There are two senses of a critical dimension at play, and both are implied when we use the term. These meanings can be regarded as internal to and external to the digital. By internal we refer to faculties of analysis and judgement as applied to the content, usage and artefacts of the technology. The external meaning relates to a position regarding the development, effects and social relations bound in technology. This position is more associated with historical and cultural analyses and operates in a wider field of technology than computers. It is concerned particularly with how meaning is constructed, by whom and for what purposes and in this sense representational and communicational modes are central. Tools of enquiry from this perspective include textual practices such as semiotic and discourse analysis. The elaborated idea of a text as an organised collection of representations brings to bear the relevance of literacy practices to describe, teach and think about such meanings; overarching this are the discourses of power around technology that can further be seen as texts (Dowling 1991). The perspectives of media education (Buckingham 2006; Livingstone 2004) are also major contributions to our understanding of a critical digital literacy.

This work was located in a national project on digital literacy (http://www.jisc.ac.uk/whatwedo/programmes/elearning/developingdigitalliteracies/DLinTransition.aspx), part of which required curriculum mapping of digital literacies. However, trying to develop methods to do this repeatedly resulted in a drift towards functional approaches (what tools and what skills) and incoherent variability (disciplinary-specific uses with no unifying concepts); nor did they address the issues of engagement, learner need and contested scope outlined previously. The perspectives of critical literacy offered a relevant approach, however required contextualisation for the digital. The critical digital literacy framework outlined below is the outcome of that work. It has been pilot tested in workshop form with a number of constituencies (academics, students, learning technologists, managers) to gauge (1) its acceptability and (2) its use in activity design. It has been positively evaluated so far (https://sites.google.com/site/dlframework/toolsandsupport/exploringdl/workshop-evaluation). Further work is in progress to produce support resources, disciplinary exemplars and course or programme-level tools.

One of the purposes served by the framework is scoping. The importance of this scoping activity lies in its role as a conduit to shared meaning. Given the many interpretations of digital literacy, developing a common understanding of its dimensions is a prerequisite to effective institutional embedding and curriculum integration. By delineating a range of filters through which digital literacies can be examined, taught or understood, a number of perspectives and angles can be considered.

Code breaking resources or coding practice involves the ability both to decipher and produce (encode) texts at a practical level. This does not imply being restricted to the physical, however, and includes an understanding of the grammars, functions and structures of texts. Typically cited examples of code breaking relate to the alphabet, phonics, syntax, spelling, punctuation and vocabulary (e.g. Ludwig 2003). These elements do not transfer directly to the digital but they do have their equivalents.

Digital texts often carry within them the conventions of pre-existing textual practice such as a common use or reference to pagination, both deriving from the practicalities of interplay between digital and analogue texts (e.g. printing a document) but also to take advantage of existing conceptual frames of reference as in the use of ‘web page’, ‘file’ and ‘folder’ (Evans 2013). Alongside practices that have been adapted from or developed out of existing ones, there are new and emerging forms reflecting the digital context. ‘Texting’ involves new forms of vocabulary and grammar as well as new practices relating to time and space (status updates and conversational text, for example). Digital texts are, or have become, predominantly multi-modal (e.g. graphical user interfaces are the norm, context sensitivity is driven by ‘spatial’ visual cues – i.e. the position of a cursor determines the result of a mouse-click). There are also new operations for creating, navigating and interacting with digital texts, for example ‘swiping’ or gesture-based and voice controls. Issues of data and identity security, along with possibilities for material corruption of texts, become part of the process of reading and understanding the narratives within these technologies. As a result, the resource learners need for decoding and making meaning from multi-modal texts have an increased scope. This includes new terminology and sign systems; presentational conventions and stylistic options; navigational mechanisms; and operational concepts and protocols such as the distinction between ‘save’ and ‘save as’.

An overview of the decoding resource and its characteristic dimensions is provided in Figure 1.

The text-participating resources (otherwise referred to as ‘semantic practice’ or by the ‘meaning maker’ role) are concerned with the reflexive processes underpinning comprehension and composition of texts. It focuses on understanding and interpreting narratives within the content rather than those around authorial standpoints.

Meaning making practices in digital texts are more likely to be multi-modal (Kress 2003). The existing purposes and conventions of digital texts are not always rigidly defined and can be subject to rapid metamorphosis or splintering; reading and creation practices may therefore be similarly fluid.

