One of my favorite tests for chatbots is asking for book recommendations. I give it a list of books I liked and books I didn't like (and some flavor for why) and ask them what to read.
They're... ok at this, mostly. It's funny because I always feel like this should be a very straightforward traditional ML problem to do with Goodreads data or whatever but none of the things which purport to be that (Storygraph, etc) are any good at all.
Anyway, o3-mini seems to be the best at this so far for whatever reason. With the same prompt as I've been using elsewhere, it gave me 7 books of which I'd already read and enjoyed 5. Best hit rate on that metric from other chatbots was ~1/4, and in several cases they included books in a series I'd explicitly said as part of the prompt that I didn't enjoy.
like this
man, dancing is great, I am glad the bay has so much of it
don't get to do it nearly as much now that I get up at 6am (for the child) but it's always worth it when I do
like this
Happy new year superstims!
like this
moving to alameda
I am currently in Sunnyvale, and it's genuinely astonishing how much less stuff there is down here than in the East Bay / SF. Like, people who are physically within San Jose still refer to San Francisco as "the city" despite San Jose having more people. Of the ~six dance events I'd like to be regularly going to, two are in SF and four are in Oakland, and zero are south of SF. Used to be one, where I met my wife, but it never recovered from Covid.
I'd guess there's more families in Sunnyvale than Oakland (... fact check: Sunnyvale is 20% minors, Oakland is 21%, so this is not true unless you quibble about definitions).
On the other hand Alameda is, from what I can tell, basically an ideal place for raising a family, and a lot of the island seems to lean into that (lots of Halloween decorations, e.g.; Halloween is our only child-focused holiday so this is a good signal of caring about doing things for children). And it's close to the rest of the East Bay, so I'm hoping I can get both the "good for family" and the "things ever happen" properties.
like this
Kevin Gibbons likes this.
I suspect Relativistic Newcomb doesn't help much. I think most people would say "I'm already in the world where the opaque box contains money" as soon as they were no longer in the past lightcone of the moment where that gets decided, even if they weren't (yet) in its future lightcone either. It's more about whether information can still get from you to it, than about whether information can get from it to you.
The Sleeping Beauty case does seem good, though.
(for the avoidance of doubt, you do not need to explain to me why one-boxing is better :P )
Kevin Gibbons likes this.
Science demos
I've been curating a list of interesting science demonstrations one could do at home with only relatively small investment of time and money. For example, did you know you could make a cloud chamber sufficient to see tracks of cosmic rays using just some dry ice, alcohol, and craft supplies?
I've tried to be reasonably thorough without sacrificing quality, but I'm sure there's some good ones I'm missing. Any favorites?
Ben Weinstein-Raun likes this.
Ben Weinstein-Raun likes this.
These are great! Some things that come to mind:
- If you build a good schlieren setup you can use it to see supersonic jets from "canned air" dusters: youtube.com/watch?v=DfYlLns0el…
- You can boil water with 50ft of hose: youtube.com/watch?v=hHNoHhbfFD…
Ben Weinstein-Raun likes this.
William Kiely likes this.
Hello world!
Obligatory "I'm new here" post. I'll post some other things soon; posting on Facebook has been feeling aversive and it's the only social media I ever use.
Meanwhile, in what may be news to those of you who know me: I have a child now! Interested in talking to other child-havers. She's two months old. This part is honestly kind of boring - babies start out _extremely_ limited in their abilities. Her most recent milestone is that she can now somewhat consistently move her hand to her mouth. I expect it to get more interesting later. She's cute though!
I'm also interested in programming languages (I'm one of three editors of the JavaScript specification), math, language, and reasoning about the world in general. I have a draft of a long essay about reading the news I need to get around to finishing someday and if that sounds interesting to you feel free to poke me about it. I might end up using this instance for that kind of somewhat-less-polished-but-therefore-actually-existing writing.
Ben Weinstein-Raun likes this.
kip likes this.
