Now
Updated 24 days ago.
(For information on “now” pages, see Derek Sievers’ original writeup on them.)
2025 — 04 — 15
Lots of change!
The biggest thing on my mind presently is that I’m almost certainly leaving KCL, with tomorrow being my last day. It’s still possible we’ll get extensions / some kind of extra bit of cash that’d keep me around — I’ve begun to joke to people that I’m an academic cockroach that can’t be killed, no matter how hard the system tries — but I suspect my goose is finally cooked.
That gives me a little time to work on some things I’ve been thinking about. I’ve got some papers I’ve wanted to write on ideas I had back in Glasgow when I was lecturing; a bunch of papers I should be able to get out on my thesis work too; and some additional work I’ve been planning based on some new ideas I’ve had in the last month or two.
Those new ideas are quite fun. I’m getting curious about metascience. In particular, I’m getting curious about using multi-agent system modelling to represent scientific work, and inject changes into the modelled system to measure the impact of different prospective changes to scientific practice. For example: if chatGPT gets used to write papers, or to review papers, or to write rebuttals to reviews, how does that change peer review? If all three of those are the case, should we expect publication quality to drop? How prevalent do those uses of AI have to be before there’s an impact, and how much AI use — of different types! — can peer review tolerate before the system completely breaks down?
I think there’s a lot of opportunity to integrate pdsf
, the library I wrote for my PhD, to produce interesting multi-agent simulations. There’s also opportunity to model science using trust formalisms, which could open up all sorts of metascience of AI (and other metascience too) by using models of trust from different disciplines to support the types of analysis that are common in those disciplines. So much we can do!
I’ll write about it more separately. I’d like to make time after my current contract ends to write a little paper on the metascience ideas I have so far — so perhaps in the end I’ll just link to that.
Running comes and goes. I hurt my knee badly just before Christmas, and it effectively kept me off the roads for about three months. It’s a shame, because I was close to a half marathon at the time. So it goes! I’m building back up now.
I’ve recently discovered Blood On The Clocktower and I’m hooked. Only played a couple of evenings so far, with about five games between them, but it’s been excellent. The game works like Werewolf or Mafia, but with a couple of clever little changes that makes the game come alive like no other social deduction game I’ve played. I’m not a huge fan of Werewolf or Secret Hitler or the like — I’ll play if folk want, just not my choice — but Clocktower is something else.
I’ve been writing about how to run games on this site; I’ll write about playing sometime too.
I’ve also got rid of my smartphone, more or less. My main phone now is a Cat S22 and I really like it. It’s an Android phone, so I’ve still got WhatsApp, banking apps, maps, etc. However: it’s a flip phone with a T9 keypad and a tiny touchscreen. That means it’s feasible to do anything I need to, but not very pleasant to do anything distracting. It keeps me from using the phone any more than I need to, and I’ve found that very liberating.
I still have my iPhone. It’s useful sometimes. For example, it’s really hard to run Blood On The Clocktower unless you buy the game (which is currently out of print and upwards of £150 used) or have a smartphone. It’s also useful for the odd message that comes through on iMessage, or for the odd evening watching youtube, or just to engage with the Apple ecosystem I locked myself into for a long time… but it’s not required at all, and increasingly I find myself frustrated when I realise I’ve brought it with me or reached for it in a moment of distraction. There’s nothing “nutritious” about using it. I don’t feel good when I do. I’m not sure I can jettison the iPhone quite yet, but I’m getting closer every day…
There are a couple of reasons why I haven’t completely got rid of the iPhone already.
- Apple Health has my entire history of health data, and I don’t know where else to put it. I want to keep it, and I want to keep it up to date… so, for now, I need something that manages it. The iPhone could do that sitting on a charger in the corner of the room, but I do need it all the same.
- WhatsApp. Yes: it runs on the S22. No: I still don’t know if I trust the device enough to make it my primary WhatsApp device. It doesn’t feel particularly flimsy, but the company that made it has gone bankrupt, so there’s no support if something goes wrong. If the device dies, I’ll just lose all of my data. I also can’t move my WhatsApp history from iOS to Android — I tried, hard!! — new versions of Android support transferring WhatsApp data between devices, but the S22 runs Android 11 (Go Edition) so isn’t much help. Only some hardware supports it too, even today, seemingly. Until WhatsApp improve their system and make it easy to transfer chat histories across operating systems, I’ll have to keep the iPhone alive or decide to scrap my entire chat history.
So, for now, I’m keeping the old iPhone alive. Still: the dumbphone life is excellent, and I recommend it for anybody wanting to spend less time on their phone, or who wants to be more present.
That’s it in a nutshell. Lots been happening, but only so much I can write, and only so much time I can make while trying to get things wrapped up at KCL.
2024 — 06 — 18
I thought life would slow down while I was getting through corrections. No such luck. I’m not sure why the final few months of the PhD have taken so much out of me, but I feel spent. Can’t say why. Feeling spent.
Corrections are all but in now, though. They’re not not in — they’ve been approved and the university’s accepted them — but getting through a couple final notes that should have taken hours is taking days. Part of that is caused by constantly switching to work on Chrompartments — that’s been most of my time, in reality — but I’ve been working on Chrompartments so much partly because summoning the two extra hours of focus to just get the corrections sent to the library in fully polished form is beyond me.
Ho hum! Nearly there!
I played with Porta 400 and Porta 800 a couple of weeks ago, and the results are stunning. Who knew film stock could make such a difference? It’s a small thing, but shorts on the Portra look far sharper than on Gold. I imagine it’s because of the boost in ISO, more than the film chemistry more generally.