The spread of the internet across areas of textual practice has meant that hypertextuality and network texts have become common, perhaps the norm. By ‘network texts’, we mean texts that are compounded of other texts, the component texts being simultaneously available both in their own right and as components of yet further texts. Working on or with such texts requires multiple concurrent levels of meaning making, constructing or interpreting narratives both within the component texts and at the network level. Furthermore, internet and ‘Web2.0’ practices can create multi-authored, rhizomatically developed texts in which there may not be a clear, internal consistency to the narratives. Not only may there be conflicting narratives within the same text, but these may fluctuate over time; the text and its narratives become more fluid (Bayne 2006; Pachler and Daly 2009). Given the developing perspectives that the reader may bring on different occasions, this creates a dual fluidity between reader and text and makes it necessary to approach texts with the understanding that this fluidity may be a characteristic of their form; the material of the text and consequently its potential meanings may have changed since the last reading (Kress 2003; Lurie 2003).

Possibly novel requirements in digital contexts are: the application of meaning making practices to online interpersonal interactions; and reading and expressing meaning through participation in communal activities. Interpersonal relationships that have been initiated and are predominantly maintained in a digital environment may exhibit similar characteristics to fictional narratives, leading to a blurred distinction between people and characters in terms of the affective resources brought to bear and to quasi-fictional relationships (Suler 2004). In consequence, it becomes more important to maintain an awareness of the underlying material situation when considering the import and impact of narratives; moral and ethical judgement around tone, voice and context may take on an enlarged role. Such considerations need to be understood by teachers in developing their students’ digital literacies.

An overview of the meaning making resource and its characteristic dimensions is provided in Figure 2.

The text using resources are concerned with the appropriate use of specific text forms for given purposes and involves understanding both the particular characteristics of different texts and the particular requirements of different contexts of use.

A number of practices have shifted from being specialist, often institutionally controlled operations to being within the scope of groups and individuals. Examples are the functions of selection and curation of materials, printing, typing, typesetting and image or video editing. This could be seen both as empowering users, who gain more control over the timescales and nature of their production, but also as burdening them with increased responsibility for its management and an increased range of vocabularies and sign systems of which understanding is expected as normal practice. In the example of typesetting, these practices have also moved from being applied to a text as a secondary process of creation to being much more intimately bound up with the writing process itself (Yancey 2004). Presentational modes such as emphasis (bold, italics, colour) and a variety of fonts are now available within a single operational domain (word-processor or email programme, for example) allowing them to be rapidly applied and removed, for design and layout to be composed simultaneously with the written narrative, leading to a more ad hoc, prototyping process of text production. Thus, design, production and dissemination of digital texts require a wide repertoire of skills and critical evaluation of choices and their implications.

The expansion of available existing texts (a result of increased efficiency and connectivity of digital storage) raises the probability of both encountering and using direct repurposing, sampling and recombination practices. This increases the value and risks attached to selection, blending and curation. As well as the number of texts, the variety of text forms that require reading and, increasingly, writing have proliferated in recent decades. In addition to more established media modes (written text, static and dynamic images, sound, moving image) there is also the game, the ‘twitterverse’, the videodiary, the several forms of website; these have their own codes and conventions that impact on how one might understand, engage with or produce them. The vast range of tools (including apps), modalities and forms enable multiple potential solutions to problems but require understanding of their functions, constraints and advantages for effective application. Additionally, the informational scale of the internet, databases and specialist datasets require highly developed search, locationing and query formulation skills.

The novelty and rapid change in the development of digital technologies means that new and unexplored purposes, potentials and problems seem likely to become available at a similarly rapid rate. The existing purposes and conventions of digital texts are not always rigidly defined and can be subject to rapid metamorphosis or splintering; reading and creation practices may therefore be similarly fluid, with individuals contributing to the establishment of conventions through challenge, experimentation and consolidation of practice norms. There are thus both practical and creative aspects to being a text user in digital contexts.

An overview of the using resource and its characteristic dimensions is provided in Figure 3.

