All my own OSS stuff I always release MIT licensed because I want to be able to use the libraries in my closed source job.

Can you explain more? Don’t leave me hanging…

(((Truth)))

https://arewereorganizedyet.com/ lol already updated

I even read this aloud in my head as “CREATE ZE VUCKING FILE” in a particularly bad German accent same as over 20 years ago when a friend I worked for drilled it in my head.

Don’t drive like my brother!

Lord you can see it’s true

I think perforce is mostly used. There certainly should be a changelog but no text diff really.

Sikhs want Punjab to be its own country, separate from India. India post Congress is ruled by Hindu nationalists who want to rewrite history books and erase anything positive that happened in precolonial India (which wasn’t really a country in the modern sense, but a changing set of integrated states) under non Hindu rule. The BJP are fascist shit stains and literally ordering hits on Sikh activists abroad for trying to promote Sikh separatism. There’s really not much more nuance than that. Fuck Modi.

Did you just post your open ai api key on the internet?

It reminds me of a Stanley Mouse graphic used for a lot of Grateful Dead art

In addition to iroh there is https://github.com/earthstar-project/meadowcap-js and other projects under https://github.com/earthstar-project. See https://willowprotocol.org/more/projects-and-communities/index.html#projects_and_communities

https://github.com/Mozilla-Ocho/Memory-Cache is the actual project if you want to use it.

Basically it’s a firefox extension to save a page as a pdf in a directory that is symlinked to your local PrivateGPT install which then ingests the docs. It doesn’t seem to me that it provides any in-browser querying of PrivateGPT but I haven’t tried setting it up to confirm that.

I think that is overly simplistic. Embeddings used for LLMs do definitely include a concept of what things mean and the relationship of things to other things.

E.g., compare the embeddings of Paris, Athens, and London to other cities and they will have small cosine distance between them. Compare France, Greece, and England and same. Then very interestingly, look at Paris - France, Athens - Greece, London - England and you’ll find the resulting vectors all align (fundamentally the vector operation seems to account for the relationship “is the capital of”). Then go a step further, compare those vector to Paris - US, Athens - US, London - Canada. You’ll see the previous set are not aligned with these nearly as much but these are aligned with each other (relationship being something like “is a smaller city in this countrry, named after a famous city in some other country”)

The way attention works there is a whole bunch of semantic meaning baked into embeddings, and by comparing embeddings you can get to pragmatic meaning as well.

It varies on who does the interview but I push for much simpler than leetcode type stuff- e.g. not puzzle problems but more “design a program that can represent a parking structure and provide a function that could be used for the ticket printer to determine where a new car should park, as well as one that can run upon exit to determine payment”

Then if they are actually solid we can dive into complexity and optimization and if they can’t write a class or a function at all (and esp if they can’t model a problem in this way) it’s really obvious.

I’m betting you aren’t involved in hiring? The number of engineers I’ve interviewed with graduate degrees from top universities who are fundamentally unable to actually write production quality code is mind-boggling. I would NEVER hire somebody without doing some panel with coding, architecture/systems design, and behavioral/social interviews.

I agree. Family of 5 many hotels require us get 2 rooms. Plus no option to cook meals makes for a much more expensive stay usually. At least until a few years ago when airbnb went insane with the cleaning fees plus cleaning requirements and all that nonsense.

Many (14?) years back I attended a conference (now I can’t remember what it was for, I think a complex systems department at some DC area university) and saw a lady give a talk about using agent based modeling to do computational sociology planning around federal (mostly navy/army) development in Hawaii. Essentially a sim city type of thing but purpose built to help aid in public planning decisions. Now imagine that but the agents aren’t just sets of weighted heuristics but instead weighted heuristic/prompt driven LLMs with higher level executive prompts to bring them together.

I fully agree with this, would have written something similar but was eating lunch when I made my former comment. I also think there’s a big part of pragmatics that comes from embodiment that will become more and more important (and wish Merleau-Ponty was still around to hear what he thinks about this)

A lot of semantic NLP tried this and it kind of worked but meanwhile statistical correlation won out. It turns out while humans consider semantic understanding to be really important it actually isn’t required for an overwhelming majority of industry use cases. As a Kantian at heart (and an ML engineer by trade) it sucks to recognize this, but it seems like semantic conceptualization as an epiphenomenon emerging from statistical concurrence really might be the way that (at least artificial) intelligence works