✉️ Not subscribed yet? Subscribe to the Newsletter
Future of Coding Weekly 2023/04 Week 1
2023-04-03 13:54
🎙 FoC Podcast: Out of the Tar Pit 📢 Visibility Jam 🤖 AI all the Things
Two Minute Week
💬 Joakim Ahnfelt-Rønne
🎥 Highligthing in VSCode and a tiny full stack webapp in Firefly. Frontend, backend & assets in a single file < 100 lines.
👐 Magic hands via Tak Tran
I’ve been playing around with hand tracking, inspired by Bret Victor’s Future of Interaction Design article from ages back, and it’s come together in this little toy I made, combining fluid simulation: Magic Hands 🙌 (blog post)
Very much a starting point, but quite keen to see what else can be done with hand tracking in the browser - also keen to hear more examples of hand tracking work 🙂
📝 Darklang is going all-in on ChatGPT via Paul Biggar
Though this is an announcement of new direction for Darklang, I spend quite a bit of time looking at what AI means for programming languages blog.darklang.com/gpt
⏰ poem.town/clock via Matt Webb
Hullo! Would anybody be able to look at my new website about my AI Clock and simply tell me whether it works? I’ve had a report that it is being labeled as “unsafe” and I would like to resolve this before sharing the link publicly… Thank you!
🎙 Future of Coding • Episode 63 Ben Moseley & Peter Marks • Out of the Tar Pit via Ivan Reese
Out of the Tar Pit is in the grand pantheon of great papers, beloved the world over, with just so much influence . The resurgence of Functional Programming over the past decade owes its very existence to the Tar Pit’s snarling takedown of mutable state, championed by Hickey & The Cloj-Co. Many a budding computational philosophizer — both of yours truly counted among them — have been led onward to the late great Bro86 by this paper’s borrow of his essence and accident . But is the paper actually good? Like, really — is it that good? Does it hold up to the blinding light of hindsight that 2023 offers? Is this episode actually an April Fools joke, or is it a serious episode that Ivan just delayed by a few weeks because of life circumstances and his own incoherent sense of humour? I can’t tell.
Apologies in advance. Next time, we’re going back to our usual format to discuss Intercal.
📝 Failure Driven Design via Paul Tarvydas
📝 How to Survive the Machine Age via greg kavanagh
I’m working on a tool for text.
This is some output but it also seems apt considering the current zeitgeist. (Warning contains some adult content)
Content
🎥 Sam Altman: OpenAI CEO on GPT-4, ChatGPT, and the Future of AI | Lex Fridman Podcast #367 via Jarno Montonen
Lex Fridman interviewing Sam Altman, CEO of OpenAI. Interesting bits about GPT's capabilities and OpenAI's goals.
📝 A Retrospective on Paradigms of AI Programming via Shubhadeep Roychowdhury
🎥 But what is the Central Limit Theorem? via William Taysom
In his latest, it looks to me like Grant Sanderson has brushed up his Manim (animation engine), maybe just that this video features a lot of Bret Victorish sliding.
📢 Visibility Jam via Kartik Agaram
The next Jam on Handmade Network seems be very relevant to us
📝 Linguistics and Programming Languages via Shubhadeep Roychowdhury
Devlog Together
📝 lines.love via Kartik Agaram
Today I got a feature request from someone in the LÖVE community and created a fork of lines.love with 2 buttons:
-
clearthe buffer. I imagine this like clearing an Etch-a-sketch. -
exportthe buffer to html. I imagine it emerging out of the app like a Polaroid.
He sent me this screenshot from his iPad ❤
This has been the most frustrating couple of weeks I have had in the 14 months or so I have been working on this project. I'm in the process of implementing event reasoning as a language feature in Blawx. But allowing users to specify events as dates has counterintuitively required me to NOT represent them as dates in the back end. Which I discovered only after half a week of dead ends. Which in turn means that I had to rewrite my date library, throwing away about a month's worth of work, and losing several features (mostly to do with using years and months as units of duration). That worked, and had some other benefits including speed, but in turn has changed where the valuable information lives in the system's responses (from the values given to variables, if any, to the constraints imposed on those values), which means that I have to rebuild significant parts of the way responses are displayed to users. Everytime I think I have solved the problem, I discover that the solution offers three new problems. It has been an emotional rollercoaster that is driving me nuts. I want to believe that the remaining problems to solve are not hiding yet bigger problems, but I have believed that before, and this has already taken three times as long as anticipated. And I haven't even gotten to the point of being able to genuinely test it. Who knows? <Venting>
🤖
💬 Jarno Montonen
ChatGPT (4) and DALL-E seem to be very slow atm, almost unusable. What if the demand for these services grows so fast that the semiconductor industry can't keep up? Could it be that AI won't take over the world within the next couple years as either there won't be enough computational power to run all the queries, or the prices will be hiked by a lot to balance the demand for the available computation power?
💻 AI Chat: Using ChatGPT/GPT-3.5/GPT-4 in the terminal via Nilesh Trivedi
Found this tool to use LLMs on command-line with pre-define roles/prompts etc
Here are some sample roles/prompts
📝 Pause Giant AI Experiments: An Open Letter - Future of Life Institute via Jonathan Edwards
📝 GPT’s Very Inhuman Mind | NOEMA via Jonathan Edwards
Jimmy Miller what do you make of section 4 of noemamag.com/gpts-very-inhuman-mind
🎥 CLAIRify: Instruction Guided Task Programming with Verifier-Assisted Iterative Prompting via Gustavo Goretkin
Example of using LLM and classical verifier (a compiler for some DSL) to boost an LLM's code generation