philovivero

[flagged]

SRE.

"You ship it, you take the pager. Once it's stable, then SRE org will take the feature. If it gets unstable again, SRE will hand it back."

If someone vibe codes something, and it works, then no reason not to merge it. So just set it up so if it doesn't work, they're on the line to fix it.

Along with their oh-so-supportive manager.

But also, if you have the clout, doing what you're doing nips the problem in the bud earlier, and so is more efficient. Good that you have the clout.

Weird.

I had LLM (Claude) work with OTF to generate an entire infrastructure HCL (from existing). It built a very nice project that seemed idiomatic from my experience.

Then used it over the course of several hours to refactor it to take variables/inputs for everything, then over a few days got it to a state where it would create entire new environments "equivalent" to the original environment. Days because you know... it's TF in AWS which is slow, so the round-trips were probably 90% of the wall-clock time here.

I'm not a hardcore veteran Infra eng, but I'm decent, and I was able to do way more with LLMs than if I'd had to do it myself.

Yeah. HN is a bubble. Hollywood has an axe to grind, and it's not a good one, but HN ideology is in-line with Hollywood ideology.

I don't know about theatres, but I do know about hotel rooms.

If you lower the price too much, you get a different sort of clientele. The sort of person who wrecks the place and annoys all the other patrons nearby.

Then the cleanup costs a lot. Often more than the amount of revenue collected on the room.

It absolutely makes more sense to keep the hotel room empty than to lower the price to keep it fully occupied.

Who grey-texted this comment? So confused who could disagree with it. Is that the problem, this comment is so obviously true, it's just redundant?

I'm assuming this is a joke of some sort?

But correlation is correlation. At least, for most people's definitions of the word.

Running a model isn't binary, it's per amount of time spent generating tokens.

First, I am also frustrated by companies trying to prevent unauthorised used.

But second, the reasons are:

(1) For AI company, someone publishing: "I asked the model a question about crime, and it talked shit about black people! Look! [damning quote that you can also get model to say/do]." Stability took the "let people do what they will" tack and now Forbes and every other major media mouthpiece slams them at every opportunity about how they are ethically-challenged.

(2) For Replika, someone chatting with their online girlfriend: "I love you more than my wife and children." Then someone hacking Replika exposing these conversations, and now Replika is in hot water because all these divorces. Replace example with 100 other similarly awful situations like talking about mental health problems, crimes, petty squabbles with their coworkers, or political problems.

Start searching SuperHOT and RoPE together. 8k-32k context length on regular old Llama models that were originally intended to only have 2k context lengths.

> While many software problems can cause you to lose money, engineering problems can cause you to lose time.

I'm guessing you were trying to say something else here. I literally cannot think of a single software engineering problem I've ever encountered that didn't cost time. By your definition, then, software engineering is engineering. Your claim and your definitions are at odds with one-another.

Also, you don't directly claim it, but you seem to imply, that software engineering can't have real-world consequences... or something? As another reply points out, sometimes software is in the critical path for things like rockets and airplanes, where mistakes cost lives.

And some people making software for less life-altering systems take their craft just as seriously. Some people think that losing $10M every single second while their software is failing is a big deal.

Are you claiming people who write HFT code, ad arbitrage code, code that powers the front page of Apple, Amazon, Microsoft, and Google are just cowboying it through the day, doing nothing special?

Overall I just find this comment very confused. Maybe you could put some thought into what you're trying to say, and say it better?

Someone else posted this down in the bottom of a comment thread somewhere, but I think it deserves to be at the top level: https://www.youtube.com/watch?v=5Dw3aKbw5Wo

This is a smart, motivated audio engineer talking about Atmos. Quick summary: it's awful when it works perfectly, and it almost never works perfectly, in fact, it rarely works at all. Good luck trying to get Atmos working, and even if this was a completely open standard, would you really bother?

> Both of those are systems which had to work right. Large language models are not even close to being safe to use in such applications. Until LLMs report "don't know" instead of hallucinating, they're limited to very low risk applications such as advertising and search.

Are humans limited to low-risk applications like that?

Because humans, even some of the most humble, will still assert things they THINK are true, but are patently untrue, based on misunderstandings, faulty memories, confused reasoning, and a plethora of others.

I can't count the number of times I've had conversations with extremely well-experience, smart techies who just spout off the most ignorant stuff.

And I don't want to count the number of times I've personally done that, but I'm sure it's >0. And I hate to tell you, but I've spent the last 20 years in positions of authority that could have caused massive amounts of damage not only to the companies I've been employed by, but a large cross-section of society as well. And those fools I referenced in the last paragraph? Same.

I think people are too hasty to discount LLMs, or LLM-backed agents, or other LLM-based applications because of their limitations.

(Related: I think people are too hasty to discount the catastrophic potential of self-modifying AGI as well)

Replying to self to answer thread questioners. No. Owen was the engineering powerhouse. Kevin was the PR front-man (the pretty face). Ron (Gorodetzsky) was the DevOps mastermind.

Who I am referring to was named Amar (his name is common enough I don't think I'm outing him). He was the SEO whisperer and "algorithm." He was literally like a spy. He would infiltrate the awful groups trying to game the front page and trick them into giving him enough info that he could identify their campaigns early, and kill them. All the while pretending to be an SEO loser like them.

There were a few other amazing people behind the scenes. I'm actually leaving out myself and my group because who wants some dude to blow his own horn? But many of us did amazing things.

There were also literally dozens of guys super high-up that were useless. Not because they were dumb, but they were too full of hubris and thought they had expertise where they didn't. Like Kevin Rose should have realised being a nice guy was his strength, and stay out of engineering, because he started dabbling in it, promoting the wrong people and ideas for the wrong reasons, and the next thing you know... BOOM. Implosion.

I even catch myself calling some of the people who were in K.Rose's event horizon "idiots" or "stupid" but when I really think about it honestly, they were reasonably bright but just given poor incentives. Hey, this NoSQL thing is awesome! Let's replace the entire (functional) MySQL portion with Cassandra. Yeah! After seven beers and two joints, this sounds like an amazing idea. Let's do it!

No.

If you find yourself, or your company founder, doing things like this, sell your equity position for whatever it's worth at that moment. Do not HODL. SELL and SHORT.

Some of the most hilarity I've ever seen was asking various 7B models: "How old is a child who was born 9 years ago?"

Paragraphs and paragraphs of complex calculations, ending up with "9.5" or "11" or some other close-but-very-wrong answer.

I wasn't a Digg founder, but I worked with them from the earliest days on Digg until its downfall.

The real story is probably way more interesting than anyone really would guess. I'll summarise it as so:

What made Digg work really was one guy who was a machine. He would vet all the stories, infiltrate all the SEO networks, and basically keep subverting them to keep the Digg front-page usable. Digg had an algorithm, but it was basically just a simple algorithm that helped this one dude 10x his productivity and keep the quality up.

Google came to buy Digg, but figured out that really it's just a dude who works 22 hours a day that keeps the quality up, and all that talk of an algorithm was smoke and mirrors to trick the SEO guys into thinking it was something they could game (they could not, which is why front page was so high quality for so many years). Google walked.

Then the founders realised if they ever wanted to get any serious money out of this thing, they had to fix that. So they developed "real algorithms" that independently attempted to do what this one dude was doing, to surface good/interesting content.

They thought they'd succeeded, or market pressures forced their hand, whatever. So they rolled it along with a catastrophic UI/UX and back-end tech rewrite all rolled up into one.

It was a total shit-show. I was involved in the "old" MySQL stack, and watched them totally fuck it up with beta software that wasn't ready for production (Cassandra, at the time, was not what it is today).

The algorithm to figure out what's cool and what isn't wasn't as good as the dude who worked 22 hours a day, and without his very heavy input, it just basically rehashed all the shit that was popular somewhere else a few days earlier.

So you ended up with a site that was ugly, fuxed, no-one in the existing wanted, and with a bland boring bunch of stories on the front-page, which was not at all compelling for anyone new showing up to check stuff out.

Instead of taking this massive slap to the face constructively, the founders doubled-down. And now here we are.

To be clear, much of the tech behind Digg was very interesting, the work Owen and many other engineers did was very interesting. The algorithm was all smoke and mirrors, though. And Kevin and his little circle of buddies were all crap engineers that tanked the business with their hubris and inexperience.

I really wish I could figure out the formula. I have had so many shirts that I wear for a year or two, then they're dead.

But there's this one shirt I bought back in 1999 that still now, today, looks about the same as it did then (I have photographs). I don't wear it any less than any of the other shirts. Just somehow... year after year... it remains decent.

I wish I could go custom-make 20 shirts out of the same material and with the same techniques as this shirt. I'd never need another shirt again in my life!

"Scott free" is way too strong. Like, a whole extra T strong. It should be scot free.

All that said, I'm not sure how this will play out. He cozied up to the right people, but so did Epstein, and he... definitely did not get away scot free.

No-one believes in the right of the individual overruling all else. I'm not sure how you are coming to that conclusion.

If you offend all of society (or more realistically, most of it), then sure, you are going to have a bad few years. No-one really disagrees there.

But if you offend some tiny segment of socieety, a single malicious and persistent "villager," then you should not have to endure that person spending all their time trying to ruin your life.

If it's one person, you would get a restraining order.

What do you do if it's 10,000 people, in a world of 7 billion? Clearly that's even smaller an influence than a single person in a village. We need a modern day equivalent of a restraining order against those 10,000 so they don't get you fired from your job, divorced from your wife, estranged from your kids, and sent packing from your bank.

This is what everyone's upset about. No-one is defending the dictator wannabe who is advocating for murdering half the planet to get socialism to finally work. Everyone agrees that guy's a nut, and it's fine if he gets fired.

No-one believes either of those things. You should familiarise yourself with the arguments and beliefs of people outside your ideological bubble.

To be charitable, I suppose what you mean is: "I don't believe gays/blacks are being genuine in their efforts to promote their agendas" and "there has been significant election interference, including against my preferred candidate (Trump|Clinton)."

