Returning to the 2017-vintage Rabbit Hole format: No grand theme this week, just some links I thought were noteworthy.
Cards of Control
If you only read one thing this week read this 👉Cards of control: How ZTE helps Venezuela create China-style social control. This is an exceptionally well-sourced long form article on the implementation and implications of ZTE ID cards in Venezuela.
Decentralized Identity: Ideology & Architecture
If you want to understand self-sovereign identity I think this is about as good a primer presentation as any. It is no more than a 15 minute read time even if you are new to digital identity and, I think, hits the major points in a sensible way.
Btw: don’t be put off by the ‘Bitcoin Association of Switzerland’ logo on the cover … there is no further mention of bitcoin (or Switzerland) in the presentation.
Going Underground, as a Toaster
Despite writing in recent weeks of the rise and reach of ‘the Chinese system’, I don’t believe we will actually get full, unavoidable surveillance in most of the real world any time soon.
For most people in the general populace (i.e. groups who are not specifically and intensively targeted) of most countries the rise in surveillance tech just means staying outside the surveillance system will be inconvenient not impossible, and so most people won’t bother.
For example, if you want to stay out of the video recognition system you (sort of) just have to put a ‘adversarial perturbation’ sticker on your head – see link here for how to get yourself classified as a toaster. For a more technical explanation of why you can get yourself classified as a toaster, or panda, or whatever, there is a good paper here.
This is also why I think real world ID and app data are currently much more effective surveillance techniques than video as video processing is still expensive, imprecise and hackable. But video surveillance still gets the headlines as it just feels so much more like surveillance.
Defining and Designing Fair Algorithms
While researching SB-10 bill (referenced in last week’s Rabbit Hole here) I found Stanford Computational Policy Lab and their work on SB-10 and then their presentation on Defining and Designing Fair Algorithms – check it out, it’s long (112 pages), but really good work.
Sinister News Readers
And, finally, there is something deeply sinister about this Chinese news bot (scroll down to the YouTube video to see it in action). I think the sinister thing is something so human-like presenting news with no conscience of regard for ‘truth’ (>> insert preferred partisan joke here<<) … I don’t know, just deeply, deeply sinister in a Minsky / Shannon Ultimate / Useless Machine kind of way.