Recently found Tom Cardy, and damn, he’s good. Every track is catchy, and weirdly motivating too? It helps to have songs that are genuinely excellent but with lyrics entirely about going to bed on time or practicing basic self-care or calling your mum. Or my favourite, which is just a pep talk through the metaphor of Pluto’s planetary status. Genius. So well written! So catchy! I can’t stop singing them to myself.
2024 — 05 — 28
It’s been a really busy couple of months…
I’ve had my viva, and passed it, subject to corrections for which the magnitude is still kind of unclear to me. Minor? Major? I’m treating them like major. I’m also tearing through them as quickly as possible. Get ‘em done, get ‘em out the door. Move on to the next project. I’ve spent pretty much all of my 20s working on some version of the ideas in the thesis, and while there’s something satisfying about really inhabiting an idea like that — and my understanding of it has matured and matured and matured — I’m ready to put this figurative chapter and these literal chapters to rest. I’m also disappointed in parts of it. Ho hum. If I had my time again I’d certainly do things differently.
However, passing — and being about to submit! — is a really positive moment. I can’t wait!l to graduate! With any luck it’ll be in a month, after some cheeky pestering of the university. Not long now.
I’m trying to keep running up too — only a couple a week, but I’m getting a good 15km/10mi in across a typical week and that’ll suit me ok for now. When I’ve got a little more mental space I’m sure I’ll push harder. For now, I just want fo run consistently. Running constantly is another thing altogether.
I’ve been reading more lately, too. Investing in a Kobo was a game changer. A bit of fiction — Cixin Liu’s Three Body Problem was a brilliant read for anybody with a passing interesting in physics, politics, and perhaps environmentalism. Recommend. Katherine Rundell’s biography of John Donne — Superinfinite — has also been a charming read, although I’m finding less time for it than I’d like.
I’m also jumping back into sections of Haruki Murakami’s What I Talk About When I Talk About Running again. I finished the book a few months ago but it’s fantastic and well worth a read for anybody creative, discipline-starved, or who writes for a living. Not really about running at all in the same way that Coach Chris Bennett’s writing, speaking, and coaching on running is not about running. It’s about life.
A useful book to revisit in the final throws of a PhD thesis.
I’ve also been getting into photography more again. Photography’s been a really useful outlet on days where academic pressure has taken its toll. The last few days have seen me in the office until ~10pm — moments like that need outlets. Something creative that isn’t work.
The most significant change in my photography has been a serious lean into film. I suuuuuuuuck at film photography — with is essential, because these hobbies and creative endeavours seem to either become gear obsessions or learning obsessions, and the great becomes a great distraction when I hit a plateau on learning. How do we keep it nutritious? Find some other angle where there’s still lots to learn. I’m glad I’ve found a new type of photography that excites me and that I have so much to learn about.
2024 — 03 — 25
I have dates for my viva! 05-08 is the big day. It’ll be a couple of hours in the afternoon — I’m sure it’ll feel like a couple years in the moment.
There’s lots to do.
I’m still thinking lots about Aspect-Oriented Programming, about composing model components, about the relationship between model building and hypothesis testing, about how AOP’s been misconstrued in the literature to date…whether I’ll get the opportunity to get into all these details on the day is likely another matter!
2024 — 03 — 11
Time’s moved quickly lately. Prep for summer trips; wait for viva dates; Marta’s finally had some appointments we’ve been waiting for after what’s felt like aeons; winter’s finally begun to leave us, my enjoyment of cooking’s returned, a little passive burnout’s stayed. Refuses to shift. The city feels like it’s opening up to us, finally, which is both satisfying and a relief.
I’ve done my best to keep on top of things the last few months, but a couple things — especially catching up with friends — have really fallen by the wayside, particularly since the sprint to wrap up my thesis.
However, work’s yielded some success! The project I’m working on (chrompartments) started with some rather slow struggles to get things to build, which have mostly resolved. Some kind of cosmic mercy. Now I can get started on the task at hand and play with V8. We’re looking to provide some security (/performance?) enhancements to it using CHERI, and while getting it to compile we’ve had plenty of ideas as to what those enhancements can be. Now it’s time to play with them — finally! — fun.
I’ve been playing with photography more, trying to find ways to improve and to have a little more fun with digital…but I keep moving back to film. Frustratingly, my film photos rarely turn out how I’d like (which I’m blaming on my OM-1n’s broken light meter), but I enjoy the process so much more that I struggle to care. The practice is about having an excuse to see the world differently anyway.
Only halfway through Craig Mod’s Things Become Other Things. It’s fantastic…but I keep getting distracted by John Berger instead. His notes on an interview with Bresson — page 157 of this .pdf on archive.org — or his notes on the theory of narrative — page 117 of the same .pdf — are just lovely, and always provoke some moment of clarity for me. Worth a read!
2024 — 02 — 07
Improbably, I’ve begun playing a little chess. I’m probablytom
on chess.com.
Also, Bluesky just enabled public registration…! I’m @tom.coffee
there. The mechanism they have for domain name verification is clever and simple; I love it! Not convinced I’ll actually end up using it, though. Is there any replacement for the Twitter of 2009? Or is that warm memory more a product of nostalgia than of the quality of the community there? Mastodon and micro.blog didn’t land any more successfully for me…maybe the problem’s less the services, and more how we use them.
2024 — 01 — 23
- Juggling! I’ve got a regular three ball cycle and reverse cascade pretty much down. Practising for consistency and getting under-leg throws working.
- Getting back into running — after a particularly nasty fall in September I couldn’t run for months, and got out of the habit…
- Reading a little — I’ve just finished Murakami’s What I Talk About When I Talk About Running and started Katherine Rundell’s biography of John Donne. Top of the list is also Craig Mod’s Things Become Other Things.