HNHacker News
TopNewBestAskShowJobs

lmm

47,766 karma · joined October 14, 2011

I'm known as m50d some places, including github; my blog (such as it is) is at https://m50d.github.io/
submissionscomments
lmm··on Dumb ways for an open source project to die
Until you run out of hardware that can run an OS old enough that the old version of the framework works, sure. Even then, you may find it hard to add new dependencies without getting current on the framework, since most ecosystems require you to pick a winning version in the case of a "diamond" transitive dependency.
lmm··on Disney erased FiveThirtyEight
Yes and no. It's a two-party system and a lot of people vote for candidate X to stop candidate Y getting in. Indeed IIRC a lot of democrat messaging around the previous election was explicitly about stopping Trump rather than the merits of whoever their guy was.
lmm··on Disney erased FiveThirtyEight
They had him at 30% at a time when most reputable media had him in single digits at best. You might not remember now, but the result of that election was a real shock.
lmm··on Disney erased FiveThirtyEight
> a good analyst can't predict the future in politics, sports, or anything else. What they can do is make good probabilistic estimates of what is likely to happen. 538 wasn't pretending to do anything more than that.

Well, sure, but how big is the market for that, really? Particularly for a binary outcome like an election, knowing who's going to win is fun, reading a pundit telling you who's going to win can be fun, but ultimately the man in the street is going to take whatever the pundit said and reduce it to candidate X or candidate Y, and you can only do so much better than replacement level at that.

lmm··on Disney erased FiveThirtyEight
Conversely weather forecasters report a 40% chance of rain when the actual chance is 10% or similar.

So I have a bit of sympathy for people who don't have a good intuition for probabilities, given that the world is constantly gaslighting them.

lmm··on Tesla's lithium refinery discharges 231,000 gallons of polluted wastewater a day
In some vague abstract sense? Wouldn't protesting specific instances where bad laws are letting companies get away with environmentally damaging practices be more effective? As and when there's a pattern of events that is best addressed by adjusting the law, politicians and voters can figure that out for themselves.
lmm··on Elon Musk has lost his lawsuit against Sam Altman and OpenAI
> Frankly, bullshit. It worked reliably, extremely so. It had remarkably few bugs too. It was also actually doing way better financially.

It was still notoriously prone to failure, and it had been a financial disaster area until the Trump phenomenon suddenly boosted it.

lmm··on Elon Musk has lost his lawsuit against Sam Altman and OpenAI
You raise all your defenses in the trial, you only get one. If they'd wanted to put all their eggs on the statute of limitations point then they could, but you can understand why defense lawyers generally don't do that.
lmm··on We stopped AI bot spam in our GitHub repo using Git's –author flag
Some say the greatest compliment a mathematician can be paid is when a concept that bears their name is so ubiquitous that we no longer capitalise it. (E.g. Abel with the abelian group).
lmm··on I don't think AI will make your processes go faster
> Almost 2 decades in IT and I absolutely do not believe this can ever happen. And if it does, it’s so rare, it’s not even worth talking about it.

It happens completely routinely, IME. Just compare how much effort it takes to clone a system that's already written versus making that system from scratch.

lmm··on AI is a technology not a product
> Are people in the habit of asking their admin to order a pizza or a Uber?

They're in the habit of their admin telling them the Uber has arrived to take them to their 3 o'clock yes.

lmm··on I don't think AI will make your processes go faster
> Their compiler fails to compile (well, at least link) some C programs altogether, and in other cases it produces code that is 150,000x slower than a real C compiler with optimisations turned off

I would bet that those things are also true of at least one expensive commercial C compiler.

lmm··on Mercurial, 20 years and counting: how are we still alive and kicking? [video]
> I have the feeling that Git winning the war hinges heavily on GitHub being the way to do open source projects

Nah. At the time BitBucket was the better way to do open source projects, and they were Mercurial-first. But eventually they had to add Git support because there was so much demand.

lmm··on Access to frontier AI will soon be limited by economic and security constraints
In theory yes. They've got a bargaining chip with TSMC. But it's unclear how much use that would be without a safe shipping route between Europe and Taiwan and/or a navy capable of maintaining such.
lmm··on Access to frontier AI will soon be limited by economic and security constraints
Happy to pay for? Or happy to spend other people's money on?
lmm··on Claude for Legal
> How's this any different than any professional license? You're basically paying for preferential treatment from the state in a given subject area.

