qudat

21 ms·

361.
by qudat 2y ago
Try perplexity. It’s the only LLM use case that actually worked for me. If you want to research a topic it is awesome
362.
by qudat 2y ago
Try perplexity and then come back and tell us how you feel
363.
by qudat 2y ago
> The 0.05 threshold is indeed arbitrary, but the scientific method is sound. Agreed. A single published paper is not science, a tree data structure of published papers that all build off of each other is science.
364.
by qudat 2y ago
I'm guessing the syncing process between instances is really brutal on resources. Imagine constantly syncing external databases for your service to function properly.
365.
by qudat 2y ago
That's a really good point! That might work if you have a single VM but it's a little more complicated when you have an app on multiple instances. pipe is a multicast pubsub which means you can have many subscribers.
366.
by qudat 2y ago
ActivityPub seems to require a lot of hardware resources in order to run properly, which is unfortunate. It’s not something I would ever want to run myself, especially to the public.
367.
by qudat 2y ago
Here's my one-weird-trick: On my VM, keep this running: while true; do ssh pipe.pico.sh sub deploy-app; docker compose pull && docker compose up -d; done On my local machine: docker buildx build --push -t ghcr.io
368.
by qudat 2y ago
Agreed on all fronts. Go has nailed the stdlib and every other language is in its shadow
369.
by qudat 2y ago
While not directly competing with mqtt, https://pipe.pico.sh is a great pubsub tool that leverages ssh for communication, effectively creating an authenticated, networked *nix pipe system. It tries to be the simplest way to send
370.
by qudat 2y ago
Checkout https://pipe.pico.sh which is a system for networked Unix pipes using ssh.
371.
by qudat 2y ago
I think that’s a smart approach. You might be interested in https://starfx.bower.sh which has similar design goals
372.
by qudat 2y ago
I had the same thought recently: https://bower.sh/on-writing
373.
by qudat 2y ago
That’s a wild take
374.
by qudat 2y ago
One core issue is that we need to convert spoken/written languages (e.g. english) into more formal math languages since sometimes the underlying mathematical problem is written using prose. The example in the paper: > When Sophie
375.
by qudat 2y ago
We just released a go pkg that uses the new iter pkg. We were so excited by the interface in large part because of how simple iterators are to use. https://github.com/picosh/pubsub/blob/main/pubsub.go#L18
376.
Show HN: Pipe – Authenticated Pubsub over SSH(pipe.pico.sh)
11 points by qudat 2y ago|0 comments
377.
by qudat 2y ago
That's interesting because the Genealogy of Morality was his greatest contribution as far as I can tell. Is there something in particular that feels most egregious about master/slave morality or do you have some reading attempting
378.
by qudat 2y ago
I think your technique is working because a lot of people got triggered by your friend test. I kind of love it.
379.
by qudat 2y ago
The use API is very simple; JSX is a bastardized version of html that is atrocious to read and write. Just imagine all the dev time spent dealing with jsx/tsx. From IDE support to babel, typescript, etc. it is a burden on the JS/T
380.
by qudat 2y ago
JSX is the worst part of react so I find that to be a pretty wild take.
381.
by qudat 2y ago
The biggest issue was with using older versions of ruby or older versions of the bundler itself. We had to pin our bundler to a specific version and it caused all sorts of issues when installing gems.
382.
by qudat 2y ago
This is such a wild take. I’m constantly running into issues with bundler and literally never have issues with npm or yarn. Not to mention the amount of innovation happening within that world with module resolution algorithms.
383.
by qudat 2y ago
Iteration speed is significantly fast on the client. Perf is an afterthought — for better or worse
384.
by qudat 2y ago
Companies love it because it’s an all-in-one knowledge base
385.
by qudat 2y ago
I see this as a pruning process and an inevitable part of science. But I would further argue against what others were saying about personal ethics. Science must remove the human as much as possible from the process.
386.
by qudat 2y ago
While I agree this is a big problem, science should never be defined by a single article. I was always taught that science is a tree of knowledge where you build off previous positive results, all of which collapse when an ancestor turns ou
387.
by qudat 2y ago
> I hate that the AI pundits have succeeded in popularizing the notion of "hallucination", anthropomorphizing these balls of statistics into something that seems like it's actually in some sort of deep thought process akin
388.
by qudat 2y ago
The default IDE workflow is like a python "notebook" where code can and is run in whatever order the creator wants. Every R code I've read treats it as such and it results in an absolute mess to read and manage.
389.
by qudat 2y ago
Do you actually know that’s what’s happening? The details are extremely fickle the last I read (a couple days ago). For all we know, they are doing model routing and prompt engineering to get o1 to work.
390.
by qudat 2y ago
In the primary CoT research paper they discuss figuring out how to train models using formal languages instead of just natural ones. I'm guessing this is one piece to the model learning tree-like reasoning. Based on the quick searchin

More ›