Hacker Newsnew | past | comments | ask | show | jobs | submit | azalemeth's commentslogin

I know that the interim releases had issues with zfs and trying to update gave the message "Sorry, cannot upgrade this system to 25.04 right now System freezes have been observed on upgrades to 25.04 with ZFS enabled. Please see https://wiki.ubuntu.com/PluckyPuffin/ReleaseNotes for more information. "

The release notes don't seem to mention zfs. I hope these issues have been fixed?


Just FYI - That defect only impacted systems which boot from ZFS - mine did not.

The installer would stop an upgrade if it found a mounted volume - it apparently checks for zfs volumes as follows: sudo zpool list -H

I could get around the installer's ZFS check by unmounting the drive: sudo zpool export poolname

I upgraded.

Then remounted the volume: sudo zpool import poolname

Then I was up and running...


I really like hexyl [1], which does this by default.

https://github.com/sharkdp/hexyl


The author uses hexyl as an example of trying, but not doing it right.

One thing I noticed is that github's "clone" drop down menu used to include instructions on how to do so via git git (i.e. `git clone /path/to/repo`). Now there are only instructions provided by default for `gh`, their client.

I can't help but guess if these are related.


I do often wonder about stories like this in the context of forensic science – my (incomplete!) understanding a lot of the time suspect DNA samples are taken from small areas and amplified significantly with high-cycle count PCR. I'd worry that any jury presented with a statistical argument about a fragment of somebody's DNA being very unlikely ("1 in 100 million") to be different to the sample found at the scene would not be aware of all of the potential systematic reasons why the actual true probability may be much, much higher.

Probability seems to be one of those things humans habitually mess-up at.

"The chances of this person's unique DNA showing up at the scene are a zillion to one!"

"What does that really mean when the sample also contains unique DNA for a hundred other people? Did all of them commit the crime as a group?"


Depends on how they're using it. To find an unknown person and prove they were at a scene - yeah you'll have the 100 person's worth of DNA to sort through and then match against a (presently) incomplete DNA database. But if you already have a suspect and need to place them at the scene, if their DNA is one of the 100 then they have shown that.

But we’re they at the scene or did they just bump into someone or something that was there?

That's something that would have to be addressed at the trial by the defense attorney raising challenges.

If the DNA is present, it's present - barring any procedural mistakes by the forensics technicians (mislabeled sample, dirty lab equipment, didn't follow manufacturers instructions, etc). Or deceit by one or more members of the forensics team to implicate the suspect.


>That's something that would have to be addressed at the trial by the defense attorney raising challenges.

This is the wrong point at which to correct the problem. When prosecutors are allowed to introduce "science" as their evidence, jurors give this way too much deference. It's the prosecutor basically saying "you're too stupid to understand it, so indulge my appeal to authority" and jurors tend to happily acquiesce. This is why judges are supposed to gatekeep expert testimony and not just let any quack step in and make outrageous claims... and despite their attempts, quacks have repeatedly done just that these past few decades.

If is true that the presence of DNA means essentially nothing, then it shouldn't be permitted to be introduced at trial. Protection from wrongful conviction shouldn't rest on someone having an attorney expensive enough to be able to unsway too-easily-swayed jurors with counter-specious arguments.


Except DNA may not be present: the probabilities involved with partial DNA matches become very problematic.

Forensics never relies on a single piece of evidence. It takes several corroborating accounts and pieces of evidence to reach "beyond reasonable doubt."

This is why even breathalyzers have to be done twice, with a gap in between (and preferably with different sensors). It's also why several witnesses are examined. It's why fingerprints and lie detectors alone don't pass the muster. Nearly anything can be faked, or misinterpreted. All these things have to be used together to create an unbreakable story.

So there is next to no risk with DNA in the air.

As an example, there recently was a guy who went into a police station and claimed that he was relocating there from another province. The officer on duty was suspicious, and found there was no record of this officer in the entire country, and arrested him. In court, the guy got off the hook, even after impersonating a police officer and trying to infiltrate a station, because the officer on duty did not collect the appropriate evidence. The camera footage looked entirely normal. It became entirely hearsay.