Because it's got nothing to do with the professional part? Licensing should affect their practice of law, sure, but it shouldn't grant random other privileges.

lmm··on Mullvad exit IPs are surprisingly identifying
> I imagine there are a bunch of things on the internet that break if you start trying to connect to them from varying IP addresses. Things like the various CAPTCHA schemes and rate limiting etc, IP reputation etc.

Given how much of the world is stuck behind CGNAT now, I would expect any major sites to handle it.

lmm··on Have a Coherent AI Policy
Is that wrong? My assembly programming sucks, but I can do it (slowly) if I need to. I'd expect most serious developers to have that level of knowledge.
lmm··on Have a Coherent AI Policy
> using the Social Media "vibe-coding" as a strawman risks anchoring against something that's a mirage.

It's not a strawman. A lot of people are doing it.

lmm··on The Power of a Free Popsicle (2018)
> Now, I worked for a company who last year cut all the parties, food, drinks, team events, 3 year HW refresh cycle, even the color printers, to ensure we'll still get to keep above average salaries through the tremulous times our industry is going through. Absolute respect. I'd rather have more money to pay the ever increasing bills, than pizzas and a 50 Euro softshell jacket.

It's amazing how you can do without the necessities of life, provided you have the little luxuries. Yes, money is always good. But a company that isn't willing to spend a small amount of money on "frivolous" things that have a disproportionate impact on the daily experience of working for them is either an ineffective company, or one that genuinely doesn't value its employees.

> I wonder if it's possible to tell this story on how dehumanizing it felt to not get free beer with a half million dollar salary, to an average laborer, with a straight face, and expecting any reciprocating "working class" empathy.

Oh look, now the mask comes off.

lmm··on First public macOS kernel memory corruption exploit on Apple M5
The dangers of technical debt and the importance of mitigating it have been known for a long time. Unfortunately a lot of entities now ignore all experience and best practices as soon as you say the "AI" buzzword.
lmm··on Do teachers need advanced degrees?
They're already paid better than adjunct professors or grad students which is the normal career path for people with advanced degrees.
lmm··on AI is making me dumb
> you would be stupider every time you go on vacation

Are you sure people aren't?

lmm··on Int a = 5; a = a++ + ++a; a =? (2011)
They don't. In the culture some kinds of undefined behaviour are taken seriously and some aren't. If you want to write code that "works", you emulate what popular performance benchmarks do (whether their code is undefined according to the standard or not), since those are the thing that C compiler developers actually care about.
lmm··on Int a = 5; a = a++ + ++a; a =? (2011)
> OK. What is the value of a spec to which compliance is impossible?

It lets you tell people you have a spec? It makes it easy for compiler developers to dismiss bug reports with "your code violated the spec"?

lmm··on What the Hell Was Going on with Cigarette Ads in the 70s? (2024)
"Error establishing a database connection", apparently? Groovy.
lmm··on Twin brothers wipe 96 government databases minutes after being fired
Yes, but the company may well be judgement-proof at that point.
lmm··on The US is winning the AI race where it matters most: commercialization
> To align the interests there has to be a substantial negative consequence that would be felt by the grifters if the endeavor fails.

Like what though? The failure in Iran has had pretty substantial consequences that are being felt. If that's not good enough, what is? You were talking like you thought there was a realistic path to a better military, but consequences for the US aren't going to come much bigger than this.

lmm··on The US is winning the AI race where it matters most: commercialization
> The Iran war doesn't count as the alignment of interest requires an actual threat of being defeated.

That's starting to sound a bit no-true-scotsman. If we need an existential threat to the US, that's not going to happen - realistically China conquering Taiwan or even building an empire around the Pacific would still not be felt as such a threat.

lmm··on The US is winning the AI race where it matters most: commercialization
> The structure of the US makes it basically the single most secure democracy anywhere right now or in history. No country in Europe or Europe as a whole is even competitive by comparison.

How do you figure? I hear you have roving gangs of masked thugs beating up random citizens with the backing of your government, that doesn't sound very democratically secure, especially with what healthcare costs over there.

← PreviousPage 8 of 34Next →