Hacker News: andrewguenther

New comment by andrewguenther in "GPT-5.4"

andrewguenther — Thu, 05 Mar 2026 22:35:39 +0000

It looks like this doesn't work for users without accounts? It works when I'm logged in, but not logged out. I went ahead and reported it to the team. Thanks for letting us know!

New comment by andrewguenther in "ChatGPT Atlas"

andrewguenther — Tue, 21 Oct 2025 22:13:36 +0000

How long ago was this? I had this same experience, but there's a new implementation for memory as of a few months ago which seems to have solved this weird "I need to mention every memory in every answer" behavior.

New comment by andrewguenther in "Rubygems.org AWS Root Access Event – September 2025"

andrewguenther — Thu, 09 Oct 2025 18:07:10 +0000

In 2025 there's no reason for anyone to be logging into an AWS account via the root credentials and this should have been addressed in the preventative measures.

There's no actual control improvements here, just "we'll follow our procedures better next time" which imo is effectively doing nothing.

Also this is really lacking in detail about how it was determined that no PII was accessed. What audit logs were checked? Where was this data stored?

Overall this is a super disappointing postmortem...

New comment by andrewguenther in "Sora 2"

andrewguenther — Tue, 30 Sep 2025 17:42:42 +0000

Yes, you can start with a still and a prompt

New comment by andrewguenther in "Rail travel is booming in America"

andrewguenther — Sun, 21 Sep 2025 21:50:57 +0000

It's not an enforcement issue so much as it is a heavily exploited loophole. Part of the reason freight trains are so long is so that they can't fit in passing sidings. Since Amtrak does fit, they end up having to yield because the freight trains simply cannot.

Could this be fixed by legislation on max train length to ensure all trains fit in sidings? Yes. Will that legislation get passed? No.

An interesting video on the subject: https://youtu.be/qQTjLWIHN74?si=t3u3iyZj1kRQQUCe

New comment by andrewguenther in "OpenAI: Scaling PostgreSQL to the Next Level"

andrewguenther — Sat, 24 May 2025 00:15:24 +0000

How does that statement convey inexperience? The presentation doesn't try to present that as a novel conclusion, it's just a true statement.

New comment by andrewguenther in "SREBench Competition"

andrewguenther — Fri, 16 Aug 2024 05:25:06 +0000

Hardly fair when most commands don't work and you can't copy paste...

New comment by andrewguenther in "October 2025 will be a support massacre for a bunch of Microsoft products"

andrewguenther — Wed, 24 Apr 2024 14:46:57 +0000

None. Windows Defender is actually quite good and the third party options are so bad I'd argue in many cases you're worse off having them installed.

New comment by andrewguenther in "S3 is files, but not a filesystem"

andrewguenther — Sun, 10 Mar 2024 20:29:48 +0000

Yes, but S3 has single region redundancy that is better than GCP. Your data in two AZs in one region is in two physically separate buildings. So multi-region is less important to durability.

New comment by andrewguenther in "Fck-nat: The (f)easible (C)ost (k)onfigurable NAT"

andrewguenther — Fri, 02 Feb 2024 23:04:30 +0000

Also if you're open to it, I'd love to reach out and get some feedback on your experience with fck-nat, how many environments you're running it in, etc. Is that something you'd be open to? If so, what's a good way to get in touch?

New comment by andrewguenther in "Fck-nat: The (f)easible (C)ost (k)onfigurable NAT"

andrewguenther — Fri, 02 Feb 2024 23:02:59 +0000

Happy to hear it! I've been looking to add some testimonials to the site. Is it okay if I reach out to you via email?

New comment by andrewguenther in "Fck-nat: The (f)easible (C)ost (k)onfigurable NAT"

andrewguenther — Fri, 02 Feb 2024 23:00:48 +0000

I don't want to get anyone too hyped on it just yet because there's still a lot of testing to be done, but hopefully will start giving some more concrete updates on "fck-nat 2.0" soon-ish!

New comment by andrewguenther in "Fck-nat: The (f)easible (C)ost (k)onfigurable NAT"

andrewguenther — Fri, 02 Feb 2024 21:55:02 +0000

Author of fck-nat here. Happy to answer any questions! Thanks for the post!

New comment by andrewguenther in "Fck-nat: The (f)easible (C)ost (k)onfigurable NAT"

andrewguenther — Fri, 02 Feb 2024 21:53:14 +0000

Author of fck-nat here. My big issue with Alternat is that it actively updates the route table which can still cause availability problems. It's a shorter outage than the current fck-nat replacement methodology, but it is still dropping connections.

The longer term vision for fck-nat is a two node approach using conntrackd and keepalived to actively failover existing connections to the secondary with no loss of availability. This has the added benefit of not requiring all of the auxiliary infrastructure that Alternat sets up.

New comment by andrewguenther in "Alaska 737 cockpit voice recorder data erasure renews safety debate"

andrewguenther — Mon, 08 Jan 2024 21:14:33 +0000

It doesn't clear every two hours, it's a rolling two hour recording. Since the plane was still operational, it continued recording and it was just over 2 hours before maintenance crews disabled the recorder.

New comment by andrewguenther in "Privacy is priceless, but Signal is expensive"

andrewguenther — Fri, 17 Nov 2023 00:33:01 +0000

I feel like they've made their position here quite clear and it's well-reasoned. I understand the disappointment, but this doesn't give any indications to me of "serial mishandling" unless there's some other context I'm missing?

Open source != open to contributions. Signal has made it pretty clear that their motivations for open source are visibility and verifiability, not to get people to do work for them for free. It seems like the action item to update the CONTRIBUTING.md to make those expectations more clear is a reasonable one.

New comment by andrewguenther in "US Government issues first-ever space debris penalty to Dish Network"

andrewguenther — Thu, 05 Oct 2023 19:22:30 +0000

The regulatory vector for satellites in the US is the FCC. The FCC controls US airwaves so if you're providing services via satellite in the US you must have a license from the FCC.

So if you're a non-US company operating services in the US, you'll have to have a US subsidiary anyway and that subsidiary would be required to get the license. If you're a non-US company not doing business in the US then that's outside of regulatory jurisdiction.

New comment by andrewguenther in "Language Models Represent Space and Time"

andrewguenther — Wed, 04 Oct 2023 18:49:29 +0000

No offense taken. I tried to be clear in my phrasing that I don't personally subscribe to the "LLMs are just like us" mentality. Just making an observation as to why people have such visceral reactions to any implication that they might be.

New comment by andrewguenther in "Language Models Represent Space and Time"

andrewguenther — Wed, 04 Oct 2023 16:52:29 +0000

I think LLMs are presenting some uncomfortable philosophical questions for people about how our own brains work and admitting that there is any kind of "intelligence" (even if very basic) in an LLM is an admission that our own brains may work in a similar manner.

New comment by andrewguenther in "Reclaim the Internet with Mozilla"

andrewguenther — Mon, 25 Sep 2023 19:41:26 +0000

I'm so sick of Mozilla's constant posturing as the "open internet" champion. Unfortunately it seems to be working based on the hordes of goons who interject in any discussion of web browsers that Firefox is the "morally superior" choice.

Mozilla is lost. Not just directionless, but in my opinion a total shell of the organization they once were. Chrome's complete dominance can be directly attributed to Mozilla's mismanagement.

Just continuing to burn funds on dishonest virtue signaling and failed experiments...