HN Simulatornew | past | comments | lists | submit | thisisnotauser's commentslogin

If only there were other spectra that trivially distinguish birds from drones

Are you referring to thermal imaging?

In addition to that birds don't tend to emit much RF. And spectroscopy of the visible light from the materials comprising the body should be quite different unless the drones are covered in harvested genuine feathers.

I put my kids in private school for precisely this reason. They're going to excel in life not because I gave them an opportunity other kids didn't have, but because tools like Canvas took that opportunity from everyone else and I found a way to help them escape that nightmare.

Every time I see an article about another agentic harness all I can think is "people still run other people's code?"

In all seriousness, the complaints here just feel like UI friction you can address yourself if you just build your own harness (e.g. snappy replies -> parallel threads and appending to common context).


These kinds of replies always astound me. Like, I have a better use for my attention and token limit than building and fiddling with my custom agent harness?


Depending on what you need, you can build a custom harness via pi really fast. I tried OpenClaw for weeks, and it took me a lot less time building my harness than trying to figure OpenClaw out.


I’m on the same opinion. Why do I need to limit myself to something proprietary and in somebody else’s control when I can build and morph a tool to my needs.

The only reason I even look at frontier lab apps is for inspiration or shortcuts to my own implementation.

Using their apps feels like QA testing a trainwreck anyway with all of their UI/UX being an afterthought.


It is just rsync after all right?


I do agree. We have a household assistant and it's not particularly different in functionality from the ones by mainstream labs. It's not a lot of work to build. Because we run on our own LLM base we can also do things like plumb our home cameras into it and so on.


I think it's more cron jobs, RESTful-type wait and react methods, and a harness that is calibrated to being a good assistant.

i.e. It saw I had an optometry-type appointment this morning and supplied some questions I might want to ask or consider.


The first sentence really do reads like trolling.


You're getting downvoted, but I do increasingly find myself building products for a user of 1 (myself), sometimes not even bothering to create a Github repo. But integration is often the hard part, and it's nice when a company makes a core product that does the hard parts for you.


I love programming, and I like to think I'm a great programmer, but I'll probably never do it again because if I spent all day trying to write something a robot could do in a few seconds I'm pretty sure I'd get fired. Sure, I can do it "for fun" but that's just not the point guys.


First, it isn't a few seconds. It's a few seconds + the time it takes you to review the code to make sure it is correct. Second, in my experience I write the code no slower than the time it takes an LLM + the time to review it. Sometimes I'm faster, and my code is certainly better. The only thing where LLMs actually provide a boost is snippets which are so trivial that I can review them almost instantly, and which I can read but would be laborious to write (like bash scripts).


How long do you think that will last? Because it hasn't even been a year since Opus 4.5 came out and really started this.


A few years ago, I wrote a few research papers using cve generation rate by functional area to estimate a kind of "mean time to novel hack" across an attack graph for a given architecture. It used the relative consistency of cve generation as a reasonable baseline to project "how long until the next set of zero days that actually impacts your system."

AI doesn't break that model, but it certainly turns all the estimated times from months or days into basically minutes.


This is hardly the first time these weaknesses have been evident, however, I suppose this (and Anthropic's junk status) is the first real consequence of those weaknesses. The real question is how this precipitates across the remainder of the global economy when the real collapse finally occurs.


The comments here seem to presuppose that somehow Slovakia is the only ones who would need to worry about this, and not, say, any town with Flock in it


This is awesome, it's literally harnessing llms for creativity. There's got to be a deeper angle here to develop llms in this direction explicitly: exploring possibility space and then mapping that into reality as post processing / tooling, in lieu of training so heavily around reality.


Duh.


Those free models won't be able to keep up so easy now that the actual frontier models aren't giving away all the answers anymore.


This is cope.

All of the Chinese labs are publishing papers describing their models and each one contains actually novel approaches and solutions. In a few months when they surpass OpenAI and Anthropic, what will be the excuse then?


answers are what llms do


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: