at a scale of 1″ = 40′ imperial or 1:500 metric, a building plan at ¼1/4″ =1′0″ or 1:100, and construction details at 3/32″ =1′0″ or 1:5 scale, depending upon the level of information required by the intended audience for the drawing. These are still relevant when a drawing is to be printed; yet, on screen, a digital scale bar can be the best guide. Many born-digital drawings continue to be printed out for practical purposes such as presentation to clients or for use on-site by builders, although the format of these too has developed from the large blueprints of the past to smaller, more manageable sizes printed in black and white.

To be able to read born-digital design records, understanding some of the historical context for their development and use is helpful. The first CAD software programs were developed concurrently in the 1950s and 1960s at the Massachusetts Institute of Technology (MIT) and in industry at companies like IBM, General Motors, Lockheed, and Boeing.⁷ Sketchpad, developed by Ivan Sutherland (see Figure 6) as part of his PhD thesis at MIT in 1961–1962, is commonly regarded as the first CAD system.⁸ By the 1970s, several two-dimensional (2D) drafting programs were available on the commercial market. Many of these early systems used input devices such as light pens that presaged the advent of graphical user interfaces and have since largely fallen out of favor.⁹ Due to their high cost and reliance on expensive mainframes, early CAD systems were typically developed and used by large companies in fields such as automotive and aerospace engineering and were generally out of reach for most architects.

Computers with sophisticated drafting software began arriving at many architectural firms during the 1980s and 1990s. Taking advantage of advances in computing such as the development of microprocessors and the new accessibility of computers in businesses and homes, companies such as Intergraph, Bentley Systems, SolidWorks, Autodesk, Micro-Control Systems, and Diehl Graphsoft released affordable architecturally focused drafting software for personal computers that found wide market traction through the 1980s and 1990s.¹⁰ Several of these systems, including Bentley's MicroStation and Autodesk's AutoCAD, remain popular software today and have developed 3D modeling capabilities in the years since they were first released (see Figure 7). Their original use, however, was two-dimensional: a translation of the hand drawing into digital space.

Though CAD and, later, BIM software would eventually come to be standard in architectural offices, the shift from analog to digital practice did not take place uniformly, or all at once. The twenty-five projects dating from the 1980s to 2000s collected and studied as part of the Archaeology of the Digital project at the Canadian Centre for Architecture curated by Greg Lynn collectively demonstrate that the use of digital tools has varied significantly between designers, contexts, firms, and schools of architectural education and practice, and has at times supplemented rather than replaced analog architectural media.¹¹ The history of design software is similarly nonlinear. Some features of design software such as parametricism and scripting that are commonly thought of as being recent developments have antecedents in the earliest days of design software—by way of example, Ivan Sutherland's Sketchpad was parametric by design in the early 1960s, and AutoCAD has supported scripting in AutoLISP since the late 1980s, long before the advent of visual scripting and mainstream generative design practices.

One particularly important moment in the history of design records is the emergence of powerful 3D modeling software beginning in the late 1980s. Although developments in 3D modeling began decades earlier, during this period it became more widely available. Pro/Engineer, first released in 1987, brought parametric 3D solid modeling to a commercial market and found widespread use in engineering contexts. In architecture, Autodesk added 3D support to AutoCAD in 1994 and faced competition throughout the 1990s from new 3D-first competitors such as SolidWorks, Softimage, and form·Z, and later Rhinoceros 3D (colloquially known as “Rhino”) and SketchUp. Simultaneously, advancements in software for animation, computer-generated imagery (CGI), and video games pushed the envelope technologically and provided inspiration and new possibilities for recent generations of architects on projects such as Greg Lynn's Embryological House (1997–2001) and Asymptote Architecture's Three Dimensional Trading Floor and Advanced Trading Floor Operations Center for the New York Stock Exchange (1998–2004).¹² These new technologies have changed not only the process of designing and documenting buildings but also the buildings themselves, enabling structures such as Frank Gehry's Bilbao Guggenheim (opened 1997), Walt Disney Concert Hall (2003) (see Figure 8), and Zaha Hadid's Phaeno Science Center (2005) that would have been nearly impossible to design, document, and construct just decades before.

The earliest and simplest 3D models, known as wireframe models, simply added a third dimension to the x-y coordinates of drawing space familiar from a paper environment (see Figure 9). Surface models more closely resemble objects in the real world by approximating curved surfaces, implementing algorithms such as non-uniform rational B-spine (NURBS), which was developed independently in 1958 by Paul de Casteljau at Citroën and in the early 1960s by Pierre Bézier at Renault.¹³Solid models add semantic data such as volume to models, turning them into objects with properties rather than simply collections of connected points, lines, and curves. This additional data allows for increased computational possibilities from solid modeling, which today's CAD and BIM software and related ecosystems of computational tools use for structural analysis, automatic clash detection, simulations of environmental conditions such as wind and sunlight, and similar analyses.

The move from 2D digital drawings to 3D digital models, and solid modeling in particular, significantly changed the practices and possibilities of architectural design, while simultaneously offering a familiar interface to designers who had long used physical three-dimensional models in their work for schematic design, design development, and presentation to clients and other stakeholders. Digital 3D models introduced new possibilities such as generating photorealistic renderings and 2D construction documents on demand; performing complex computational analysis on designs; and automating fabrication and model-making processes through direct fabrication and computer-aided manufacturing.¹⁴ Technological developments in virtual (VR) and augmented reality (AR), sometimes collectively known as extended reality (XR), have likewise affected how building designs are communicated to clients and other stakeholders, as well as the architectural design process itself.¹⁵

The relationship between designer and software tool can be both complex and multidirectional. Matthew Allen's exhibition Tell Me about a Rhino Command. Software and Architectural History, held at the Harvard Graduate School of Design in 2016, explored “the feedback loops that exist between architect-users and their software” by examining the stories implicitly told through the change log of the first six years of Rhino's development.¹⁶ For experienced architects, “software is sometimes visible in the architecture,” meaning that the hallmarks of the software used to develop a project can often be identified by looking at the finished building.¹⁷ Visual programming platforms such as Grasshopper and Dynamo in recent years have given even more control to users of design software and have lowered the barrier to entry for parametric modeling and generative design.

Although most CAD file formats have historically been proprietary, several file formats have been developed over the years to support interoperability and exchange of geometry and related information between CAD systems. Initial Graphics Exchange Specification (IGES), developed in the 1970s and 1980s, is an ASCII-based exchange format for two- and three-dimensional geometry that has historically been a popular format for exchange of architectural geometry but suffered from a lack of standardization in terms of how software vendors applied it.¹⁸ IGES was superseded by the Standard for the Exchange of Product Model Data (STEP), first published in 1994 and standardized as ISO 10303. STEP, defined over nearly six hundred published parts and implemented in a collection of file formats such as AP 203 (sometimes referred to as a “STEP physical file”) and AP 242 (Managed Model-based 3D Engineering), has attempted to address the shortcomings of IGES by including more information and emphasizing testing and compliance.¹⁹ Unlike AutoCAD's more fully featured native DWG (drawing) file format, the specifications for Autodesk's DXF (drawing exchange format) are openly published, making it another CAD exchange format, albeit a lossy one, with support for a much more limited range of potential CAD file data than STEP.

The most recent shift in born-digital design practice is toward the use of BIM. A BIM is a virtual model of a building intended to serve as the definitive base of knowledge for that building from initial design and construction through facilities management, renovation, and demolition. BIM is standardized as ISO 19650, which offers the following definition: “Use of a shared digital representation of a built asset to facilitate design, construction and operation processes to form a reliable basis for decisions.”²⁰ It is commonly said that BIM adds new dimensions such as time (4D), cost (5D), energy/sustainability (6D), and facilities management (7D) to the three geometric dimensions of a 3D model. In theory, BIM offers the capability for all of the stakeholders in a building—for example, architect, landscape architect, engineer, contractor, owner, facilities manager—to use and extend a single shared model, selectively viewing the data necessary for their work.

The central concepts of BIM can be traced back to Charles Eastman's academic work in the 1970s.²¹ Some have argued that ArchiCAD, released in 1987, was the first BIM software available for the personal computer.²² Yet, for most, BIM software emerged as a phenomenon in the 2000s and began to see adoption and standardization in the 2000s and 2010s.

In parametric BIM software such as Revit, objects are both semantically rich and linked, so that changes to an object in the model will automatically propagate to any related objects. Changes made to the length or height of a wall, for example, are automatically reflected in any adjoining or otherwise related walls, windows, doors, ceilings, floors, electrical fixtures, HVAC (heating, ventilation, air conditioning) ducting, and so on. Objects within a BIM are often pulled from linked “family files,” which are libraries of objects that can be used within a BIM. These family files may be developed locally, such as sets of objects that are commonly reused within a project or firm, or pulled in from external sources such as online product catalogs.

