2 HN users recommended this talk2 months2016–2024score 2
Years with an HN recommendation, 2007–2026
HN cites it in ECC memory debates and bit-flip discussions as concrete evidence that memory errors have real security consequences.
Artem Dinaburg's DEF CON 19 talk on bit-squatting, where flipped bits in DRAM cause devices to contact domains that differ by one bit from legitimate ones. He shows how registering such domains can hijack traffic without exploiting any software bug.
“There was a good Defcon talk about some security implications of this phenomenon” (HN, 2021)
I'm sure all of you who love debating ECC will enjoy this defcon video: [this video] "DEFCON 19: Bit-squatting: DNS Hijacking Without Exploitation (w speaker)"
Google is quite famous for running their data centers hotter than others. * Title makes it obvious, but 95f : [link] * Increased in dns request failures(likely due to said heat) and bad routing cause internal iGoogle services to request this guy's stuff: [this video]
Bit flips causing requests to bit-squatted domains
Depends on how you encrypt over HTTP. You can achieve similar benefits of verification with PGP and similarly rich object level encryption. Be careful of gotchas in how PGP or similar work, if you choose it (notable what is or isn't plaintext). Object level encryption typically allows for better separation of concerns as it does not bottleneck to a unique domain:port to seperate encryption contexts. It also eases separation of public facing encryption risks from private internal ones. If a public server is breached, all HTTPS traffic can be read, but if it proxies a more secure, perhaps simpler application environment (perhaps a microservice for the specific functionality), then that application environment is not affected by vulnerabilities in unrelated public facing features (like a backup microservice might not care about how to render HTML/JavaScript, so wouldn't be vulnerable to common web XSS attacks). It allows other benefits like: - it faciliates only user can decrypt data being sent (good for backups) - intermediate services to load balance and block DOS don't have to be exposed to the plaintext data, only the necessary service that needs it - many to many encrypted messages over one connection (TLS is many to 1 decrypting endpoints) - intermediate caching and redelivery of messages can be done with ease - it likely is harder to be at risk of bitflips ( [this video] ) - similarly verification of the data with signatures reduces risk of data corruption being missed that ma
Bitflip risk cited as a point in discussion of encryption layers
On the post, if Linus is referring to low-end Xeon and workstation non-Xeon lacking ECC, I think it's completely idiotic. Nearly all infrastructure should be baking in ECC as standard rather than some "option" to not have silent, unknowable, probabilistic corruption that is realistically realizable on a near-daily basis given the orders of magnitude of network, storage and computational volumes of today (and the future) by ever lower tiers of end-users and common examples of network infrastructure. DEFCON 19: Bit-squatting: DNS Hijacking Without Exploitation (w speaker) [this video]
I want a laptop with 64 or 128 GiB of Chipkill ECC RAM. Any of this soldered-on, anti-R2R/-consumer Apple trend-copying is BS. This crap about "YAGNI" or "it costs too much" have no idea what they're talking about and don't understand the reasons or use-cases. At home, I may have a 96 core EPYC with 512 GiB of Reg'd ECC RAM, but it doesn't have the locality or reachability of something that's with you. Network presence and low-latency aren't a given. [link] Edit: Resources: [link] [link] [link] [PDF] [link] DEFCON 19: Bit-squatting: DNS Hijacking Without Exploitation (trivia: at the time of filming, I was excessively-hungover and laying across 6 seats in the middle of an empty area to the extreme right, from the podium's perspective. I did watch the presentation with sunglasses on indoors because my eyes weren't cooperating nicely with photons.) [this video] Edit follow-on: And I don't want to lug-around a 17" Russian tank laptop that needs wheeled tracks to move. 13-15" max.
Indeed. See this one for the results of Bit-squatting the Windows Update domains: "DEFCON 19: Bit-squatting: DNS Hijacking Without Exploitation" [this video]
Bit flips are totally real, at scale you will definitely see them on large queries. There was a fun talk at DEFCON on bitsquatting, the process of buying 1 bit off domain names and then accepting all incoming connections. Attacks like rowhammer similarly abuse erroneous bit flips. Supposedly microsoft can detect solar activity based on the number of windows crash logs they receive. DEFCON Talk: [this video] [link] [link]
> [this video] Interesting talk. I did find that <takes a drink> later in his talk he kept <takes a drink> taking a drink every 10 seconds or so <takes a drink> , which ended up being more than a <takes a drink> little <takes a drink> irritating to watch and listen <takes a drink> to.
Video list and Hacker News comment labels from HackerTube (prompted by @karpathy), released under ODC-By 1.0, built on the open-index/hacker-news archive. Hacker News data is provided by Y Combinator; comments remain the work of their authors. Video titles, channel names and thumbnails belong to their creators and are shown via YouTube. HackerTube Watch is an unofficial edition. Data export of 2026-10-31.
10 Hacker News comments
From the HN archive; up to 20 kept per video, favoring different commenters. Recommendations first.
Citations, criticism and jokes
Read more
Read more