Hacker News: maverwa

New comment by maverwa in "Websites have a new way to spy on visitors: analyzing their SSD activity"

maverwa — Sun, 31 May 2026 18:38:10 +0000

That’s timing the cache, that’s old stuff by know. As I understand, this writes a relatively large file („Gigabytes“) using this OPFS api, which is different from the „localStorage“ api. This seems to use actual filesystem storage on the client, instead of living completely in memory (which may be reasonable given the size of files supported). This allows to actually time SSD IOPS latency by doing random reads.

Collected enough of these samples, together with the information of what else runs on the host, put that in the ML-Blender and the result will be able to tell you, with some accuracy, from a given set of samples, what’s running on the host.

I am sure i misunderstood some things because there are so many caches and unknowns in that setup that I struggle to understand how there could be any correlation, but that’s my understanding so far.

New comment by maverwa in "Safari and Firefox change how big sites render based on the domain"

maverwa — Fri, 15 May 2026 23:29:52 +0000

The reason stated in the article seems sound to me: if it’s broken ins safari/ff but works in chrome, users conclude that the browser is the problem and switch to chrome.

New comment by maverwa in "Gzip decompression in 250 lines of Rust"

maverwa — Fri, 27 Mar 2026 16:20:03 +0000

Where do you see the lack of respect? The author wanted to learn how gzip works and chose to implement it in a language they like to do so. As a learning tool, not because the world needs another gzip decompressor.

New comment by maverwa in "GitHub appears to be struggling with measly three nines availability"

maverwa — Mon, 23 Mar 2026 16:47:39 +0000

True! Technically even 9.99% would be three nines!

New comment by maverwa in "Fyn: An uv fork with new features, bug fixes, stripped telemetry"

maverwa — Mon, 23 Mar 2026 13:50:18 +0000

If I understand the description of this „telemetry“ in fyns „MANIFESTO.md“ correctly, it does not make outbound connections you did not asked for. It sets the user agent http header to something that identifies your OS, CPU, python version and if you are running in Ci when communicating to the package registry. It does not send any of that to astral, not ist any of that highly personal.

Sure, it should not be there by default, especially OS & CPU imho. But it’s not really what I’d call „invasive telemetry“.

New comment by maverwa in "JPEG Compression"

maverwa — Wed, 18 Mar 2026 10:38:07 +0000

With CRTs I would think that the problem may be that they do not see a full picture at all. Because the full screen is never lit all at once? Don’t know how persistence of vision works in this case…

New comment by maverwa in "Separating the Wayland compositor and window manager"

maverwa — Mon, 16 Mar 2026 15:42:31 +0000

Except if you need WSLg, because then you can add RDP issue to all your Wayland issues, and, not infrequently, also xwayland issues! You can have all the fun!

New comment by maverwa in "Show HN: Renovate – The Kubernetes-Native Way"

maverwa — Wed, 11 Feb 2026 15:55:30 +0000

Mend is not a competitor, renovate is the software, mend is the company.

They are tools that automatically check your repo for dependencies and create PRs when there are updates. It supports a wide range of package managers and other places dependencies may be specified.

Dependabot is another solution which is more „GitHub-native“ maybe.

New comment by maverwa in "Kidnapped by Deutsche Bahn"

maverwa — Mon, 29 Dec 2025 13:20:39 +0000

Maybe worth mentioning that this „1.50€ compensation“ rule only matters if you use the „Deutschland-Ticket“ which is a fixed price ticket for a whole month, unlimited travel on the Regio lines (short distance trains, i.e. non IC)

If you bought a regular route ticket you get 25% and more than an hour delay, and 50% at more than two hours. Not sure how it is with other multi-use tickets.

This, combined with the certain delays CAN make traveling by train quite affordable… /s

https://www.bahn.de/service/informationen-buchung/fahrgastre...

New comment by maverwa in "(Social) media manipulation in one image"

maverwa — Mon, 22 Dec 2025 15:02:33 +0000

The difference being that while people have a good intuition on one may spend their vacation and what one would talk about, it’s not so easy to extrapolate for other things. Sure, everybody knows that media talks more about homicides and terrorism than the thousands of people dying or „old age“-kind of things, but it’s not always easy to keep in mind on how far apart these numbers are.

I think „manipulation“ may be to strong of a word here, since it assumes intent to manipulate, which is not necessarily to focus on „out of the ordinary“ events. But I think nonetheless that this infographic is interesting and important, because it reminds us that these biases exist and how big they are.

New comment by maverwa in "Tell HN: HN was down"

maverwa — Wed, 17 Dec 2025 16:56:25 +0000

Thank you for HACK. I love it!

