yodon

10 ms·

91.
by yodon 8mo ago
Is there a "detect infection and clean it up" app from a reputable source yet (beyond the "version 8.8.8 is bad" designator)?
92.
by yodon 8mo ago
If customers start asking Claude first, before they ask Thomson Reuters, that's a big risk for the later company.
93.
by yodon 8mo ago
Status page reports it's back up
94.
by yodon 8mo ago
This looks amazing! Curious if you (or others) have dug into the upcoming claude swarms feature? It looks like that would let you remove the dependency on beads, as claude seems to be getting native task tracking and inter-agent messaging c
95.
by yodon 8mo ago
> "Most" There are news stories that are of particular interest and relevance to the type of system designer that makes up much of the HN audience. The community should be allowed to discuss them intelligently and politely.
96.
by yodon 8mo ago
>Why would you share this here? These are clearly important times, regardless of your politics. The HN guidelines define on topic as "On-Topic: Anything that good hackers would find interesting." Up-votes are a measure of what
97.
Not a Drill(nytimes.com)
18 points by yodon 8mo ago|9 comments
98.
by yodon 8mo ago
Steve Yegge is hella smart, and I've spent many hours digging into his recent work on GasTown and Beads, but he needs to read up on business strategy. I'd recommend starting with Stratechery's articles on on Platforms and Agg
99.
by yodon 8mo ago
I read it. He used many words. Did he say anything?
100.
by yodon 8mo ago
This isn't a taxes issue. The assault on higher education and the sciences by this administration is inseparable from the assault on minorities and free speech. This is the authoritarian playbook 101. Mao went as far as locking up all
101.
by yodon 9mo ago
It's still worth knowing that this letter was written, even if the letter is only significant because of how embarrassingly little it says.
102.
by yodon 9mo ago
It's hard to imagine a joint letter that says less than this one does.
103.
by yodon 9mo ago
What is the risk profile when running untrusted code in a GitHub codespace under VS Code (other than access to and env vars or secrets attached to the code space)?
104.
by yodon 9mo ago
Are there plans for an academic paper on this? Super interesting!
105.
by yodon 9mo ago
Is anyone running Claude in a GitHub Codespace container? There was this HN post[0] last week on a tool for automatically shutting down the codespace container when idle. [0] https://github.com/wandb/catnip
106.
by yodon 9mo ago
>I'm making a joke about... Probably a good time to read up again on Poe's Law[0], which does always pertain to attempted sarcasm in online forums. [0] https://en.wikipedia.org/wiki/Poe%27s_law
107.
by yodon 9mo ago
Genuine question: Is this karma farming? Why would one try to karma farm on a site where karma brings no value (other than a low one-time threshold where you gain an unimportant power)?
108.
by yodon 9mo ago
Clearly other HN readers consider news of as many as 12,000 people possibly having been killed by their government to be important, and consider the discussion happening about it on this page to be interesting to them.
109.
by yodon 9mo ago
Sounds similar to steve yegge's beads[0] 0: https://github.com/steveyegge/beads
110.
by yodon 9mo ago
Re: #2, I think I was looking for something different and likely simpler than what you were thinking... The session would still be running in a codespace, I just want to be able to start on my phone, then move to my Mac, which has a nicer k
111.
by yodon 9mo ago
Nice - Two questions (1) Are you able to send a push message to the phone when Claude Code is waiting for a permission approval or other user input? (2) Any thoughts on making this available on desktop as well, so we can access the session
112.
by yodon 9mo ago
> The thing is, you could probably create something FAR more horrific by simply mixing spent nuclear fuel with TNT... Nope. It turns out you need astronomical amounts of spent waste to noticeably impact a large population. The trial of J
113.
by yodon 9mo ago
Perhaps being more intentional about adding a use case to your original prompts would make sense if you see that failure mode frequently? (Practicing treating LLM failures as prompting errors tends to give the best results, even if you feel
114.
by yodon 9mo ago
> I have never came here for the moderation. The moderators (and the algorithm they support and tune) are why the conversation on HN is compelling enough to attract 99th percentile experts on just about every subject.
115.
by yodon 9mo ago
The real question is "could she have written that problem statement a year ago, or is a year's learning by the team baked into that prompt?" There are lots of distributed agent orchestators out there. Rarely is any of them go
116.
by yodon 9mo ago
Fond memories of Sun workstations in the 80's showing images like these when no one was logged into the console.
117.
by yodon 9mo ago
HN hug of death? Can't open the site
118.
by yodon 9mo ago
Do you have a citation for that 20 years estimate?
119.
by yodon 9mo ago
>Economics rules everything Physics rules everything, once you start trying to run at scale. The density of carbon per unit volume in solid materials of interest doesn't vary that much, whether you sink it in trees or in exotic mate
120.
by yodon 9mo ago
Viewing your site from a phone, I can't see any indication of what the models look like or what their quality is. You should probably put some time into communicating what your app does. At present, with no sample images, there is noth

More ›