Hot take: that XKCD comic is as much if not more a commentary on software peoples' Exceptional Thinking, rather than "yeah haha bitscoin elections are bad!"
We, as an industry, are decades behind the rest of the world, in safety and resiliency development, are actively hostile to improving that situation in this mistaken belief that velocity and Logic will somehow save us. We don't even have vocabulary to talk about verification and safety, we just rub magick solutions on it and hope
it works out.
Those of us who don't just blindly present their solutions take on this nihilist view that there's no fixing software, that We Are Truly Lost.
No vision for what a world where computers can be trusted for the important things we're already using them for, as much as they can be trusted to post our fucking cat pictures. Just a firmly held belief that it's hard and thus impossible to make a system resilient to active attacks and passive failures.
Plane crashes are weird and unpredictable not because planes are somehow immune to simple failures. Avionics engineers spent *decades* building systems and vocabulary that allow them to reason about their system in a way that is Correct.
Assholes who work at Facebook or Amazon or Google or Uber will scoff at them and cast them aside because ~velocity~ as if the quarterly numbers are the only things that matter. It's no different than the people who rub blockchain on the problem.
laughing about the stupid fucking clowns who think blockchains will solve our problems is easy. Taking an honest look at how immature and underdeveloped the very underpinnings of society are is hard, let's make fun of the Others
@rrix a generous interpretation is that the humor is used as a sugar coating that could help the spread of the comic which contains the seeds of the "what if software, but with actual responsible engineering principles" train of thought
spose it's somewhat unlikely that was intended tho
i guess more succinctly: election systems aren't harder to build than planes or industrial systems. Pretending that they are and either 1) disclaiming any ability to do it or 2) rubbing magick like bitscoin on them are shitty responses compared to 3) grow our profession and our fundamentals enough to be able to construct systems which are trust-able. They don't need to be Provably Correct And Written in Habskrell, but we need the same sort of verification systems and processes that Real World engineers have
Replace "election systems" with "hard computer science problem du jour"
@rrix it's funny how my perception of the comic is totally different from yours.
my reading is that clearly we know how to build safe election systems; after all, we have mastered making huge metal tubes full of people stay up in the air reliably, which is self-evidently much more difficult!
so by drawing attention to the fact that we as a society have chosen not to do this thing that we know how to do, it implies "why is this? why have we chosen this path?"
@technomancy I think we're taking the same point, I'm only prescribing an answer to the last one that is "empty nihilism and the inability to talk about hard problems in solvable vocabulary like other engineering professions" I have seen many people here and twitter and even in work engineering chat not getting even "why is this?"
@rrix I see what you mean. my conclusion is more like "it's fascinating how perverse incentive structures can lead to results like this; how can I be more aware of incentive structures to avoid similar traps"
less a "brokenness is inherent to computering" and more "brokenness is inherent to society", but even more "here is one particular failure mode which has manifested; what can we learn from it?"