Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
cube00
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
31 ms
·
631.
▲
by
cube00
1y ago
Just goes to show how greedy AWS has become with pricing now that everyone is locked in. If anything they should be able to use their scale to make it cheaper than doing it on-prem and all the costs that go with that. Strangely at the time
632.
▲
by
cube00
1y ago
Good luck getting past the sign up flow to even get an account if your fingerprinted device is on their ban list.
633.
▲
by
cube00
1y ago
The eternal AI hype excuse, the next model will fix that, now we're seeing new models hallucinating more then ever.
634.
▲
by
cube00
1y ago
Unleashing a hallucinating LLM to make edits will creates so many subtle problems on such a scale that it may not be possible to clean it up once other edits are made on top.
635.
▲
by
cube00
2y ago
> Grok is FIPS complaint, right? Safe for use with confidential information? PII/PHI? “Laws without enforced consequences are merely suggestions.” -- Ron Brackin
636.
▲
by
cube00
2y ago
I'm surprised you can tolerate the mobile website at all with the non stop full screen "better in the app" modals.
637.
▲
by
cube00
2y ago
Assuming the space was being consumed by attachments, DAOS can help with this. There's reports [1] of people getting up to 2TB once you get the attachments out of the NSF. [1]: https://ds_infolib.hcltechsw.com/ldd/
638.
▲
by
cube00
2y ago
I only just came across his speech to Congress in 2019, highly recommend it, https://www.youtube.com/watch?v=_uYpDC3SRpM
639.
▲
by
cube00
2y ago
> “You are not the first person to ask for us to give them fire retardant. It happens. It’s not something we do.” Scary to think what other discoveries were missed if those other investigations had been given the samples they asked for.
640.
▲
by
cube00
2y ago
I'm so glad someone dug into this properly after the tj-actions maintainer started locking threads and refusing to look into how the PAT was leaked. https://github.com/tj-actions/changed-files/issues/2464
641.
▲
Attack on Coinbase Expanded to the Widespread tj-actions/changed-files Incident
(unit42.paloaltonetworks.com)
3 points
by
cube00
2y ago
|
2 comments
642.
▲
by
cube00
2y ago
The vibe bros would have you believe your prompt is at fault and that you need add "don't make typos".
643.
▲
by
cube00
2y ago
> because in 6 months when the AI labs release a more capable model How many years do we have to keep hearing this line? ChatGPT is two years old and still can't be relied on.
644.
▲
by
cube00
2y ago
> Vibe coders focus on writing tests From the boasting I've seen, Vibe coders are also using AI to slop out their tests as well.
645.
▲
by
cube00
2y ago
> I have friends who write a lot of this code for tasks like data analysis for retail and "vibe coding" isn't that crazy in such a domain Considering the hallucinations we've all seen I don't know how they can be
646.
▲
by
cube00
2y ago
I'd almost forgotten what it's like for a blog post to embrace that it's a short story and not drone on for pages to pad it out.
647.
▲
by
cube00
2y ago
AI heavy now
648.
▲
by
cube00
2y ago
> But no. It's just pure incompetence. While I don't agree with how he conducts himself on the mailing lists; I respect he uses the same language when talking about himself.
649.
▲
by
cube00
2y ago
It looks like teachers are actually creating presentations in minutes with this tool. I'm an Aussie teacher, but in the same boat as everybody else. Swamped and time poor. Recently we had some pastoral care activities for National S
650.
▲
by
cube00
2y ago
> The teacher generating the presentation will review the content for accuracy before presenting it First header on front page: "Craft Engaging Lessons in Minutes." Further down: "Seriously. You will be done prepping a
651.
▲
by
cube00
2y ago
It's a ChatGPT wrapper, it is not version 1.
652.
▲
by
cube00
2y ago
This feels incredibility risky, especially given you're expecting users to put in their own efforts editing the slides. That could all disappear tomorrow if you decide to wind up this business.
653.
▲
by
cube00
2y ago
It's heart breaking to think a whole generation will soon be brought up on learning from AI generated slop. Here's my attempt at using this tool: Learning Intention: To understand the differences between MP3 and WAV audio formats
654.
▲
by
cube00
2y ago
Depends on your SecOps. Ours shuts down apps with critical vulnerabilities if they're not patched within 48 hours.
655.
▲
by
cube00
2y ago
> This stability does mean that old React (or Knockout, or whatever) applications will still work just fine for the end users, likewise without a single line changed. Not in the current enterprise cyberops environment of needing to pass
656.
▲
by
cube00
2y ago
After giving me continuous wrong answers ChatGPT decided it would try allow me to indulge it in a "learning opportunity" instead. I completely understand your frustration, and I genuinely appreciate you pushing me to be more accu
657.
▲
by
cube00
2y ago
Considering Go is the only language with a garbage collector out of the three languages you mentioned, I'm not sure how you reach the conclusion they're all as close to the metal. C and Rust both have predictable memory behaviour,
658.
▲
by
cube00
2y ago
Great until your dependency scanner picks up a vulnerability in the framework or a dependency that can't be upgraded without breaking v4, no exemptions can be given and the only updated version is 5. [1] [1] https://endoflif
659.
▲
by
cube00
2y ago
Make sure you go into that eyes wide open, I misguidedly thought there was some communication between Bitwarden and the open source vaultwarden but there is not. I've been burnt by things breaking as Bitwarden updates the client and va
660.
▲
by
cube00
2y ago
I used to think I had low quality EMs because they didn't have an IT background. However I started coming across EMs with IT backgrounds that had completely checked out of anything technical. They'd let their source control licenc
More ›