Image: Gustav Metzger Recreation of First Public Demonstration of Auto-Destructive Art 1960, remade 2004 Hydrochloric acid, white nylon, glass, found objects 3000 x 2500 x 1000 mm © Gustav Metzger

In the previous post, I began to review Luciano Floridi’s ideas linking Dedomena–> Data–>Information. I’ll follow up on these now by focusing on points 2 and 3 of his data-information framework before going on to ponder some potential creative approaches.

1 – Information consists of n data, for n ≥ = 1 (Diaphora).
2 – The data are well formed syntactically
3 – The well formed data are meaningful

Syntaxical Data
As Dedomena cannot be grasped independently of abstraction, they require ordering through the application of cognitive and observational processes that draw upon agreed rule sets or syntax (and by implication methods and technologies). Syntax functions to determine the shape, construction, composition and structuring of data in ways that make it ‘well formed’ in Floridi’s terms, which I take to mean comprehensible and usable; an architecture. Interestingly he chooses to give an example from culture to illustrate this referring to how pictorial space is rendered using the rules of vanishing point perspective.

So if Dedomena are ”raw data’, then data as we might understand it to be, (i.e. graspable, mobile and useable abstractions), are ordered according to agreed upon syntactical practices. Agreements are of course a social construct that again highlights how data are formulated culturally inheriting all manner of ontological biases (as opposed to a view that they pure or neutral entities).

Semantic Data
The move from Dedomena to data, (and then from data to information) is then understood as an increase in abstraction (of the world) between nested elements a bit like a russian doll. Information relies on data, and data link to ‘data in the wild’ Dedomena. The final ordering of this chain of elements is semantic whereby data translate in Floridi’s terms as meaningful.

This is a more complex and problematic differentiation, but my reading of Floridi is that this is essentially a semiotic issue, i.e. the application of cultural conventions that enable and guide meaning construction from a given data set. Referring back to the example of perspective given above, this might be expressed as an overlaying of pictorial codes of representation (e.g. photo-realist, expressionist etc.).

What”s important to stress here is that codified form of data, the semantic level then functions to underpin meaning construction independently of the original author, and by extension the original data set and reference point. As discussed below.

Data-Information musings
The three levels of the data-information framework are useful in that they clarify a chain of relationships between elements ‘in the world’ and what we might broadly call the cultural constructions of data. It’s important here to emphasise both the connection and separateness of these levels and the consequences and opportunities these offer. While each link and inform each other, they are also separate elements in their own right which explain how information, while conceptually linking to a referent via a chain of material and conceptual processes, is also easily decoupled from it. This it what gives data/information its particular mobility and potential for articulation across many different formats both analogue and digital.

The consequences of this for creative practitioners are manifold. Each level (dedomena, syntaxical, semantic) offers opportunities for critical and aesthetic interventions that play across ontologies, devise novel data collection techniques, mangle syntaxical frameworks, devise new semantic codes, and explore differing material expressions of data. Here’s an example to ponder:

Dedomena (observed differences in the world)-> Application of syntax to data set (conceptual/performance art)—> Symantec application (Auto-Destructive Art)

As a critical approach an auto-destructive data set is probably much needed in certain settings.