In many ways, BIM can be seen as the realization of the dream of Douglas Englebart, the computer scientist of “mother of all demos” fame, expressed in his 1962 paper “Augmenting Human Intellect”:

As BIM matured, competing BIM products such as Gehry Technology's Digital Project also emerged, and BIM standards have developed and been published in locations around the world including the United States, the United Kingdom, Australia, Canada, Belgium, New Zealand, Norway, Germany, Singapore, Hong Kong, and Finland.²⁵ Many of these national standards are based on the concepts of openBIM articulated and refined by buildingSMART, an open organization composed largely of members from BIM software vendors. buildingSMART has led the development, specification, and publication of the Industry Foundation Classes (IFC), an open and interoperable object-based file format for BIM standardized as ISO 16739-1:2018. IFC enjoys wide support in modern BIM software systems, and its use has been legally required for publicly funded building projects in several countries, including Norway, Denmark, Finland, and the United Kingdom. In the United States, the General Services Administration (GSA) mandates the use of BIM for all building projects under its mandate, although IFC specifically is not required.²⁶

The history of the development and use of born-digital design records can be helpful in understanding these records, both from a technical as well as a contextual point of view. Visual literacy allows us to explore the content of this type of born-digital design record.

Visual literacy has been described in its simplest form as “the ability to recognize and understand ideas conveyed through visible actions or images.”²⁷ More specifically, the Association of College and Research Libraries writes that visual literacy

It is worth emphasizing that visual literacy involves more than simply the evaluation of the aesthetics of a record. Embedded within any visual image are historic, scientific, technical, social, and cultural values, all of which contribute to the value and potential uses of the record. As outlined in the Library of Congress's Prints and Photographs Division's “Every Photo Is a Story”²⁹ video series and exercises, the visual literacy competencies of creators are often similar to the skills that researchers need to appreciate, understand, and contextualize what they are seeing.

Visual literacy competencies are a core feature of the Information Competencies for Students in Design Disciplines published in 2006 and updated in 2007,³⁰ as well as of ARLIS/NA Core Competencies for Art Information Professionals,³¹ both produced by the Art Libraries Society of North America (ARLIS/NA). Discipline-specific work has been further developed for architectural history, architecture, art history, fashion design, studio art (June 2018); graphic design, interior design, photography, urban and regional planning (June 2019); and landscape architecture (May 2020).³² These reports demonstrate the importance of developing visual literacy skills for creators, scholars, and individuals interested in interpreting the design process and end products of designers, as well as information professionals who assist in those pursuits. They also demonstrate that each discipline has both shared and separate competencies that may need to be developed by creators, scholars, and information professionals.

Born-digital visual records, which are created using digital tools (see Figure 11), may have embedded descriptive, structural, and/or administrative metadata that reflect the designer's decision-making in selecting specific software for creating particular files, accounting for both the technical functionality of the digital tool and the communication of the design intent. It is important to document these additional layers, as compared to digitized records, which have been created on paper and scanned or photographed to be saved as raster image files.³³ While some work has been done to develop a comparable framework around digital visual literacy,³⁴ that work does not address the unique features and context of born-digital design records, particularly those related to the built environment.

Understanding user-driven demand for architectural and design records may be pertinent in relation to born-digital records, as both visual literacy and digital literacy may be sources of bias and currently have an impact on the users of these files. With the exception of a 2016 survey of users of born-digital archives at the Canadian Centre for Architecture (CCA), scant research exists on the researchers of born-digital design archives, their research questions, skill sets, and experiences.³⁵ Martien de Vletter of CCA has found that those requesting the CCA's born-digital architectural files are largely “researchers from the digital humanities or media studies (rather than architectural historians) . . . [whose] research seems to have a more quantitative focus.”³⁶ Possibly, however, with the passing of time, those who might be requesting files will change in profile, as digital literacy grows within the architectural history, heritage preservation, and historical research communities.³⁷

Architectural historians and built heritage professionals, as traditional user groups of paper-based architectural archives, are often interested in the history of the building rather than in the drawing or model as an object in itself. These users most likely are researching when and for whom a building was constructed, when alterations were made, and the functions of the building over time. These questions differ from those of a builder or engineer charged with adding an extension to an existing building, who might need to know the depth of foundations, the strength of steel, the materials specified, soil test results, location of utility infrastructure (e.g., power, heating, cooling), or the structural loads on the building. These different needs show how users can look for varying types of information in the same set of primary records.

Some users of design records may also be interested in projects that were never built, such as planned projects stopped short due to economic or other reasons, unsuccessful competition entries, or student projects whose authors' main aim was to learn by exploring the design and drafting process (see Figure 12). Other works in this arena might include theoretical exercises and explorations of fledgling ideas as architectural prototypes and feasibility studies. Such works can be quite useful in the study of architecture as a field, for example in studying how practice and intellectual understandings of architecture change over time, and they may serve as inspiration to practitioners for future projects.

