foobiekr

You should perhaps consider that most people would rather die than be tortured to death and perhaps we feel the same way about animals even if we eat meat - especially primates.

Now you're killing other animals with the collateral damage.

If you must use traps, use snap traps and take the responsibility for cleanup.

Rat-X isn't a poison and mostly doesn't work.

You forgot (d) which is that the people involved are often terrible people.

The issue of people on mind-altering substances being involved with "facilitators"/therapists who use that power for their own ends showed up even in the recent MAPS trial with alleged professionals.

LSD and other psychedelics like mushrooms of mescaline are best used with someone who actually cares about you in a setting you control for self-exploration and absolutely devoid of the kind of people who become psychedlic-assisted underground therapists.

I find that it is an effective rat repellent - a neighbor has a rat colony they will not address - but while it was effective for squirrels at first, they seem to have gotten over it, and we now see them eating dropped seeds without any pause at all. I think the first generation never overcame it but now they do eat whatever the birds spill.

It makes it impossible to write a general purpose dissector that takes captured messages or bytes and figure out how to parse it.

All they needed was a varint at the head of any marshaled from to at least provide some scoping clue.

It is _very_ unfortunate there is no PDF version.

they are just bullshit machines. bullshitters can cite wikipedia and are still bullshitters who are bullshitting.

Semantic versioning is an illusion. It's a human-managed attempt to convey things about the surfaces and behaviors of software systems. Best case, it isn't completely misleading and a waster of everyone's time.

There is no perfection here, but the correct way to reason about this is to have schema-based systems where the surfaces and state machines are in high level representations and changes can be analyzed automatically without some human bumping the numbers.

Figma absolutely sucks.

It is slow, but more importantly, by encouraging UX designers to work in high fidelity designs where live changes are frustrating and feedback is entirely incorporated as a bunch of mostly-hidden comments in a sandbox that makes it hard for reviewers to follow up on or reference back to, all it's done is made everything worse.

There's almost no HCI or interaction design in what passes for UX today. They are mostly - probably 90+% of them - visual designers who have memorized some canned responses to use when asked why they aren't doing basic things.

Seriously, anyone who things software as a discipline is bad, well, they're right, but they have no appreciation for how appalling "UX" is today vs even 1990.

sshfs doesn't work for things where you're developing for linux and working on macos - run the code, run the debugger, step, etc.

I don't use IDEs at all and work on sshfs almost exclusively, but I totally understand where people are coming from and the difference between editing files and running code.

I don’t think I’ve ever seen an llm in any context ask for clarification. Is that a real thing?

Very much so. Pretty much all of these protocols are simplifications of asn1 and in some cases (like protobuf) there are a handful of things that got lost because the wire formats didn’t have them as they didn’t need them. A schema indicator being the single biggest flaw in protobuf.

Have you tried any of the specialty services like Augment? I am curious if they are any better or just snake oil.

The connection between canker sores and SLS is pretty well known and not rare or caused by some other allergy.

More like three years. Even in the best case the retained value curve of GPUs is absolutely terrible. Most of these huge investments in GPUs are going to be massive losses.

agreed. making the world worse.

At least on Intel, it was quite a significant difference in some ways, in kernel terms.

Honestly whatever the hell Google offers at this point has been disguising itself as google search for years. It sure as shit is not what people expect from google.

Are you familiar with Centralia?

For the many reasons I listed. Pre sharesd keys are almost always global and you can’t do forensics to find the leak.

I am very familiar with reporting bugs on the Linux kernel, as you may imagine from the previous comments.

But you’re just reiterating that the kernel development itself lacks sufficient testing which necessitates this. Everyone is a volunteer; the result is as you’d expect.

Look, there’s nothing wrong with this. In reality all the people who are electing to run Linux without it actually having sufficient quality for their purposes are the actual problem, who’d which there are hundreds of companies that just don’t care. We created this situation by undervaluing quality vs cost.

But it’s also the reason it won’t be a success on the desktop. The only way that happens is if windows takes an even more substantial nosedive. Things are going to have to get pretty bad for that to happen.

I think you’re missing the point. Unless you were actually measuring the content of your file system with a reasonable quality hash function, you likely had actual data corruption that Fsck will never see. Fsck can’t check data on ext*.

Unless you were routinely running mtree or something similar, you’d honestly have no idea.

My point here is that software can appear to be fine when it’s actually shit because most people won’t notice any kind of corner case. All you’re saying is that you’ve very likely missed the corner cases.

Pki here does not mean a global CA. You can run your own CAs (and should).

You don’t need to bring up quantum computers. Almost all protocols in the networking industry are basically running with a shared secret that is service global. Pop any box at all and you have the world for any traffic you can capture.

The problem with the shared secret model isn’t that it can be stolen, it’s that it is globally shared within a provider network. You can’t root it in a hardware device. You can’t do forensics to see from what node it was stolen.

We are talking about an industry where they still connect console servers, often to serial terminal aggregators that are on the internal network alongside the management Ethernet ports, which have dumb guessable passwords, often the same one on every box, that all their bottom tier overseas contractors know.

It’s just sad.

More likely they got access and then snooped any of the many insecure protocols used to manage network devices.

Anyone who has ever worked in networking will understand what I mean.

The networking industry is comically bad. They use ssh but never ever verify host keys, use agent forwarding, use protocols like RADIUS or SNMP which are completely insecure once you pop a single box and use the almost always global shared secret. Likewise the other protocols.

Do they use secure boot in a meaningful way? So they verify the file system? I have news for you if you think yes.

It’s kind of a joke how bad the situation is.

Twenty years ago someone discovered you could inject forged tcp resets to blow up BGP connections. What did the network industry do? Did they institute BGP over TLS? They did not. Instead they added TCP MD5 hashing (rfc: https://datatracker.ietf.org/doc/html/rfc2385 in 1999) using a shared secret because no one in networking could dream of using PKI. Still true today. If deployed at all, which it usually isn’t. 2010!!

If you want to understand the networking industry consider only this: instead of acknowledging how dumb the situation is and just using tls, instead we got this - https://datatracker.ietf.org/doc/html/rfc5925 - which is almost as dumb as 2385 and just as bad in actual deployment because they just keep using the same deployment model (the shared tuple). Not all vendors that “support” 5925 support the whole RFC.

As an aside this situation is well known. People have talked about it for literal decades. The vendors have shown little to no interest in making security better except point fixes for the kind of dumb shit they get caught on. Very few security researchers look at networking gear or only look at low end junk that doesn’t really matter.

Yes, they are very careful about not breaking userspace. That has nothing to do with holding the _user_ in high regard, though.

The sweeping changes in Linux are amazing, but they also break a lot, frequently, and the testing approach leaves a lot to be desired. I am old and a former embedded coder, I understand how the 90s developers feel about unit testing, and I am sympathetic to the idea that test-driven-development and heavy UT in modern teams is often completely out of control, but Linux is distinctly lacking in real testing of independent units in isolation. I am aware of kunit, kselftest, etc.

Your experience is selection bias. And if you never had issues with ext3fs, I assure you it's because you never checked.

No coder should ever, ever be shocked that broken code mostly kind of seems to work, especially if you don't notice things like file corruption because you're not looking, or memory corruption that doesn't happen to matter at the moment. A lot of the failures and bad code are in the unexpected path.