HNHacker News
TopNewBestAskShowJobs

delroth

3,137 karma · joined December 3, 2009

NixOS developer and infrastructure maintainer. Dolphin Emulator (https://dolphin-emu.org/) developer. Ex-infrastructure Security Engineer, now on a long break away from full time employment.
submissionscomments
delroth··on Laptop stolen from Pelosi's office during storming of U.S. Capitol, says aide
Against what threat model? The general answer is "yes, most of them, including all the ones being used as defaults in major distros" -- they will not allow a user that isn't in possession of the password to log back into the session.

They don't necessarily prevent against other threat models, like "the attacker dumps the laptop's RAM" (which you could technically protect against if you froze all session processes and encrypted their working set with a key held in a secure element).

delroth··on Linux Hardening Guide
Some of the recommended steps like removing machine-id to prevent fingerprinting are especially funny when the rest of the article perfectly describes how to get your system in a completely uniquely fingerprintable state via obscure knobs most people don't touch.

As you've mentioned, it's a hardening "guide" without specifying a threat model, so its value is very limited anyway.

delroth··on Leaked documents from Nintendo investigation of hacker in 2013
> 1). This group of modders was planning on profiting from their hack

That's wrong, or at least unsourced and unsupported by current facts.

> 2). Copyright exists and generally must be enforced lest the holder wishes to lose their rights.

That's also wrong.

delroth··on Manifest V3 now available on M88 Beta
And 1/4 (Safari) already only implements declarative request blocking, and has done that for years while being touted as the alternative for privacy conscious folks.
delroth··on Would you be willing to fund a Linux port to Apple Silicon?
Aren't you basically describing big.LITTLE, which has been standard in most mobile SoCs for years now? (And was probably introduced in Android before iOS.)
delroth··on Goodbye PowerDNS
You cannot run any internet service without collecting personal information. IP addresses are personal data per GDPR, and by serving TCP connections you are collecting IP addresses of your clients. In the case of DNS, your DNS query is also likely personal data in some cases, and it definitely is personal data when linked to a timestamp and an IP address.

There is a huge distinction to be made in collecting vs. storing. The 8888 privacy policy also calls out that some personal information is stored, for how long, and for what purpose. But most people who claim that they're not collecting personal information are just not being accurate in their privacy statements.

Disclaimer: googler, opinion is my own and not my employer's.

delroth··on Guidance to developers affected by effort to block less secure browsers, apps
> To protect our users from these types of attacks Google Account sign-ins from all embedded frameworks will be blocked starting on January 4, 2021.

(emphasis mine)

So I don't follow how this would have anything to do with banning youtube-dl, which doesn't require login? And as the blog post mentions, you can still bootstrap auth through a normal web browser, and pass the auth token to your command line / less secure browser / ... app.

(Disclaimer: I work at Google, not on anything related to this blog post or to your hypothetical scenario.)

delroth··on Open Letter from Facebook Content Moderators
My understanding is that there are certain legal requirements associated with content moderators who may have to review illegal content, especially around child abuse / sexual imagery. This is what I've been told in the past by friends in other companies working on content moderation -- I don't have any specific evidence for or against it, but it passes the smell test.

The whole thing in the article about Facebook suddenly trying to automate this part of the content moderation job seems to also point in that direction.

> Management told moderators that we should no longer see certain varieties of toxic content coming up in the review tool from which we work— such as graphic violence or child abuse, for example.

delroth··on Google Widevine Content Decryption Module DMCA
Repository mirror: https://archive.softwareheritage.org/browse/origin/directory...
delroth··on Dear ImGui – Bloat-free graphical user interface library for C++
No, it does not.
delroth··on The world needs nuclear power and we shouldn’t be afraid of it
Do you realize that a single reprocessing plant releases more than 10x more tritium in the ocean yearly than the amount currently contained at Fukushima? This has been happening safely for decades, under constant monitoring for impact on fish and other ocean life. The whole recent "Fukushima tritium release" news cycle is entirely FUD, and all nuclear safety experts agree that there is basically no risk involved. Please stop propagating the FUD.

(https://www.japantimes.co.jp/news/2020/02/25/national/social... has some of that information, but you can easily find more refs comparing tritium amounts at Fukushima vs. normal release at facilities like La Hague.)

delroth··on KeePassXC 2.6.1
Browser extensions have the benefit of being more resilient against phishing (since they can perform origin checks), which I would definitely recommend for most users.
delroth··on How to contact Google SRE by dropping a shell in Cloud SQL
As far as I can tell, there is nothing in the writeup that suggests the file wasn't there from the beginning and present on all Cloud SQL instances. I know this was a standard practice in a few of the products I worked on where we expected the first few layers of security/obfuscation would be peeled off by curious outsiders.

This seems much more likely than OP's exploration being detected in realtime, then someone SSHing into the container manually to put a message.

Disclaimer: I work at Google, not on Cloud SQL, and don't know anything specific about that greetings.txt.

delroth··on Epic Games has filed legal papers in response to Apple [pdf]
This is blatantly wrong. Apple's CPU and GPU designs have been years ahead of Qualcomm/Samsung/... for years now, especially regarding single core performance. Check any mobile device benchmark.
delroth··on Epic Games has filed legal papers in response to Apple [pdf]
You're generalizing your own opinion. If I could buy an iPhone and run Android on it, I would do that. Instead I'm stuck choosing between good hardware with crappy OS (Apple) vs. outdated hardware with an OS that fits my needs (all Android devices).
delroth··on Intel ousts its chief engineer, shakes up technical group after delays
I don't know where you're getting your info from, but Google salaries in Switzerland are extremely competitive with California. The average for software engineers at Google Zürich is around $300K at L5 I'd say. And the housing situation, while still very expensive, is a lot less fucked up than San Francisco.

Source: I'm a Googler in Zürich. Current total comp is at $320K after 3 years at L5.

delroth··on Storage Matters: Why Xbox and Playstation SSDs Usher in a New Era of Gaming
Instead now you have most online services allowing you to pre-download the game (and often the day 1 patches too if any) and just do a quick install on release day. The experience isn't significantly worse, it's just different. Even Nintendo's e-shop supports this, and god knows how far behind they are in terms of online feature support compared to everyone else.

I'm pretty happy to not have to keep a large amount of game boxes and worry about scratching discs too.

EDIT: To your specific example, FO76 released on Nov 14, PS4 preload with base game + day1 patch started on Nov 11 according to online sources I've found.

delroth··on Revolut staff claim they’ve been told to quit their jobs or be fired
Or rather "we give you the option to work flexibly and be present either mornings or evenings and still get benefits". My employer provides breakfast and dinner in the office, and I don't think I actually know anyone who is present for both (but there is a 60-40 split on breakfast/dinner). I'm sure it happens, but interpreting it as a requirement is just bad faith.
delroth··on Linux Security Hardening and Other Tweaks
2FA that actually carries a payload to verify like Duo will be much harder to exploit, indeed. As to whether making your life harder for routine system administration tasks will actually make stealing your personal data harder, I still have my doubts.
delroth··on Linux Security Hardening and Other Tweaks
If you've got persistent malware on the machine (-> keylogger), 2FA with a hardware key will barely help. The malware can just notice your "sudo" call, pass through your password + 2FA token to its own "sudo" invocation, and make your "sudo" invocation say "invalid password" -- which you'll likely interpret as a typo.

Not that getting root is too interesting for malware running on your personal machines that runs all the interesting code as uid1000. https://xkcd.com/1200/

delroth··on Linux Security Hardening and Other Tweaks
This is cargo-cult hardening without specifying a threat model. Some of the things in there make no sense in most contexts -- for example, logging in as root on a tty to perform administrative actions because you might have an X11 keylogger protects you from pretty much no real attack vector. Your X11 keylogger still runs as your user, with full access to your data, full access to your browser accounts and passwords, etc.

Similarly, many of the L1TF-style mitigations make very little sense on a single-user computer. They make sense on shared tenancy servers / VM hosts where you want to prevent lateral movement, but there is close to no lateral movement to be had on your random Linux laptop. The mitigations that do help are mostly those at the web browser level, since this is for almost all users the only place where they run sandboxed untrusted code.

delroth··on New Google SRE book: Building Secure and Reliable Systems
How do you measure "is it worth" without a good idea of the risks and costs involved in the decision making? Just because it doesn't directly answer your question doesn't mean it's not applicable.

In fact, I'd argue the risk factor is significantly different across startups, so exploring the tradeoffs is the only way to approach the problem generically.

(Disclaimer: I work at Google and was involved with some aspects of the new book.)

delroth··on How SNES emulators got a few pixels from complete perfection
Nintendo's emulators are usually equivalent to what the state of the art was 15-20 years ago. They have plenty of bugs, usually fixed with duct tape and game specific hacks. They are designed to run the set of games they sell and nothing else -- once you get out of what the QA team has explicitly tested, it's not uncommon to find unemulated or badly emulated hardware features.
delroth··on How SNES emulators got a few pixels from complete perfection
Counting the GameCube and the Wii as two consoles is kind of cheating, the Wii is barely more than an overclocked GameCube with an irrelevant OS running on the side exposing some I/O through a fairly well defined I/O interface. The hardest part of emulating a Wii is emulating a GameCube.