Due to a myriad of reasons ranging from limited access to niche software to the continued use of paper-based records by design firms creating hybrid analog/digital project records, questions persist around the future value of born-digital architectural records. The valuing of traditional paper-based architectural drawings based on their artistic merit is one of the reasons such drawings survived. While recognizing the artistic beauty and value of physical drawings, especially those undertaken with the skill of a practiced hand, is easy, it remains to be seen whether similar qualities will be appreciated in computer-mediated drawings. Whether born-digital files will ever be viewed as digital art, which has become a recognized genre of the contemporary fine art world, is as yet unknown. We are still waiting to see if digital renders could become artifacts to be appreciated and hung prominently on a wall, public or private, either as digital objects on screen or as printed graphics. Such questions point to an unknown future for these artifacts and one that warrants further research.

Archivist Terry Cook, in his 1997 article reviewing a century of archival practice, states that “the principal justification for archives to most users and to the public at large rests on archives being able to offer citizens a sense of identity, locality, history, culture, and personal and collective memory.”³⁸ Architectural documents have the potential to provide a record of the crystallization of society's many identities, and the social and cultural systems on which they were based, in built form—demonstrating how social boundaries became walls, organizational groupings were housed in rooms, power relations were built into plans, and human well-being was affected by spatial design. And, while buildings and their records hold the potential to act as powerful reminders of past times and societies, they can also offer the possibility of impacting not only current behaviors but also conceptual systems.³⁹ Specific types of buildings can be seen to act as containers for determined functions; as an example, the built form of a jail can provide tangible evidence of a specific judicial and penal system through its walls. Thus, the built environment and its records can be seen to act as a keeper and modifier of social, political, and ideological standards, guiding activity patterns and setting precedents for future use. The significance of the role buildings play has been recognized by psychologist Mihaly Csikszentmihalyi, who sees them as holding the potential to “give a permanent shape to our views of ourselves that otherwise would quickly dissolve in the flux of consciousness,”⁴⁰ thus providing something tangible to which society can attach meanings, both individually and as communities.⁴¹ By extension, the records of the built environment play an important role, as they hold these histories and identities when the bricks and mortar have been lost or destroyed.

Buildings and their records also have the ability to arouse curiosity within users about the layered histories behind their facades. These reminders are present in both monumental and everyday places from statues to hospitals—although not all histories of use are remembered or leave traces. Significantly, geographer Doreen Massey recognizes some of the imbalances present in history, stating that “The identity of places is very much bound up with the histories which are told of them, how those histories are told, and which history turns out to be dominant.”⁴² Design records add to these histories, as they help tell a place's full story. Design decisions documented and intent articulated provide the necessary context to understand why and how our environment is what it is today, as well as what its potential future may have been. However, the risk of selection bias and historical imbalance applies equally to the records of places as it does to their built preservation, with the potential to either prioritize or erase them from the cultural record intentionally or by neglect. This leads to the next important consideration.

Archivists, it has been argued, “are literally co-creating archives” as “appraisal . . . starkly determines which documents are destroyed, excluded from archives, their creators forgotten, effaced from memory.”⁴³ Any archival assessment must be recognized as ideologically and culturally driven, rather than as an objective way of ranking and assigning value to items. Additional aspects that archivists must therefore understand and consider are the visual and digital literacy of parties involved in decision-making around record retention, which can unintentionally have a profound impact on the retention or loss of records, as can the stage of their life at which they are retained. Additional research is needed in this area to better account for the myriad factors impacting the collecting, processing, preservation, and access provisions of the wide range of archival institutions that have or soon will be addressing collections of born-digital design records.⁴⁴

Jakob Beetz et al. in 2013 developed a sawtooth curve of knowledge loss in building and construction that shows how knowledge about a building diminishes over time.⁴⁵ They found the drop-off once a building is constructed to be significant. They demonstrated that the best time to capture knowledge about a building is immediately after it has been built and while it is in use, and most certainly before its decline toward demolition.⁴⁶ This is particularly pertinent for born-digital architectural records, which, like all digital records, are in danger of becoming lost if not intentionally selected for preservation. They are especially prone to software and file format obsolescence, hardware protections, and severe licensing restrictions, which collectively make digital preservation challenging for older born-digital design records.⁴⁷ Information about authorship, provenance, ownership, and rights, especially when the records have been produced collaboratively, is also easier to capture while staff are still employed or traceable. Current custodians need to ensure colleagues or archivists are aware of the existence of records in digital format and develop strategies to care for them, lest invaluable data become virtually irretrievable or their very existence forgotten.

Traditionally, within a paper-based archive of architectural records, acquisition would occur at the retirement or death of an architect or the winding down of a practice. However, if, as Beetz et al. warn, archives acquire them within a few years of creation to avoid knowledge loss, deciding what to prioritize as soon as possible becomes salient.⁴⁸ In 2003, UNESCO stated with regard to digital heritage that “the main criteria for deciding what digital materials to keep would be their significance and lasting cultural, scientific, evidential or other value. [noting that] ‘Born digital' materials should clearly be given priority.”⁴⁹ These values include their worth as evidence and containers of information; their artistic or aesthetic significance; their ability to demonstrate innovation, historical, or cultural associations; their usefulness to potential users, and their culturally significant characteristics.⁵⁰ By 2009, such files were increasingly recognized as heritage at risk of being lost, and, in its “Charter on the Preservation of Digital Heritage,” UNESCO again emphasized that “these resources . . . constitute a heritage that should be protected and preserved for current and future generations.”⁵¹

The ability to read born-digital design records, like any other skill, is best learned incrementally and through practice, and this approach can be applied to better understanding records across all archives, from large specialist architectural collections to smaller local history or organizational archives. Although many archives' holdings currently do not include born-digital design records, free software readers and sample files can be found on the websites of several CAD software vendors.⁵² These sample files are often created expressly for the purpose of teaching and can make excellent learning materials. We hope that the following visual literacy approach to born-digital design records will demystify these records and empower archivists to feel more confident in appraising, describing, preserving, and providing access to them. The process for achieving increased visual literacy with regard to born-digital design records begins before a file is opened and moves through opening the file, viewing the screen, exploring the content, and understanding its meaning. At each of these steps, the visual cues must be read to get the most out of the files in question. The approach is summarized in chart format (see Table 1) and illustrated in the sections to follow with relevant examples.

Before opening any files, proper appraisal of records of all formats that may be donated together is crucial, as many of these records can provide considerable insight into a project. In many archives, derivative versions of born-digital design files in commonly accessible image and video formats may already exist. These may be very useful for exhibition, as well as for access by users whose research questions may not require, or whose skill sets may not support, operating CAD and BIM software themselves. Additionally, paper records related to the project may exist, including rough sketches, handwritten notes, designers' notebooks, and physical models. For current projects, many records may be stored in cloud-based subscription services and project management tools rather than on local file servers and may need to be exported as part of the transfer process.

Identifying a design file's file format will provide initial clues into what type of information will be rendered when the file is opened. Identification can be as simple as checking the file's extension (e.g., .dwg, .3dm, .skp, .rvt, etc.) or can be done more robustly by using one of the wide range of free and open-source format identification and profiling software utilities such as Siegfried, DROID, Brunnhilde, fido, Apache Tika, and the Linux file utility.⁵³ These programs identify file formats by comparing the contents of a file's header and/or footer against known byte signatures (also known as “magic numbers”) collected in file format registries such as the National Archives' (UK) PRONOM. This approach provides more precise and accurate identification than relying on file extensions, which can be arbitrarily changed and are not required in all operating systems. DROID and Brunnhilde (which uses Siegfried for format identification) in particular can create high-level human-readable reports with aggregate statistics for entire directories, drives, and disk images (see Figure 14).

In addition to identifying a born-digital design file's file format, it may be possible and desirable to extract technical metadata from CAD and BIM files, a process commonly referred to in digital preservation practice as characterization. Although characterization tools for CAD and BIM formats have been few and limited to date, Micky Lindlar of Germany's Technische Informationsbibliothek (TIB) has presented work on the possibilities of using existing open-source tools developed through the EU-funded DURAARK (Durable Architectural Knowledge) project to extract useful metadata from IFC BIM and E57 files.⁵⁴

The file format and originating software of a born-digital drawing or model can offer valuable insight into its intent, function, and historical context. Although designers do not use software in identical ways, some established practices and understandings of what certain software types are designed to do exist. For instance, AutoCAD (.dwg file extension) is predominantly a two-dimensional drawing software that can and has been designed to be used throughout the life of a project from conceptual design through construction drawings; while SketchUp (.skp file extension) is used for a wide range of three-dimensional modeling activities, from video game design to animation, and is commonly used by the built design field to explore conceptual design layouts with less emphasis on measurements or specifications. Stereolithography files (.stl file extension) are often used for 3D printing and CNC (computer numerically controlled) routing, where simplified vector-based representation of a model's geometry enables machine fabrication of components for an object or building. Knowing this information before opening the file garners the viewer valuable context to interpret the file's content.

