The critiques of OO in the comment thread, referencing for example the classic problem of object/relational impedence (the
Vietnam of CS!) reminds me of some of the postmodern critiques of logocentrism (made famous but not invented by Derrida), which leverage some of the same species of problems in set theory/incompleteness, in order to emphasize the subjectivity of knowledge and the fluidity of borders between substances and so forth.
A little Pomo 101, for my remedial benefit more than anything: The Platonic Ideal, in which the Logos (the original 'Word' or 'Reason') grounds and comprehensively justifies language and abstractions, is shown to be impossible even in theory, much less in practice. Where the natural standpoint (for instance, of classic structuralism) with regards to language is to assume that a 'sign' ('word') actually presences and mediates the thing signified, the underlying 'play of difference,' or 'impedance mismatch' to borrow from CS theory, shows how this is a lie. In order to ground and validate any system of signs and signifieds, we must assume there is some Ideal outside the system (outside "the text") that is not subject to the play of difference. Without this, all systems of signs are fundamentally malleable, and this malleability is what makes prejudice and oppression possible.
One of the problems I can see with introducing this line of thinking is that there is a historic tension between the post-structuralist critiques of Derrida and his followers, and feminist theory. While both share a common goal of critiquing Western hegemonic language, there is not a 1:1 sympathy, and there has often been antagonism between various feminist camps and various postmodern camps. Some feminism affirms the post-structuralist program, others see it as introducing a value-neutrality that actually makes it
more difficult to critique patriarchy. "Non-normative" is a two-edged sword when you're talking about social justice.
So to my mind, there's nothing specifically feminist about introducing 'non-normative' into the question. It's also worth noting that feminists may value and emphasize qualities like fluidity, openness, and ambiguity as a function of their critique of patriarchy/male hegemony, but those qualities are also emphasized in process-oriented philosophies like that of Bergson, Whitehead and Deleuze. If the ideas of 'non-normativity' and set incompleteness are not peculiarly 'feminist', how could a programming paradigm that incorporates them be referred to as 'feminist' in any meaningful way?
Additional questions:
* Are we rejecting (as many post-structuralists advocate) the principles of identity, antithesis and excluded middle? If so, what grounding is left for logic itself, much less a robust core CS theory? It's one thing to talk about how "
object oriented programmed reifies normative subject object theory," it's another altogether to follow through to the level of CPU instructions and the logocentrism inherent in logic gates.
* How does introducing the language of feminism or post-structuralism advance CS theory? Things like O/R impedance, set theory (database) incompleteness and so forth are already very prominent questions in the CS world and have been for decades. You could argue that few domains are more aware of the practical limitations of logocentrism than CS theory.
* Which feminism is to be preferred? This is a huge question. There are nearly as many feminisms as there are feminists, some sympathetic to 'non-normative' approaches, others -- and it could be argued most -- demand a normative standpoint in order to identify, judge and correct social inequalities.
It doesn't seem to me from Arielle's abstract that she has really grappled with these kinds of problems, and doesn't have a trajectory in place that is very promising. Maybe I missed something, but without a more serious engagement she's open to being caricatured (as we saw in that puerile and painfully unfunny parody.)