Dolphin is also nowhere close to the accuracy of emulators like bsnes. There are many games that plain don't boot, mysterious bugs that have resisted debugging for years and years (some game breaking, e.g. if you have too many characters unlocked you cannot finish Fire Emblem Radiant Dawn on Dolphin), etc.

delroth··on Judge orders Google to turn over a full year of actor’s data
Just because law enforcement asks for it doesn't mean Google has it. They might just ask for it because many companies actually do keep that data, but Google provides in its privacy policy some fairly strong guarantees on data deletion. The documentation at https://policies.google.com/technologies/retention?hl=en is a good read to get an idea of what is kept and for how long.

Disclaimer: I work at Google, some of my work relates to privacy infrastructure, but I don't speak for the company.

delroth··on Google claims copyright on employee side projects
For example, because I don't want people to have to sign a CLA to contribute patches to my weekend side projects? That's incredibly dev hostile.
delroth··on Google claims copyright on employee side projects
> most have not complained the OSS releasing process is heavyweight

Citation needed. I have been at Google for 6 years and I have seen many people either 1. quit in frustration at our OSS policy; 2. stop contributing to OSS projects on their free time because of frustration with our policy; 3. just ignore the OSS policy at the risk of getting fired because the precedent is that most people ignore the policy. Most people I talk with do (3), nobody will admit to it publicly though.

I'm right now waiting for an IARC approval for a small HTML+Typescript that took me 6h to build on a weekend and that I want to MIT-license. This has now been pending for 2 weeks. This is completely ridiculous and it means there is just no reasonable way for someone to follow the policy while doing small side projects on the weekend. Going through the releasing process would have taken roughly the same time, except that Google would also probably not be interested in owning that code for multiple reasons.

delroth··on Better password protections in Chrome
Do you see any other alternative than trust for such a system?

- Building a fully local system will not provide the same coverage and doesn't allow sharing findings across users (which means that when a new phishing page appears, it will appear as "new" for every single user, instead of being new for N users and then known bad for the rest of users). It also heavily reduces what kind of analysis can be done since you can't just store large datasets on every single device and/or run expensive algorithms on every single web page load on a mobile phone.

- Making it open source would not help. You can't know whether the code you can see is what is deployed remotely, so in the end you just end up trusting a different assertion instead (if you can't trust a privacy policy, why could you trust that the deployment is not backdoored?). It also has some significant cons: malware / phishing / abuse detection is in essence a cat-and-mouse game, and secrecy is unfortunately a key requirement in how everyone is building anti-abuse systems across the industry (not necessarily because they want to, but because nobody knows how it could work otherwise).

- You could even go all the way and have e.g. remote attestations, reproducible builds, etc. that allow proving that indeed the code running remotely is the open source code you want and can audit. This is barely doable with available technology these days, and even if someone was to do it there would maybe be 1K people on this planet able to understand why this is trustworthy. A prime example of this is looking at people in this very thread not understanding the differential privacy scheme for detecting compromised passwords.

Not trusting Google is a personal opinion, and I completely respect that. But implying that there is an alternative to trust for this kind of system is IMO misleading. Using DDG or Protonmail or any other service doesn't change the fact that you have to trust someone, it's just a different someone. You might personally believe their word more than Google's word, but if e.g. DDG started logging your identity and log requests and sell that to ad companies you would have very little way of learning about it either.

Disclaimer: I work for Google, not on Chrome, but I have worked on anti-abuse systems in the past.

delroth··on Dex: A research language for array processing in the Haskell/ML family
The N one or the G one?

They're both surprisingly similar to the Nix configuration language: https://nixos.org/nix/manual/#ch-expression-language

delroth··on NixOS 19.09
> I exclusively use the unstable branch on all my Nix/NixOS machines. I rarely do rollbacks. But it's important to note that I'm a developer with commit access, so I can basically "just" fix things if they're problematic. Not everyone has that luxury. (Many Linux distro maintainers likely operate this way, I guess. :)

An important thing with NixOS is that since the whole distro is built from a monorepo, it's also very easy for an advanced user to just git clone and fix the problematic things too (and hopefully send a pull request!).

← PreviousPage 5 of 12Next →