Table 2, although far from comprehensive, attempts to provide one such high-level overview of commercial design software since the 1970s, indicating prevalent design software by discipline for each subsequent decade.

Upon opening a file, archivists should pay attention to what, if any, error messages or prompting screens appear. These are critical clues to whether additional applications or files are necessary to view the file as it was intended to be seen. An externally referenced file (xref) might need to be reattached to the file being opened to see the topography of the site, the texture of a surface, a title block, or other supporting information. Other errors might indicate the special font originally used for the file isn't available or display a notice about the compatibility of the file with the software it was opened within. In many cases, such messages might be resolvable after research. Where error messages are not resolvable, they should be noted and taken into consideration during appraisal and description.

After working through the initial prompts to reach the main viewing screen, archivists should aim to describe what can be seen by identifying written, drawn, 3D-modeled, familiar, and unfamiliar elements that are visible (see Figure 15).

This can be at a granular level:

• Are there several color lines on the screen?

• Are there three-dimensional objects?

• Is there any legible text?

Or at a more general level:

• What information is being communicated?

• Who could be the intended audience (e.g., the contractor or client) of this information?

• Are building, landscape, or wayfinding components identifiable?

Looking for familiar design elements can help an archivist begin to unpack the information being communicated in the digital file. Associating the new and unfamiliar digital file with the (potentially) familiar records created in a paper-based design workflow can be safe as long as it does not become limiting. While the majority of born-digital design files do have a close relationship to the paper-based design processes, some file types or visual representations (e.g., visual scripting or virtual reality simulations) do not have an obvious or direct connection with the more established design record types, such as those outlined in Architectural Records: Managing Design and Construction Records.⁵⁵

Continuing to look closely at the representation on-screen, archivists should notice perspective (e.g., aerial, ground level, angled), level of detail (e.g., general blocks known as massing, textured surfaces like brick or paneling), and indicators or information keys, as well as relationships between components. This information can help determine whether the content is a plan, section, elevation, or more robust 3D model.

This close viewing makes it possible to identify clues worth further exploration, such as when a title block is present on a virtual drawing sheet. The title block is a standard architectural drawing labeling system, typically found in the bottom right corner or along the entire right edge of a printed sheet. Software such as AutoCAD, Revit, and MicroStation offer a “sheet view” that provides the space for components such as company name and address, drawing description (e.g., first floor plan), revision number, drawing author's initials, scale of the drawing, date of the drawing, and a few other pieces of data (see Figure 16). Reviewing this standardized information can provide additional context about intention of the drawing or how this iteration fits into the larger project's evolution.

After identifying a file's format and the software that can render the file, archivists should look closer at the navigation bars and the functionality of the software. Are there tools to change the view of the digital object on-screen or to select and deselect layers? Layers are the digital equivalent of transparencies being layered atop a drawing and will be familiar to those who have worked with image editing software such as Photoshop or GIMP. Architects and other designers use layers to organize parts of a drawing by function or purpose, as well as to select which elements of a drawing are visible at one time. The layer panel in most CAD and BIM software enables users to set the visibility of layers, to lock layers to prevent further editing, and to change the display color of elements in the layer. Layer colors, which are typically very bright neon shades, are often though not always intentional—for example, layers related to mechanical engineering may be grouped together as one color, while topographical site plan lines are grouped as a second color, architectural elements such as walls are grouped as a third, and so on. Layer names may be standardized at a project-wide or a firmwide level, be entirely at the discretion of the designer, or, most typically, some mix of the two (see Figure 17).

When first opening a digital drawing or model, checking the layers before deciding that the file is devoid of content is wise. Architects commonly save complex 3D models with many layers turned off in an attempt to speed up loading times for their models on CAD workstations, which are frequently pushed to the limits of their hardware.

Understanding the content of the file involves understanding both the file format as well as the structure that the drawing or model communicates. It is helpful to understand who made the file and when, where, how (what software), and why (the intended use of the drawing/model). Once these basics have been established, archivists should endeavor to determine information about the building itself: its location, name, client, functions, and size.

Additional clues valuable to understanding the file's content can be found by examining the external context of the file and software. Contextualizing research is not a foreign concept to the archival field. Depending on the level of mitigation and support being attempted to contextualize born-digital design files, archivists may need to understand how the software works and how the designers intended to use that software, as well as how to read plans, sections, elevations, and details, and to navigate within a 3D model. While this may sound intimidating, it can be achieved mostly through traditional research methods applied to a broader range of information sources.

