HN user
Seems like some of those problems could be partially addressed. Such as a retractable metal nozzle that is sanitized with heat or steam after use, not using warm water, reducing spray velocity, adjusting angle of spray, limiting spray time and throttling subsequent sprays, etc.
Feel free to share resources, but I am speaking purely in terms of practicality related to my day to day.
I guess it probably depends on what you are doing. Outside of layers on top of these things (tooling), I personally haven't seen much progress.
Are people really doing coding with agents through PRs? This has to be a huge waste of resources.
It is normal to preempt things like this when working with agents. That is easy to do in real time, but it must be difficult to see what the agent is attempting when they publish made up bullshit in a PR.
It seems very common for an agent to cheat and brute force solutions to get around a non-trivial issue. In my experience, its also common for agents to get stuck in loops of reasoning in these scenarios. I imagine it would be incredibly annoying to try to interpret a PR after an agent went down a rabbit hole.
Seems like that is taking a very long time, on top of some very grandiose promises being delivered today.
These rodents were huffing vaporized ethanol for 16 hours a day at ~.2 BAC on weekdays and then they left them on weekends for four weeks straight. That is quite the bender.
Or just small specialized search engines for categories of knowledge with some shared advertising / revenue model.
Can always go back to the days of these:
In my experience, OneNote syncs more reliably than Obsidian. Although, I still prefer using Obsidian for other reasons.
LLMs will read it, and it will be distilled into the great pool of knowledge until the end of time.
A terrible typescript
There is an unwritten social contract here.
There isn't, no matter how many times you say there is.
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
Still, the author is making a deterministic leap by stating that they wouldn't have been X if Y.
That said, the arXiv is rolling out an experimental LaTeX-to-HTML
Some history: https://www.arxiv-vanity.com/
A lot of disease has environmental and intrinsic causes.
Categorizing disease as having a set of characteristics in the body doesn't make it solely rooted within those characteristics.
I think there will always be a desire to solve disease with a single pharmaceutical, as that would be an incredible thing. However the existence of drugs that do not work for everyone doesn't necessitate that the signals for characterizing disease are off or that a pharmaceutical isn't useful to some people.
Organisms are incredibly complex, even those composed of relatively few cells. Thinking there is always some magic bullet to address a biological or emergent phenomenon is probably not the answer, but neither is being dismissive of treatments that don't meet that criteria.
I wouldn't have gone down any of that path
Yet, if you actually believe in biological determinism, there is no way you could have taken anything but that path.
I have been using the internet since 1995, and I have yet to see the cycle. From where I sit, it has been more of just a steady trend toward compartmentalization and aggregation.
(Not to say that there aren't a large number of great projects off the main path.)
It sounds interesting, but it really feels like they are downplaying the risks here.
I'd be hesitant to put anything mixed in a DIY device with off the shelf peristaltic pumps into my body without some additional analysis.
If something like automated analysis was a possibility, then maybe this would be more alluring.
I have burned out at least 10 times in my career. The last time, it was like I lost half a decade, even though it didn't last quite that long. It was as if I had amnesia. I fell into a deep depression. I considered killing myself just because everything seemed so difficult and pointless.
Give yourself some more time. You might not emerge exactly as who you remember, but its possible to mostly recover from the pretty extreme cognitive impairment that you are likely struggling through.
It looks like you need special sensors and a lot of light in certain wavelengths. Would like to read the paper.
Around June of this year.
Just because we have receptors that have affinity for molecules that we call cannabinoids doesn't mean much in terms of evolution related to plants. There is a wide range of molecules that have a shape that bind to the wide range of receptors that are a part of what we call the endocannabinoid system.
it is much better are correcting grammar
Mmmhm
https://github.com/mendel5/alternative-front-ends
Some of these are open source and easy to run, like https://github.com/iv-org/invidious
No, the burden is on you. The web is public by default. If you join it and think there is a burden on your visitors to pay you by viewing some bullshit that you serve them from some megacorporation who sells their identity, you are the one being unethical.
Do you think that any significant number of users are circumventing advertisements on YouTube?
No one has the right to control my client just because I visited their server. If Google doesn't want my client, they can easily block me.
Acting like people running custom clients are thieves or something is ridiculous.