STAR VALLEY (SIRIUS) (2018) In a centennial technology arc, from the start of the radio era to the software defined radios (SDRs) of today, the wide bandwidth transmitter utilized in the work is modeled on the earliest method of radio transmission, the spark-gap transmission apparatus, which was implemented by physicist Henrich Hertz to prove James Clerk Maxwell’s theories of electromagnetism and later used by Marconi for his transmission experiments.
The system generates a pair of codes words representing an imagined operation, and the first plinth’s corresponding screen displays the code words, converts them to morse code, and the spark-gap transmits the morse code. Finally, the code words a deep learning model which returns a description, and the first plinth displays it. The second plinth then receives the same set of code words, which are seeded into the same model, generating its own unique description of the operation. Each Spark gap apparatus has a corresponding monitor displaying the current nickname and description with a list of the recent generated nicknames and descriptions, n order that the visitors can see a long term record of the inner workings of the system as well as follow its evolution.
The underlying system is a natural language processing machine learning algorithm that has been trained on the US Department of Defense and NATO nicknames and descriptions of programs, activities, exercises and special access programs. The work is in part inspired by the 1975 Joint Chiefs of Staff Code Word, Nickname, and Exercise Terms System (NICKA), which automates and tracks such assignments. The work reflects on the present state of encoding, decoding, secrecy and transparency by transmitting these imagined nicknames in morse code through the wide band spark-gap apparati. For STAR VALLEY we have paired the earliest form of wireless transmission with contemporary computational systems to open a window on the landscape of the accountability of secrecy and artifical ‘intelligence’