Due to the continuous development of born-digital design software, industry resources for the technological community as well as for the architecture and design communities can be leveraged to provide insights into a software's use, the influence of professional standards, and how these resources and work-flows are discussed within the design profession. Websites of software vendors (e.g., Autodesk or Rhinoceros 3D) and professional organizations (e.g., the American Institute of Architects or buildingSMART) offer useful insights about current and previously promoted practices. Archivists can leverage resources that designers use themselves, such as YouTube tutorials, to learn about a feature or function of particular software that may be necessary to navigate and identify all parts of a design file. As further discussed in the next section, asking the records creators about their preferred reference resources enables archivists to build a common understanding with those creators/designers about the functionality and anticipated uses of various software and the types of files they create that are present in their collections.

Information published in book and periodical formats can become dated more easily and should be carefully assessed for its applicability to the time period when born-digital design files were created. We have provided some useful resources in the notes as starting points for beginning one's personal exploration into born-digital design records.⁵⁶

In addition to exploring the software used for record creation, contextualizing the project and understanding more about the designer(s) involved are also important. Exploring the creator's website, identifying files they've provided that offer a narrative of the project, and looking more broadly for articles and promotional materials about the proposed or completed project by the client and various press outlets, from mainstream newspapers to niche design-focused publications, may be helpful.

Archivists need to contextualize their collecting policies by appreciating each institution's collecting scope and designated community of users. Part of this contextualization is remaining aware of and supporting the collecting practices of institutions with similar scopes rather than competing for collections. This holistic professional community collecting effort has the potential to better document a wider range of designers and the global built environment. The Design Records Section of the Society of American Archivists has developed an ever-improving map to provide a comprehensive understanding of where design records are being collected throughout the United States.⁵⁷ This is both a recommendation for archivists to be aware of other collections and a reminder to connect with the professional community.

Using design records to prompt questions through a close examination of the information readily evident on a drawing is one way of beginning to interrogate a record. Information that may be gleaned could include identifying who designed the building and made the file, for what purpose, and when; and determining the location of the building, its function or use, its size, the materials it was constructed of, and its aesthetic style.

Much of the basic information can be found in a well-filled-out title block on a drawing sheet, which should (but won't always) list the architect's or designer's names. This can act as a prompt to begin searching for information about the practice and could involve contacting a professional organization, looking for registration lists, checking telephone directories, searching out company histories, and undertaking online searches. Unlike artworks or written works, the design and authorship of buildings is often a team enterprise, with the practice as a whole taking credit. However, in a practice named after one of its directors, that individual often has all works attributed to them personally even if it was actually a collaboration. This is particularly the case in multinational practices where an individual architect's work carried out in a smaller branch office can become obscured. Assigning design authorship to the practice's namesake does not tell the whole story, so it is best to capture such information about project teams while it is still in the living memory of those involved. At times, the initials of the design and drafting personnel are captured in the title block, giving a clue to who worked on the job but not necessarily their role.

While access to the individuals responsible for the design can be difficult to gain, the best time to capture staff details is during negotiation and acquisition of donations into a collection. In the case of active practices, the administrative staff may hold such information in document format. As archives are rarely donated while current, interviews or oral histories post hoc may be another option, albeit an expensive and time-consuming one. Architectural staff members may recount personal anecdotes as well as provide sources of further information, including reviews of the building, citations, or awards, all of which are useful to researchers. The capture of these associated materials, be they paper-based or born digital, should also be considered. Oral histories are a valuable tool, not only for historians but also for archivists, and can be useful in recovering files.⁵⁸ Interviews can be made part of the accessioning workflow to ensure the donor provides nuanced insight about their design process, which can be supplemented with written documentation when donors are willing and able. An approach taken by some design archives is to ask that donors complete a written questionnaire to be included with any deposits, which asks for information such as who worked on a given set of projects, which software was used during the different stages of the design process, and what were the file-naming conventions and passwords, along with explaining the firm's computing infrastructure.⁵⁹

To understand why records were created in the first place, archivists should first ask for context from the records creators. Within an architecture practice, depending on its size, the team members who worked on a particular project may include the designers of the building, the drafters, and interior or landscape specialists, often under supervision of a senior architect or director. External to the architectural practice are allied professionals such as structural engineers; mechanical engineers; acoustic specialists; ventilation specialists; fire systems managers; surveyors; and fabricators of components, fixtures, and fittings, all of whom may be creating particular drawings of the same project for their own purposes. Statutory authorities, be they councils, planning officials, building certifiers, or heritage professionals, will also examine and, at times, request changes to drawings or documents after they have been submitted. Amendments to drawings are usually indicated within the title block, and checking for dates is critical to understanding which version of the drawing is being examined.