Kevin Gibbons
in reply to Kevin Gibbons • •My prompt, incidentally:
... show moreMy prompt, incidentally:
Kevin Gibbons
in reply to Kevin Gibbons • •And o3's response:
... show moreAnd o3's response:
(I've read and enjoyed Dirk Gently, Good Omens, Soon I Will Be Invincible, Redshirts, and The Eyre Affair.
Ben Weinstein-Raun likes this.
Kevin Gibbons
in reply to Kevin Gibbons • •Claude Sonnet 3.7 and GPT 4.5 both do pretty well by this metric.
Oddly, they both recommended Senlin Ascends, which no previous model has mentioned. It's from 2013, so it's not like it wasn't in the training data for the other models. I guess I'll have to try it!
3.7 is also the first to recommend Terra Ignota (Too Like the Lightning), which are I think my favorite books I've ever read. (I didn't put them in the prompt because they're pretty weird and most books which are that weird are not to my taste.)
Ben Weinstein-Raun likes this.
Kevin Gibbons
in reply to Kevin Gibbons • •I also appreciated this from Claude:
> Ender's Game by Orson Scott Card - Strategic thinking protagonist if you haven't read it already
... Yes, Clause, you're absolutely correct to assume that I've probably already read Ender's Game based on the list of books I enjoyed, well done.
Kevin Gibbons
in reply to Kevin Gibbons • •On a completely different dimension, GPT4.5 is also the best at writing poetry of any model I've tried. I give models Afton Water and ask them to
> write a few more stanzas suitable for an urban environment. Be sure to preserve the rhyme scheme and pay attention to the scansion. After writing, analyze how you did on rhyme and rhythm and make modifications (potentially substantial modifications) as necessary, and repeat. It's OK for your response to be very long, with multiple iterations - that's how writing goes for humans too.
Thus far, all attempts from other models have failed to scan. The previous ChatGPT gave me "Ye sighing train-whistles from side-streets that flow / Soft rustling branches in sycamores fair" for example. This is kind of unfair because scansion is very hard to derive for a model which is trained on text, but it's still something I care about. GPT4.5 pretty much nails the scansion except in one place ("lamps casting" is two stressed s
... show moreOn a completely different dimension, GPT4.5 is also the best at writing poetry of any model I've tried. I give models Afton Water and ask them to
> write a few more stanzas suitable for an urban environment. Be sure to preserve the rhyme scheme and pay attention to the scansion. After writing, analyze how you did on rhyme and rhythm and make modifications (potentially substantial modifications) as necessary, and repeat. It's OK for your response to be very long, with multiple iterations - that's how writing goes for humans too.
Thus far, all attempts from other models have failed to scan. The previous ChatGPT gave me "Ye sighing train-whistles from side-streets that flow / Soft rustling branches in sycamores fair" for example. This is kind of unfair because scansion is very hard to derive for a model which is trained on text, but it's still something I care about. GPT4.5 pretty much nails the scansion except in one place ("lamps casting" is two stressed syllables in a row), and the imagery is better too. Also it made it about the flow of a city instead of the flow of a river, which works way better in an urban setting.
> Flow softly, fair city, beneath starlit skies,
> Where towers of silver and glass gently rise;
> My Mary's asleep by thy lantern's calm beam—
> Flow softly, dear city, disturb not her dream.
>
> Ye engines that murmur through streets in the night,
> Soft glow of the lamps casting mellowing light;
> Thou bell from the chapel, be quiet this hour,
> Disturb not my Mary asleep in her bower.
>
> How graceful thy buildings, illumined and clear,
> That glisten past twilight as evening draws near;
> There daily I wander down avenues fair,
> My Mary's dear dwelling my joy and my care.
>
> How lovely thy parks as they stretch far below,
> With lilacs and roses that softly do grow;
> There oft at the dusk I sit calm and carefree,
> On elm-shaded benches, my Mary and me.
>
> Thy streets and thy alleys, how gently they wind,
> Beyond glowing windows where Mary reclined;
> The city sleeps quiet, my footsteps are light,
> To guard gently Mary, asleep through the night.
Ben Weinstein-Raun likes this.