Profound words from the manual
Remote
Dan Piponi
@dpiponi@mathstodon.xyz
mastodon 4.7.2Disclaimer:
👽 My opinions are not my own. They're beamed to me by aliens
Current life:
🧠 Thinking at high effort
Previous lives:
🍩 PhD in Riemann theta functions
🧬 Glaxo Group Research
🚀 Argonaut Games
💥 Mass Illusion, Esc, ILM
🎈 Google X, Verily, Google
🎮 Epic Games
Likes:
🚴 I like to bike
🏃 I like to run
🎛️ my musical tastes lie towards the electronic end of the spectrum
🚀 I like Andor *and* The Mandalorian
📈 I have an OEIS entry https://oeis.org/A387361
2708 Followers
428 Following
50 Posts
Joined April 27, 2022
Is there a word for continuing to watch a TV show just to see how bad it can get? Like Dark Matter.
Replying to
@metarecursive@mathstodon.xyz @synlogic4242@social.vivaldi.net The current "high concept" Apple TV show based on Blake Crouch's 2016 novel.
Australian spiders not scary enough for you? Well now it seems they build catapults.
https://www.abc.net.au/news/2026-06-23/ballista-spider-new-discovered-web-traps-green-ants-queensland/106831796
TIL the now defunct silverware company Oneida (who manufactured our own silverware) started as a sex cult whose membership included one (actually successful) assassin of a US president
The most perfectly post-apocalyptic beach I've ever visited. The looming nuclear reactor complex gives it all a bit of a Stålenhag feel.
Something it took me many years to appreciate is this: a lot of constructions in algebra can be interpreted as figuring out what you can precompute when you're not sure what's coming next.
The first example I was taught may have been the free group on some generators. You're basically saying "well I don't know how to multiply these things so I'll just record whatever I need now so when I do learn how to multiply elements I've already done as much work as possible".
Sometimes there's even a kind of cost model. As I've mentioned before, a tensor product is basically what you need to store from x and y when you know that at some point in the future you're going to compute many bilinear functions f(x,y), but you don't know what f actually is, arbitrary multiplication is expensive, but memory is cheap along with addition and multiplication by a constant.
In a sense everything I said here is wrong. For example in mathematics you have considerable freedom to choose what sets actually represent a tensor product so it makes no sense to identify a construction with a particular "implementation". Nonetheless the definition of the tensor product corresponds well to the sort of interface you might attach to an obvious precomputation for the bilinear function scenario.
(I think this is basically summarising what a Left Kan extension is.)
All day long. Copying numbers from emails and texts to web sites. From little plastic cards to apps. Digging them out from old emails to copy to web sites. 6 digits. 4 digits. 12 digits alphanumeric. All day long. Make it stop! I'm not a number copying machine.
Outside of my immediate social circle it seems like a high proportion of people are conspiracy theorists. We have entire TV channels that seem to be nothing but alien conspiracy documentaries. Has "conspiracy theory" taken over as the main religion of the US?
I've never heard of this book by Fred Hoyle before (Ossian's Ride) but I was sold when I read the blurb and first paragraph.
The British Library is always worth a visit.
Here's a typewritten document by Alan Turing:
"I propose to investigate the question as to whether it is possible for machinery to show intelligent behaviour. It is usually assumed without argument that it is not possible..."
I'm sure Google can't actually be using generative AI to create plausible train times around London but after a week I'm yet to see any real train that corresponds to anything listed by Google Maps.
Vanity purchase on eBay due to having a mention in it. But check out these ads. I tried doing something with Shortcuts on my iPhone this morning but I think that Sharp might do it many times better. Only problem is I now have to figure out how to restore failed glue in a "perfect bound" magazine from the 80s.
Replying to
@drewph@ieji.de Apparently it's already a bit of a catch phrase and I'm surprised the t-shirt doesn't exist.
The US has basically abandoned democracy and replaced it with some kind of weird game involving fractals.
I enjoyed the metaness of the moment as I sat down in Pret à Manger, King's Cross station, surrounded by Pottermania, to sip my cappuccino and start my new book which coincidentally started at King's Cross station with the protagonist surrounded by Pottermania.
(Slow Horses, if you were wondering.)
Movie idea: elaborate plot to hijack an aircraft involving hijackers making a movie with the sole purpose of winning a large hunk of metal that they could use as a weapon on an aircraft. Clearly there is an audience for whom this is a plausible story.
An Emmy might be better than an Oscar. A friend of mine had a bit of hassle trying to bring his back one year...
I'll be listening many times more but nothing on Inferno has really grabbed me so far. Tomorrow's Harvest is still the greatest album ever made.