Issues of provenance are complex when authorship is mediated digitally. Bayne (2006) argues that individual authorship is deprivileged by collective forms such as wikis, whilst tools such as RSS feeds, and the potential of anonymity (e.g. in chat rooms, online ‘handles’ or avatars) all serve to distance the author/s from the text/s. Critical evaluation becomes central in these contexts and learners may require the decoding skills referred to previously to trace and track provenance. They will also need to discern the ideology inherent in a source (Buckingham 2006). The submerging of qualitative and quantitative methods underlying aggregated or processed data (including search engines, ratings, databases) may need to be exposed, for example. Developments such as ‘liking’ and ‘trending’ offer new propositions in relation both to the form and quality of evidence they present. Many developments in digital technologies are perceived in terms of challenges to traditional academic value hierarchies (Bayne 2006) and learners will need to develop their own values as researchers in this context.

The increase in access to publishing opportunities accompanying the development of the worldwide web means that individuals may be more directly and personally confronted with legal, ethical and moral decision making in relation to how, where and to whom they should publish. This requires a careful reading of the systems and processes available to them, both technical and social. For example, since personal profiles and online identities function as texts, we may now be dealing much more with consequences relating to self-image and public standing (Aula 2010).

An overview of the analysing resource and its characteristic dimensions is provided in Figure 4.

One of the challenges presented by new technologies, and social media in particular, lies in the presentation of self. The disembodiment of virtual spaces can render a decontextualised identity whilst, paradoxically, choice in presentation, locality and juxtaposition act to recontextualise, and hence change, that identity. The extent to which that presentation is managed (e.g. in terms of language, image, content) and control over those choices is exerted (e.g. privacy settings, placement, disclaimers) are new skillsets. The multiple potential audiences of friends, colleagues, employers, clients or markets and the variation of profile which may therefore be required necessitates understanding of how to manage both the varied, disaggregated texts which may go to form such profiles, and the characteristics and security features of the environments in which they may operate (‘impression management’, boyd and Ellison 2007). These aspects are qualitatively different from traditional texts; although more present in conventional media it has not generally been the case that a majority of students would need educational support to manage self-presentation on radio or television; nor would proliferation and potential alteration be so rapid and possibly uncontrollable. The multiplicity of avenues to publicity [‘masspersonal messages’ (Walther et al. 2011, p. 34)] and the varied and globalised communication channels such as formal discussion groups, personal blogs, microblogs, videologs, and so on also requires choice and judgement about how, where and when to participate and for what purposes. These factors have a prominence in the digital which is not reflected in the four resources model. One may debate the inclusion of these practices within the scope of literacy; however, as they cannot be disengaged from the practices of reading and writing in and for the digital environment, and since participation is mediated by multi-modal textual forms, we consider them to be key in characterising digital literacy. A fifth resource, persona, was therefore introduced to take account of this.

The lack of transference between academic and personal or social practices by learners, noted earlier, is of more concern when one considers the elision of these domains in the digital. Reputation, image projection and interpersonal relations are both mediated by and become repositories in the digital. This has lasting impact on professional profile, both positively and negatively (note the rise of ‘digital forgetting’ consultancies). Development and management of one's online persona/e can be considered both a professional and a graduate attribute but can only be effectively developed in conjunction with the other literacy resources.

An overview of the persona resource and its characteristic dimensions is provided in Figure 5.

The four resources model articulates each resource at a conceptual level based on ‘families of practice’ but does not detail specific competencies. These are seen to be areas for interpretation and management by the teacher. Whilst we share the goal of maintaining teacher autonomy, we were aware of a significant difference in context regarding digital literacy: teachers can be considered expert practitioners in reading and writing, in literacy, and are more skilled than their pupils in this domain. Conversely, many academics may not identify themselves as expert in the use of computers, particularly at the level and range being considered, whilst their students will vary in expertise as highlighted previously. Additionally, terms such as ‘meaning making’ may lack appropriate significance for individuals where this is not a theoretical construct as it is in literacy/literary and media domains. Finally, whilst wishing to avoid specification of a curriculum, our project was to develop staff practices and curriculum application in ways that would help to support and scope digital literacy development in practical contexts. It was therefore necessary to articulate some dimensions to the resource categories in order to provide conceptual delineation and to concretise avenues of application, whilst maintaining consistency with the original model. These component dimensions were arrived at through a process of testing the original four resources against the digital, identifying equivalence and working through example applications and contexts.