This week I’ve been supervising folks writing code for our system, and reflecting on what would have been useful for me to do before then. I’ve also been narrowing down thesis topics. They’re all broadly security-related, and now I just need to read a lot. With luck I’ll find a paragraph at the bottom of a paper that says “We really wish someone would do this research”.
Last week I wrote about becoming an uncle in the coming weeks, and the weeks apparently noticed this and moved quicker for my prompting. So now I’m an uncle! At least for the moment. The only permanent in life is change.
I had a wonderful, cross-planet collaborative session last weekend. I am on darling-murder duty, where I reduce every character to their barest motivations and drives and consider them in relation to The Plot. And if they fall short, I threaten them with termination. It is a really peculiar lens through which to view media. The audience won’t experience the work I’ve been doing – particularly if my suggestions are accepted – because they simply won’t know what they’re not seeing. Additionally, they probably won’t consciously register that, for example, in Act 1 Scene 3 the motivations of two characters seem aligned but are then revealed to have been misaligned later, in Act 2 Scene 7.
Inevitably, when analyses are written about it (they will be) and Sparknotes gets stuck in and students read the abbreviated version, someone will point out that the line in Act 1 Scene 3 is repeated in Act 2 Scene 7 but with a different meaning. And when I read that analysis I will go, I kid you not, absolutely fucking feral. Because every writer is just a puzzle maker desperate for someone to solve the puzzle. We are cowardly serial killers.
My personal writing is still all in pursuit, more or less, of the thesis. I’ve not written much more than notes at the moment, little electronic post-its stuck to electronic papers about electronic things. “Is this related to SbD?”. “Consider in context of quantum”. “Are developers arrogant or is everyone?”. I’ll leave you to try to figure out the actual papers.
I’m slowly gravitating towards a couple of broad areas. One is a question that consistently bothers me, which is Why are developers bad at writing secure software?. Another is a theoretical question (that I’m struggling to find literature on) which goes something like How do you put identifiers on interesting data in a way that doesn’t identify the data as interesting?. Then there’s Cybercrime is a business now, and that’s interesting, and what kind of people does it attract? and How do you defend against real-time generative AI putting your boss’s voice on the phone? and What are the implications for the cybersecurity industry when Twitter goes down the –
Alright, so I’m gravitating in the same way that someone approaching a black hole might be gravitating: spinning myself into loops and apparently not going anywhere.
I discarded one of my extracurriculars this week, and feel lighter for it. I think I need to drop one more though, not least because the matching algorithm may be about to make a reappearance. Someone new has joined the project and they were surprised when I reeled off figures like 3,628,800. Nobody else was, of course, because I say these numbers a lot in the context of this project. The problem we are trying to solve is harder than people think it is; not just harder than it looks, harder than you think it could be having looked at it. There is word in the team that someone has independently approached us claiming it can be solved in a week, which is odd: at the moment it’s either 4 weeks of staff time or less than an hour of compute time. This is the first time we’ve had someone pitch a solution somewhere in the middle. Still, it’ll be good to find out the angle they’re coming from.
For me, I’m keen that we get out and do some consultation and maybe a hackathon. Maybe create some fake data with the same statistical profile and throw it out to some eager folks who want to test their mettle.
Ah. No. See. That would be more things, not fewer. But…
Caffeinated Punctuation 