genmon

9 ms·

61.
by genmon 4y ago
I found that Dewey was borderline acceptable — more categories seemed to degrade it, though admittedly I didn’t spend much time prompt tuning. I also tried tags and these weren’t reliable (not at a consistent “scale” from episode to episode
62.
by genmon 4y ago
The BBC is pushing its Sounds app by lagging the podcast. Disappointing but understandable I guess given they want the opportunity to suggest other programmes.
63.
by genmon 4y ago
I think Angie Hobbs will take it! Looks like she is referred to with a few different names in the episode notes and I’ll need to merge them manually (will be 25 appearances) (btw thanks for datasette. I use SQLite as an intermediary db and
64.
by genmon 4y ago
Great list! Some personal faves in return, in no particular order: - The evolution of teeth https://www.bbc.co.uk/programmes/m0003zbg - The fish-tetrapod transition https://www.bbc.co.uk/programmes&#x2F
65.
by genmon 4y ago
Just to make sure that others see this, I found your clustering technique super interesting and very effective, and something I want to try myself. Your technical writeup is here: https://weekendcollection.substack.com/p&#x2
66.
by genmon 4y ago
Excellent, new to me and I'll give it a go, thanks! I gravitate to PCA for terrible reasons (undergrad so it's what I think of first) and like you say, it's beguiling yet disappointing, the components rarely have any human me
67.
by genmon 4y ago
Have you read the New Yorker articles about it? On _In Our Time_ -> https://www.newyorker.com/culture/podcast-dept/escape-the-ne... Profile of Melvyn Bragg -> https://www.newyorker.com/cultur
68.
by genmon 4y ago
Credit where it's due: I was working with structured data as JSON for the completion, and the Typescript definition hugely increased reliability. I took that from helpful advice (on Twitter) from Noah Brier who afiak came up with the a
69.
by genmon 4y ago
I've started thinking about LLMs as a "universal coupling" , if that makes sense? It's wild to be able to conceive of APIs to plain text, and natural language queries on structured APIs, but that's what we've g
70.
by genmon 4y ago
I hadn't noticed that, very much appreciated.
71.
by genmon 4y ago
No way! This is incredible. That h1 "Hello," <3 Is the tagging manual? It's really good.
72.
by genmon 4y ago
It turns out that "closeness" is usually hard to visualise/explore when you're dealing with a 1,000-dimensional space... and PCA has the failures mentioned above. It's weird -- it's locally useful to navigate,
73.
by genmon 4y ago
Colour is the 3rd component -- I wanted to see the difference between overlapping episodes. As for the 3D plot... here you go! https://interconnected.org/more/2023/03/in_our_time-PCA-3D-p... Basic PCA + Plotl
74.
by genmon 4y ago
Some unlinked features... If you put the Dewey division in the URL, the directory auto-opens. e.g. here are episodes about prehistoric life (my current jumping-off point) https://genmon.github.io/braggoscope/directory#5
75.
by genmon 4y ago
Now that's an interesting regression! I don't remember seeing it there before. (The worst I've noticed before has been Lawrence of Arabia under History of the Ancient World. Very much 20th century really.) Several other class
76.
by genmon 4y ago
I've been considering this, but my assumption is that it would be tripped up by the specialist words. I wonder... is there a way to "prime" Whisper (e.g. with the embedding of the episode synopsis) so that it "listens ou
77.
by genmon 4y ago
There's more on the About page but in summary: extracts the synopsis, guests (name, affiliation), and reading list (title, author, publisher, year) as structured data. Not massively hard with a bit of web scraping, but tedious and resu
78.
by genmon 4y ago
One of the prompts is Extract the description and a list of guests from the supplied episode notes from a podcast. Also provide a Dewey Decimal Classification code and label for the description Return valid JSON conforming to the
79.
by genmon 4y ago
In the strict definition, it's deterministic: you get the same response for the same prompt, each time (given the exact same model). But the prompt is parameterised. The bulk of the prompt is requesting a list of guests and speakers to
80.
Show HN: BBC “In Our Time”, categorised by Dewey Decimal, heavy lifting by GPT(genmon.github.io)
688 points by genmon 4y ago|171 comments
81.
by genmon 4y ago
Wouldn't that be similar to the parent comment, but outputting `.Content` in the range loop?
82.
by genmon 4y ago
Site creator here. I was just submitting this as a Show HN! I'm a big fan of In Our Time. There are episodes on everything from Cleopatra to the evolution of teeth to plasma physics, all available to listen to, so it's my starting
83.
by genmon 4y ago
Because it’s enormously dangerous to stand a meter from, and people near it won’t even know (eg it could have been picked up in a tyre tread so someone is carrying it around), but radiation and therefore ability detect drops according to th
84.
by genmon 4y ago
There’s open as in “can make” and there’s open as in “can repurpose”. I believe the parent is talking about the latter. The great thing about HTML is that it is machine readable by default, and that allows for permissionless innovation like
85.
Show HN: Action Cat – a spreadsheet-enabled voice bot for Zoom/Google Meet(interconnected.org)
1 points by genmon 4y ago|0 comments
86.
by genmon 4y ago
For anyone interested in this topic, I went back and looked at the earliest von Neumann feasibility paper recently, and came away wondering… what if we are the von Neumann probes https://interconnected.org/home/2022&#
87.
by genmon 4y ago
It works in the browser. For example, my RSS feed is processed with XSL to provide a friendly landing page: https://interconnected.org/home/feed (view source, second line)
88.
by genmon 4y ago
This is a cool story about reverse engineering a protocol… combined with some really unnecessary (and sexually graphic) insults of the Stem Player/Kano founder. The combined hardware and service seems to be highly experimental, and thi
89.
by genmon 4y ago
Serious q for any astrophysicists here: could the JWST resolve any surface features of Proxima Centauri? (In infrared of course.) Would be mind-blowing to see if so.
90.
by genmon 4y ago
Working Copy has got it right: a purchase unlocks all features to date, plus any new Pro features for the next year. After that point, you don't get any new Pro features -- but you don't lose the ones you paid for. The app is unde

More ›