Note that the latter belief is also widely held by those who preferred eg Clinton in 2016, including Clinton herself, who has loudly and proudly declared repeatedly that she would/should have been president but for election interference.

So if the latter belief is enough to get someone canceled from their job, and you support that, then I guess that includes about 90% of the population. I don't think you're being reasonable here.

As to the former belief, I think looking back at how Black Lives Matter has turned out should be enough to convince most reasonable people that some of the biggest/loudest agendas promoted over the past few years aren't genuine.

As to your last assertion, you're correct. No-one is getting canceled over minor quibbles over tax rates. I don't know anyone who is claiming that.

We are in a recession. Unless the redefinition of the word is successful, and there's a strong push to redefine the word.

So you're going to see some confusion on this topic until the politics around it settle down a bit.

Nothing. Dealers know to check the owner button.

Whenever I'm buying something on CL, like say a car, I always ask the person who answers the phone: "I'm calling about the car. Tell me X about it."

If they reply "Which car?" then you've reached a dealer, not an owner.

> Nothing discovered since the 70s compares to the advances in science and technology made in the first half of the twentieth century.

Not buying that in the slightest.

Reusable rocket boosters? Battery tech? International money movement? Everything in tech from 2000 to present?

Sure, if you just define "everything since the 70's doesn't spark my imagination" then you win this round of nihilism, but the last 50 years have been amazing.

Spend 7 years in a fever pitch screaming the worst partisan things you possibly can, and eventually even the people inside your filter bubble will start wondering... where is that evidence of Russian collusion that was 24x7 coverage for 3 years?

I thought that whole Hunter Biden thing was disinformation? We literally banned it from the internet! Wasn't that... and the whole lab leak thing supposed to be the delusions of right-wing whack-jobs?

Didn't the Buffoon President committed the worst faux pas ever in front of the Japanese prime minister? Didn't he say Charlottesville racists are OK? Hold on, if I do five minutes of research, I find some disturbing video in the moments preceding the video I was shown.

The media sound like literal sixth-grade children talking smack about the seventh-graders. The level of discourse is unbelievably immature.

I'm only surprised because I know so many intelligent people that are so deep into the filter bubble they haven't lost confidence yet. I'm glad enough people are paying attention to tell Gallup they are on to the ruse.

> Do I need to make it painfully clear that it's my subjective opinion?

Yes. Because a sentence written in the Arial font, as much as I might find that font distasteful, is more readable than the same sentence written by a Captcha-generator.

A bubble-sort written in Golang is easier to read than the same bubble-sort written in Brainfuck.

Yes, there are exceptions to the rule (I imagine someone who knows Brainfuck, but doesn't know Golang for example), but the rule is very clear in both cases. The clickbait flamewar-inducing headline is saying, essentially, Ruby is the former in the examples I gave, and Python the latter.

We're not junior-high students trying to one-up one-another with cute gotchas. We're technical professionals. We should expect higher standards of communication skills from other technical professionals.

If you mean "I have a useless opinion that no-one should care about," then state it. If you mean "I have statistical evidence that most technical professionals will be able to comprehend programs written in Ruby better than the same programs written in Python," then state it. But don't expect a good reaction when you say the former, while tricking us into thinking you were saying the latter.

Oh. So this is why clickbait and trolling have increased so dramatically? The kids are being encouraged to do it by their teachers?

Eternal September gets worse when their teachers are teaching them bad habits from the get-go.

Indeed. I started working with Ruby for at least a year before I started working with Python. To this day I still can't do anything useful in Ruby, and I'm proficient in Python.

Python is far more readable and comprehensible than Ruby.

Haven't even bothered to read TFA because it's just weird flamewar bait.

Like many others, I don't think I will bother talking about JPEG2000 as a preservation risk. I guess it is, given the bugs and difficult implementation.

But for me, the problem is just how unwieldy the files are. Nothing seems to work with them well, except ImageMagick, to convert the files to some more-usable format.

I do a lot of Blender renders, and for a while I thought I'd use 12-bit JPEG2000 so that I could capture more details in the shadows for subsequent photo editing.

Big mistake. I ended up having to write a series of very complicated scripts to convert all to regular JPEG for 99% of whatever I'm doing with the images. GIMP doesn't work well with them. My own image viewer, based on Kivy, doesn't work at all with them. MacOS generates thumbnails on them at maybe 1/10th the speed of comparable JPEGs.

In the end I just decided if I really want to work with the darks in an image I will go find the original .blend and re-render the scene. Now I render straight to regular JPEG and I save massive amounts of time.

I like how this comment just sits here, unceremoniously ignored. Perhaps you're a little too cynical in how you've framed this, but it has a correct central thesis.

The reason most people aren't as bothered by this situation as they "should be" is... things were worse under the old Taxi system for everyone we should have sympathy for.

Uber didn't make everything perfect, but they improved things dramatically.

So sure, let's keep improving things. But if the "anti-Uber" fanatics take down Uber, they won't get an improvement. They'll get the old system back, which was worse for drivers and passengers alike.