From Intended Archivists to
Intentional Algivists.
Ethical codes for humans and
machines in the archives*
If the digital future is to be our home, then it is we who must make it so"
Shoshana Zuboff
Overview
Starting from the prediction that someday algorithmic archivists - or as I like to call
them: algivists - will work the archives, I describe how ethical thinking in traditional
archives could be employed to teach algivists moral values. The essay starts by
describing the digitalization of society and archives and how so-called codes of ethics
have evolved to define the moral values of archivists, characterized as the intended
archivist. I then turn to ethical thinking about algorithms, how different types of
algorithms induce entirely new classes of ethical challenges, and argue that a good
way to endow algivists with ethical behavior is to employ the same kind of
technology, algorithms, to encode ethical values directly in their core beliefs as a
bias. This results in the intentional archivist, or the algivist. In this essay I develop a
vision on the future of the algorithmic archivist and an idea to obtain algorithms in
archives that obey our human ethical values.
(1) The Coming Archivist Singularity1
[Some place, some time in the future] It took ages to get permission, but yesterday
evening I finally got THE mail. I consider myself lucky, since I really needed access to the
archives to finish my article. Other people would ask why an assistant professor in
technology ethics would like to see those old-fashioned paper documents about the
introduction of Mindbook, the company that grew out of the long-gone Facebook
Corporation. Since, their summaries are already on Archipedia. Who is interested in
paper documents anymore? Well I am. I never felt comfortable with all this digital... eh
stuff... anyway. People are physical, and they like physical things. Well... at least that's
my opinion. And besides... I don't trust Archipedia; they have appeared in so many
algorithmic trials for information manipulation, but they always use their right-to-
silence and nobody is able to crack their summarization code. I need to take a look
myself. I enter the red building next to the rocket station and turn right after getting
martijn van otterlo
The author acknowledges support from the Amsterdam Academic Alliance (AAA) on data science.
1 All links appearing in footnotes have been last accessed October 2017.
267