The Big Question in AI-driven developer experience: What's after Copilot? Everyone is convinced this is "just" the beginning but can't articulate what the next killer form factor will be. Increasingly convinced next step change won't be from optimal stuffing context windows
@swyx
-
Searching for canonical references to the bitter lesson
By
–
was looking for more references to the "bitter lesson" – is this the canonical source? http://
incompleteideas.net/IncIdeas/Bitte
rLesson.html
… -
Dealing with GPT-3 Limited Context Window Constraints
By
–
how do you deal with the limited context window of GPT3? afaik, ChatGPT a limit of 8092 tokens, whereas GPT3 is lower (3-4k?) any langchain solution to that sort of thing?
-
Amazon’s 1998 Memo Origins of Workflow State Tracking Models
By
–
so excited to find the origins of the workflow model in this 1998 internal Amazon memo – I think this directly led to AWS SWF and Temporal. it turned out that tracking state changes wasn't just an afterthought, but should be the basis of the entire thing
-
GeneralTask: Converting TodoList Tasks into Calendar Events
By
–
came across @GeneralTaskHQ https://t.co/pGx0rjCCMk on HN
— swyx 🐣 (@swyx) 23 décembre 2022
takes the todolist -> calendar cycle very literally pic.twitter.com/cW8zdvfPdgcame across @GeneralTaskHQ https://
news.ycombinator.com/item?id=336587
63
… on HN takes the todolist -> calendar cycle very literally -
Testing Model Guidance Levels Finding Optimal Sweet Spot
By
–
randomly had this page open so tried it on the published model. guidances at 7, 10, and 13 from bottom to top. looks like 10 is sweet spot
-
Overloaded Partition Terminology and Data Chunking Concerns
By
–
feel like the word "partition" is overloaded in this context but if i understand this correctly… yes i can split them up into arbitrary length chunks.. but thats just making one long list into a hundred smaller lists – feels like something that is a routine part of data eng
-
Iterative Data Extraction with Error Recovery and Restart Logic
By
–
great post! i have a variant of this kind of work that i do often but the "extract" develops over time – run `extract` on a list of items (api calls, google searches) and running it until error, then either iterating from error index or restarting from 0.
-
ACM Journal publishes article on the end of programming
By
–
The ACM Journal now has a thing out about “the End of Programming” and… f yeah https://
m-cacm.acm.org/magazines/2023
/1/267976-the-end-of-programming/fulltext
… -
Shifting Vibes: From Throwing Errors to Intentional Error Handling
By
–
it feels like the vibes are shifting away from throwing things willy nilly and passing errors as values instead. https://
news.ycombinator.com/item?id=340883
70
… consider linting against throws by default and force people to think twice if your language makes it too easy to throw