New comment by maverwa in "Same-day upstream Linux support for Snapdragon 8 Elite Gen 5"

maverwa — Fri, 28 Nov 2025 15:25:52 +0000

if you are in the business of buying parts, assembling them, and selling the assembly under your brand (as tuxedo and others like them are), then Apple is a indeed not an option. Their chips might be the best, but you still can only get them in Apple devices, in the specs Apple provides and with no official support for any OS beside macOS / iOS.

New comment by maverwa in "Show HN: OCR Arena – A playground for OCR models"

maverwa — Tue, 25 Nov 2025 12:11:06 +0000

from my first tests it does fine with german, at least for the gastly "handwritten" font the restaurant menu I used for the test uses.

New comment by maverwa in "Thoughts on the Word Spec in Rust"

maverwa — Thu, 09 Oct 2025 08:22:23 +0000

or - if feasible - extend the existing crate instead of creating a new one.

New comment by maverwa in "TigerBeetle is a most interesting database"

maverwa — Wed, 01 Oct 2025 13:06:03 +0000

its easy as in "simple to implement and execute" but not cheap, because it may require scanning large amounts of memory. You have to visit every list entry.

Whats trivial for a very small list, may be a no-go for gigabyte-sized lists.

New comment by maverwa in "My game's server is blocked in Spain whenever there's a football match on"

maverwa — Thu, 25 Sep 2025 09:37:22 +0000

I am somewhat surprised (and happy) germany is not in on it already. There are already (afaik voluntary) block lists ISPs are applying, but I am surprised its not breaking things on a similar scale to italy and spain, an that the DFB isn't already piping /dev/urandom-domains into it...

New comment by maverwa in "I'm spoiled by Apple Silicon but still love Framework"

maverwa — Tue, 23 Sep 2025 11:04:15 +0000

To add some context: man kernel_lockdown[1] reads "Unencrypted hibernation/suspend to swap are disallowed as the kernel image is saved to a medium that can then be accessed.". And to my understanding there is currently no way to tell a (mainline) kernel that allows "encrypted hibernate", i.e. no way to tell the kernel that its hibernation disk is "secure".

So its not a direct "linux prevents hibernate on secure boot", its "linux recommends kernel_lockdown when secure booting", "kernel_lockdown prevents hibernate with unencrypted swap" and "theres no well to make the kernel believe the hibernation disk is encrypted", but the result is the same.

You can "just" run secure boot without lockdown. Its a cmdline, you can just omit it. You can run custom patch sets that add cmdline options so the kernel allows hibernation in lockdown (if you pinky-promise the swap is encrypted).

But neither of these are easily accessible to the average user.

1: https://manpages.debian.org/bullseye/manpages/kernel_lockdow...

New comment by maverwa in "I'm spoiled by Apple Silicon but still love Framework"

maverwa — Mon, 22 Sep 2025 13:33:50 +0000

Thanks, I did not knew that. My understanding was that keeping the memory alive for suspend-to-idle was the main issue here. But that also might be something a vertically integrated Apple Silicon can win vs. that x86 madness there every day.

And to be sure, I do not claim that there is nothing to gain in s2idle. I bet theres still a lot of headroom to safe energy. Its just that it would be easy to safe a lot of power if s2disk "just worked".

New comment by maverwa in "I'm spoiled by Apple Silicon but still love Framework"

maverwa — Mon, 22 Sep 2025 13:29:00 +0000

Some of the issues have been addressed. For example, iirc, there was a bug where pulling out the power plug while the lid was closed would trigger the device to wake up.

Some other issues remain. Largest I am aware of is independent from the hardware, but an issue with suspend-to-disk & kernel lockdown, which prevents deep sleep.

New comment by maverwa in "I'm spoiled by Apple Silicon but still love Framework"

maverwa — Mon, 22 Sep 2025 13:21:37 +0000

The complaint about power usage in suspend is especially sad because it’s pretty much a common problem for Linux on laptops. Not sure if that’s what applies here, but the numbers about match what I see with my Framework. Basically: if you want to use secure boot you usually also want kernel lockdown mode, and you cannot hibernate a lockdowned kernel. At least not without out-of-tree patches.

IMHO that’s a giant issue. If you can’t hibernate (aka suspend to disk) you will never be able to get that power consumption low. And telling people to not run secure boot or lockdown is not really a good answer either. Especially since the default installer already sets those things up. I get that „Linux on laptops“ is not a priority big enough to get a proper fix for that. And that it’s not an easy issue to fix. But the current state is really really sad.