You can pretty much expect him to charge that officer with wrongful arrest now.

We may never know what that guy was up to. But he certainly had chutzpah!


> Forensics never relies on a single piece of evidence

An uncharitable characterization is that forensics relies on many pieces of _stuff_ that cops and attorneys can reasonably convince a judge meets the rules of evidence and a jury that it is in fact evidence, while hoping the defense can't afford an expert (mind you, almost never a practitioner, but a professional legal consultant) to convince the jury it isn't evidence.

Most forensic analysis is complete bullshit, and it takes decades to convince judges to forbid the junk science.

So I wouldn't say there's "no risk." There's tons of risk.

What I would add is that you almost never have to worry about forensics if you're committing crimes, because the forensics are only going to be used to prove your guilt in court should you choose to fight it. If you're not committing crimes and become the focus of an investigation, you should be terrified of forensics.


This isn't unlike "the right to remain silent." It's weird that people don't fully understand that to mean stfu for your own good, 'cos the only notes they are writing are toward building their case against you.

Forensics isn't the way you characterized. There are forensic teams for both sides of every coin that needs forensics. It's literally a battle of what side can make the most compelling argument, and nothing more or less.

The most compelling argument is the one that has lots (the term "a preponderance" gets used in law where I am) of corroborating evidence through tools/techniques/people that are presumably not sharing and comparing notes or data with each other.


> it's literally a battle of what side can make the most compelling argument

So it's exactly as I characterized it, a racket. It's not science.

Almost all forensic tests are bullshit, and forensic experts simply cannot determine what they say they are able to under any scrutiny.


Samples are normally things like your semen being found in the dead body or your blood on the broken window, or the victim's blood on your shoes.

Air contamination is very unlikely. You aren't going to get air contamination inside the victim's bedroom... And if somehow a whiff of your air travelled a hundred miles into their bedroom and all over their corpse, other evidence would be used to rule you out.


Human DNA is relatively large. What's floating in the air is not going to be intact or complete.

Testing relies on amplifying many parts of the human DNA not just a small fragment.

The smaller the fragment the less precise the comparison.

So no, floating DNA does not meaningfully impact testing. It can impact PCR assays that target small regions for identification eg of viruses.


Very ironically, they seem to have upped their game. Trying to read TFA on an older version of firefox gives me the lovely message:

Failed to verify your browser Code 11 Vercel Security Checkpoint, arn1::1776759703-rtDgRAtRyXvjD4IoU4RbqvkGmvQQCP7H

Gah.


This is excellent news. Now make them have user-unlockable and user-relockable bootloaders...

You wish.

I just viscerally do not want this and will frankly refuse to do business with anyone who requires it.

I just want this whole idea to kindly please bog off. We shouldn't be further creating the apparatus of the surveillance state.

Yeah I don’t like how the discussion is shifting to implementation details, instead of debating whether any of this is good or necessary

IMO the implementation is crucial. If everything is locally on the device and I can confirm digitally that I'm older than 18 BUT NOTHING ELSE is leaked, like the German eID supports (I think).

Why/how would this be a bad thing?


Because that's debating the mechanics of it rather than the need or the ethics. Nobody gives a shit about the mechanics because if there's a debate about the mechanics that means the discussions about the need and the ethics have already happened. Yet those discussions are in fact still going with few in favour of the ethics or need for these systems. The mechanics are the final step after everything else has been settled.

Implementing this is fascist.

I want corrupt politicians to bog off and people to think long term. I guess we’re both going to be very disappointed.

My understanding is that this is much more privacy friendly than showing your id

Makes no difference in the fundamental dislike i have for the concept

Do you also dislike the concept of requiring to be a certain age to say enter a strip club or a sex club?

If not, what is the difference between those controls and having to be a certain age to enter porn sites?

Genuinely curious. To me, the primary objection to the online controls has been the implementations. The EU implementation will be[1] even better than the strip club, where the bouncer sees your ID and can remember it, when they move to zero-knowledge proofs.

[1]: https://digital-strategy.ec.europa.eu/en/news/commission-rel...


> Do you also dislike the concept of requiring to be a certain age to say enter a strip club or a sex club?

You can't compare someone checking your document before entering a strip club (or even a pub, or asking for alcoholic drinks) vs a computer system getting and logging an attestation and verifying it against a government database (or third party) where the Govt knows who the credential belongs to, and who's checking it. Along with my government "compelling me" to run software (even if it's open source itself) that requires me to have a binding contract with a foreign third-party company known for privacy violations, and running their proprietary software stack on my device for said government software to work, so I can participate in most parts of the digital society.

Of course the existing "identity verification" done by scanning yourself and your ID document (passport, national ID or driving licence) is not acceptable, unless counted exceptions where said documentation is needed (banking and others, because of KYC/AML)


First of all yes.

Secondly, it's the dumbest comparison anyone could possibly make.

The difference with a porn website is as follows:

- the age check on porn sites are notoriously dumb and useless, it's literally a meme. It was a meme before there were memes.

- I choose to go on porn sites. It's not exactly a requirement that I get access to a porn site. Access to my OS on my device to work, have fun or do whatever I want privately is kind of a lot more necessary.

"Zero knowledge proof". Yeah OK. I've got a few dozen bridges to sell you. Interested?


> First of all yes.

That at least explains a fair bit.

> the age check on porn sites are notoriously dumb and useless

That they have been useless is hardly an argument that's relevant to the current discussion.

> I choose to go on porn sites. It's not exactly a requirement that I get access to a porn site.

Indeed.

> Access to my OS on my device to work, have fun or do whatever I want privately is kind of a lot more necessary.

Sure, but that is an entirely different discussion.

Anyway, if you had some actual substance to your flippant dismissal it could perhaps lead to some interesting discussion.


The author is 21 (which I find incredibly impressive) and is using a DE that was written when they were a baby.

It _is_ lightweight in that context. I also love the fact that XaoS knowledge is useful in the context of "real software" programming!


As someone who remembers E making the rounds among the BSD and Linux users in my college dorm when it first came out, there's no way he's only 21 if he was a baby in the late '90s.

From her "Short CV"

"Hello! I’m Kamila Szewczyk (iczelia). I am 21 years old. I’m an expert programmer and aspiring mathematician, primarily interested in compiler construction, data compression, esoteric languages, statistics and numerical algorithms. ... Currently I am a full-time student based in Germany." [1]

And the start of the post:

"The editor in chief of this blog was born in 2004. She uses the 1997 window manager, Enlightenment E16, daily."

[1] https://iczelia.net/cv/

edit: added the [1] at the end of the first quote


The author's name is Kamilla. She was born in 2004 (according to the article).

I literally just came to HN to ask if I was alone with the acurséd "API Error: 500 {"type":"error","error":{"type":"api_error","message":"Internal server error"},"request_id":"…"}" greeting me and telling me to get back to using my brain!

500-series errors are server-side, 400 series are client side.

A 500 error is almost never "just you".

( 404 is a client error, because it's the client requesting a file that does not exist, a problem with the client, not the server, who is _obviously_ blameless in the file not existing. )


> A 500 error is almost never "just you".

I know you added the defensive "almost" but if I had a dollar each time I saw a 500 due to the session cookies being sent by the client that made the backend explode - for whatever root cause - well, I would have a fatter wallet.


Depending on what you mean by "made the backend explode", that is a server error, so 500 is correct!

Bad input should be a 4xx, but if the server can't cope with it, that's still a 5xx.


Indeed, and also there's a special circle of hell reserved for anyone who dares change the interface on a public API, and forgets about client caching leading to invalid requests but only for one or two confused users in particular.

Bonus points if due to the way that invalid requests are rejected, they are filtered out as invalid traffic and don't even show up as a spike in the application error logs.


I know that in principle this is true. However, I have seen claude shadow-throttle my ipv4 address (I am behind CGNAT), in line with their "VPN" policy -- so I do not trust it, frankly.

> in line with their "VPN" policy

This is how I learn that they have a "VPN" policy. Thinking of it maybe it makes sense, that is if it's what I think it is, but seems scummy nonetheless.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: