In front of you lie the meandering hyper-text and hyper-media created over the course of four years by the Almat team, its invited guests, and the people that contributed to our workshops, seminars and symposia. We have embraced the open-ended structure of the Research Catalogue's graphical editor, and rather than presenting a linear narrative of what Almat was and what its finding were, we refrained from returning to a classical form of scientific report. Instead, we want to stay true to the nature of algorithmic experimentation in the form of its presentation. And true to the concept of continuous exposition, making perceivable the unfolding of the project over time.
Although this text appears in a singled out place with respect to the network of Almat expositions, it is by no means the authoritative guide to Almat. It was written in March 2022, almost a year after the project “formally finished”. In the chapter Algorithms under Reconfiguration contributed to Michael Schwab’s book Transpositions, I presented a view on the analysis of works and research that replaces the conventional boundary drawing of units along the nominal delimitation of a piece, an exhibition, ... by groupings or threads that move transparently through these units. And the selection of threads is an active move, there is no entirely separate observation from the observed. I think this is a good way for you to approach reading Almat. Formulate a question or identify an interest in the project and its objects. If you haven’t found one yet, begin by skimming through the materials, or use the meta exposition interface. Then, try to weave the threads that depart from your question or interest. Take notes, create bookmarks, go away and come back.
For a good part of the project's initial phase, we were, naturally, invested in the ontological question of algorithms. It becomes clear that our research object are not algorithms, not a reification of computation, but their processuality; what "they" do with us, to use another reifying or even anthropologising language. You cannot talk about algorithmic agency without enacting it, putting yourself into an algorithmic experiment. But in the end, do you want to talk about it? It has been noted by many that despite its open-ended structure, the computer screen and the Research Catalogue ultimately favour written text. This is not what we were after.
If you want qualifications of the nature of “the” algorithmic, I think the best ones (so far, tentatively) are reconfiguration and recursion. A lot has been said about the latter, and I strongly favour the former. I am just one voice of many voices that participated in Almat. To say the algorithmic is reconfigurative is going beyond the assertion that algorithms consists of elements that can be recombined and that, when set into motion, are performing recombination. It is difficult to do without metaphors, and so I will say that reconfigurative refers to the internal forces of the algorithmic medium that constantly push for connections and disconnections to be made, well captured by Parisi’s idea of a speculative reason and a form of contagion rooted in the algorithmic. After two years of Coronavirus pandemic, we have a good understanding of the thoroughness with which something from one domain, biology, extends into other foreign domains, the economy, the social, philosophy and thought, art. The contagion is not just between one biont and another. Sets of relationships are thrown into doubt, they become unstable before crystallising in new stable forms (forms are always what we perceive as form).