Fellow archivists, librarians, and gallery and museum professionals may be able to assist in finding materials to contextualize buildings and designers as well as pointing to sources of further information, be they technical, aesthetic, cultural, or social. The street address given on a drawing should point to the site on which it was located. Local historians familiar with the place are valuable, and local history collections often hold information about the reception and use of buildings in their area. Local chapters of architectural professional organizations such as the American Institute of Architects are another avenue to pursue, as they not only have contacts with the active professionals but also with retired architects whose living memory can prove helpful and who are often willing to share their knowledge. Additionally, if a building was nominated or awarded a prize, professional organizations may hold records, photographs, or building citations.

Understanding a building requires reading the visual elements on the drawings as well as the text. Much can be gleaned about the intended uses of the building by looking particularly at the floor plans. Examination of room names and the tracing of imagined users' paths through the building are often enlightening. For example, a plan containing rooms labeled “bedroom,” “kitchen,” “ensuite,” or “living” (see Figure 18) could point to domestic activities, but not only homes and apartment blocks contain such rooms. Alternative building types could be hotels, resorts, or health-care settings, pointing to the need to look for further contextual clues in the drawing sets. The building name on the drawing in such a case could provide clarification. Alternatively, it may be necessary to look at other drawings of the building within the same set. Does the elevation give any clues through its style or signage? Is there a common entrance to the building? Is it possible to zoom out to see more of the surrounding features on the site? Does the site plan set it among or nearby a complex of other facilities? Do other floors contain rooms that give clues to the function? If the drawings cannot confirm a building's function, then looking for other sources, such as business directories, advertising, or local council records, using the address provided on the drawing may become necessary to confirm its use or function.

It is not only documents within a set of records that can reveal a building's history. Looking deeper, as a researcher or historian would, archivists will find clues within a drawing. Assessing architectural drawings and models for cultural dimensions, including those that are historic, aesthetic, technical, and social, can also improve their readability.⁶⁰ For instance, Figure 11 (page 296), the Jeffrey Smart Building at the University of South Australia, by John Wardle Architects with Phillips Pilkington Architects and Wilson Architects, 2015, shows a multistory building with three wings around a courtyard and with a street frontage. Other drawings of the building show that it contains seminar rooms, book shelving, audiovisual screens, tiered seating, and food service areas, indicating that educational or entertainment functions very likely occur there. By counting the seats in the plan of the building, the number of people expected to use the building can be inferred. Examining the audiovisual technology and the stage setup on plan, interior section, and 3D renders can reveal what type of programs are presented to the audience, for instance illustrated lectures or film screenings. When taken with the text on the drawings, all these types of information can lead to the identification of a building's function and era—in this case, a university learning center and library with a public forum space and cafe.

Importantly, the type of detail found on born-digital architectural drawings and models gives us clues about where to look for further information, such as advertisements for public talks in print or digital media, reports of visiting speakers, or reviews of films. When a building is part of a larger organization, then institutional archives may also hold records of uses and events in such buildings, perhaps lecture schedules or opening nights. As archivists become more familiar with design records, their usefulness as records becomes more evident. We begin to see them not only as attractive illustrations, but as testaments with clues to the histories of places and the people who used them, leading to a deeper understanding of the multiple social and cultural identities of societies. Through such analysis, the significance and case for preservation and continued accessibility of visual records such as architectural drawings can be made.

In this article, we have sought to lower the barriers to understanding, appraising, describing, and providing access to born-digital design records. We first introduced some of the key historical and domain knowledge required to understand the context and content of born-digital design records. We then discussed potential users and uses of born-digital design records to help archivists understand the research questions that may be applied to such records and the many layers of embedded information and value to be found within them. Finally, we introduced the step-by-step “Visual Literacy Approach to Reading Born-Digital Design Records,” depicted in Table 1 (page 302), and walked the reader through its application to digital design records.

We hope that the knowledge and guidance imparted in this article and the “Visual Literacy Approach to Reading Born-Digital Design Records” as a stand-alone resource will empower archivists to move beyond trepidation or uncertainty and into active engagement with the contents of born-digital design files. While patience is critical for overcoming the technological hurdles, these files pose opportunities for developing new skill sets, and archivists' observational skills and a willingness to learn are the only prerequisites to getting started. Building these skills on individual and organizational levels will be necessary to ensure that archives collect and preserve born-digital design records in a timely and systematic manner, so that records of architectural practice and the built environment will continue to be preserved, described, and available for research even as design practitioners continue to embrace new technologies in their work.