Replying to
Maybe you'll enjoy the continuation onto the next page.
I'm a both sides person here. I grew up on those magazines (and Basic).
Anyway, this is from a book I've mentioned here before: The Creative Computer by Don Michie and Rory Johnston. Published in 1984 but relevant today.
Replying to
@johncarlosbaez@mathstodon.xyz @richardelwes@mathstodon.xyz You can still write a book about -1/12
I enjoyed watching The Mandalorian and Grogu even if the target audience is about 8 years old. I guess I was mildly annoyed by the whole gladiator has to win just one more fight story that's as old as the hills and always plays out the same way. But Grogu and the Anzellans made up for that and it was cool to see some instantly recognizable Phil Tippett stop motion.
I watched it in Borehamwood where the original Star Wars was made, one day shy of its 49th anniversary.
There are a lot of District line stations that are place-names in the US:
Monument, Richmond, South Kensington, Temple, Tower Hill, Victoria, Westminster and Wimbledon.
Replying to
@johncarlosbaez@mathstodon.xyz @richardelwes@mathstodon.xyz I'll proofread it a couple of hours before you start.
Bank holiday, hottest day of the year, and yet this beach on the south coast of England is almost empty...
Replying to
@futurebird There's a clearly visible gap between baseball and softball. Pity Americans don't play cricket.
Replying to
@theohonohan@graphics.social I love John Harris though I only realized the connection between various book and magazine covers and Sinclair relatively recently.
Replying to
@jac@types.pl But I have to keep memorizing again every time they get revoked which happens too often.
Replying to
@secretasianman@types.pl @jonmsterling@mathstodon.xyz Said paste buffer automatically transferred wirelessly from desktop to phone. Small mercies.
Replying to
@rich@mastodon.gamedev.place IIRC up to 4 might receive Oscars for a VFX award. Which gives some shape to the plot now.
Replying to
@BartoszMilewski@mathstodon.xyz 😂
Seriously though, this is meta in a way I'd love to see!
Replying to
But don't miss the sequel suggested by @BartoszMilewski@mathstodon.xyz that is better than 99.9% of movies:
https://mathstodon.xyz/@BartoszMilewski/116501265594153440
Replying to
@tpfto @DataKnightmare @au_hasard I had it as a kid and I've really enjoyed revisiting it recently. Live long enough and you get to use some of that hindsight people talk about :) Surprised to find I even knew the creator of one of the images in it.
Replying to
@peterb It's even more extreme over the page, but it was a bit awkward having a long section with a page break in the middle so I just posted the summary before the main quote!
Replying to
@mjd@mathstodon.xyz @tpfto@mathstodon.xyz There's a two-stage process here.
One is computing things when you don't yet have all of the information. But at the first stage these are literally the computations you might perform with pen and paper.
At the second stage you reify those things. For example, an expression you might write when doing group theory becomes, itself, an element of the free group. It's an object that formally captures this idea of computing something with partial information.
(I think you mentioned something about closures and monads the other day. This is closely connected. If you reify an expression written in group theory you get an element of a free group. But if you now write that down it looks identical to the original expression. This is the MM -> M of a monad.)
Replying to
@sjb@mstdn.io Thinking aloud here: I think what I'm saying is we write things like 3x+2 all the time. But when learning about free groups we actually "reified" those kinds of expressions for the first time.
Replying to
@sjb@mstdn.io Exactly right. But I think there's an interesting story here which is that when we write down something like 3x+5 that expression is itself a free construction and in a sense that's closer to what I'm talking about than the x itself.
Replying to
@schrotthaufen@mastodon.social I like to think I can do that - I can do sub 1hr for a 10k race on rough forest terrain with small hills and tree roots...
Replying to
@tristrambrelstaff@mathstodon.xyz I have no idea what the consequences of such a system would be. TBH I think it would be worth modeling!
Replying to
@theohonohan@graphics.social The design of Practical Computing was all round pretty ugly. IMO that ZX81 ad really shows it up.
Replying to
@synlogic4242@social.vivaldi.net Did you mean S2 E1? It really made no sense to continue after the end of S1.
Replying to
@SnoopJ @futurebird It's pretty straightforward to play with "raw" LLMs, eg. with ollama or llama.cpp.
BTW If we're being pedantic "they reproduce the statistical distribution of tokens in their training corpus" isn't quite right. Inductive bias is crucial otherwise the model grinds to a halt on novel inputs. (And I'd really like to know what it looks like when you do this but I don't have the resources to find out.)
Replying to
@synlogic4242@social.vivaldi.net Oh, you couldn't get into season 1 at all. I didn't dislike season 1 but 2 is another matter.