It's the only website I know of where queries can just randomly fail for no reason, and they don't even have an automatic retry mechanism. Even the worst enterprise nightmares I've seen weren't this user unfriendly.
There are sectors where pre-ML approaches still dominate.
Among screen reader users for example, formant-based TTS is still wildly popular, and I don't think that's going to change anytime soon. The speed, predictability and responsiveness are unmatched by any newer technology.
To a European like me, United was such a weird airline to fly.
There were actual commercials played before the safety video, the cabin crew warned passengers to make sure children cannot see the adult content they're watching (can you get more American than that?), and their credit card was offered multiple times during the flight. At least the WiFi was reasonably cheap.
Over here, that stuff would never fly (no pun intended), except maybe on Ryanair or other extremely low-cost carriers. On e.g. a Lufthansa longhaul flight, which are priced similarly and cover the same route I flew (fra-ord), it would be unthinkable.
What has become much more degraded in Europe than America is the domestic flights.
Try flying Lufthansa (or one of their half dozen subsidiaries created almost entirely to give worse service) anywhere inside of europe. Everything is a money grab and the service and boarding are terrible.
United maintains a relatively consistent experience between domestic and international, minus the free alcohol.
> Try flying Lufthansa (or one of their half dozen subsidiaries created almost entirely to give worse service) anywhere inside of europe. Everything is a money grab and the service and boarding are terrible.
FWIW, I just took such a flight and didn't notice anything that compares unfavorably to a domestic U.S. airline. (To be clear, it certainly wasn't better either.) Is there anything specific you can point to?
Every drink is for sale, even coffee. They aggressively use the bag sizer and try to take bags away at the gate, under the ruse that their tiny Airbus overhead bins cannot fit them (they can).
Just put it in the hold. I can’t stand waiting for the cheapskate to find somewhere to put their 45kg “rucksack” in the over head bins just to save a tenner.
Checked luggage is lost luggage, in my experience. I think I've only checked maybe two or three bags in the past 20 years (after two lost-luggage incidents prompted me to switch to a carry-on-only packing regimen), and then only because I needed to transport liquids in excess of the in-cabin allowance.
I also like to leave the airport after my plane arrives, not stand around a conveyor belt for some unguessable amount of time.
But I get your frustration; I'm the kind of person who barely breaks stride out of the aisle and into my row as I sling my bag up into the bin. It makes me want to scream when someone is standing there in the aisle for 30+ seconds, holding up the boarding process.
Then again, the airlines are to blame as well, most of them having terribly inefficient boarding processes.
The airlines deserve no money above the price of a ticket. The way i see it the hold comes free with the rest of the plane. I see no reason why i need to pay another 40 pounds to bring a case, so i will shamelessly abuse my allowance. (if it was a reasonable amount like a tenner, that would be a different story, as it stands it's a cash grab)
"Put it in the hold" is a decent argument for point-to-point flights, or when you do gate checking. Otherwise it's a crap-shoot whether your stuff makes it (which can end a two-day business trip before it begins), what shape it will be in, and how long you'll have to wait. As soon as you have a connection, all bets are off.
How long you'll have to wait is mostly a function not of the airline, but of the arrival airport and the competence of the handling company.
I had to walk away from a $600 ticket that I booked at the last minute b/c in the 30 seconds between the time I paid for the ticket and the time the booking returned, the connecting flight filled up and I had to wait a day I didn't have for the next one. Couldn't get a single consideration from anyone, they said they couldn't cancel the ticket b/c the first leg was still available. Just had to walk away from the money and find another airline.
I'm sure it happens on every airline but man I was pissed. They go to the bottom of the list until the next tomfoolery occurs.
I had paid $150 or so for a extra legroom seat (on lufthansa). They did some last minute tomfoolery and my seat changed on my phone as I boarded the plane.
I was now not even in a premium economy row, but just a normal seat. I tried telling the flight attendant, who gave no shits - she just said "the plane isn't full, sit anywhere when we are in the air."
Fine, I found a premium economy aisle (still no extra legroom exit row seat, but whatever), and then filed for a refund when I landed.
They responded to my request for a refund with a form letter apologizing, but no refund. Then ignored any follow up. I had to do a charge back (no joke).
Them:
> I hope this message finds you well. I am writing to sincerely apologize for the seating issue you experienced during your recent flight. I understand that you requested an exit row seat and were not able to be accommodated, which is understandably frustrating.
> We strive to honor all seating preferences, and I regret that we were unable to meet your expectations in this instance. Please know that your feedback is important to us, and we will review our seating allocation processes to prevent similar situations in the future.
> Thank you for your understanding and patience. If there is anything further we can assist you with or if you have any other concerns, please don’t hesitate to reach out.
> We hope that you put your trust in Lufthansa and partners also in the future and continue to fly with us.
Me: This wasn't a seating preference, I paid $130 to reserve this seat. This is something I purchased. I'd expect you'd at least issue a refund and extend me some sort of credit to make up for this.
I love how everyone is dunking on airlines (it's easy - I do too!), yet airlines are one of the least profitable business sectors to be in.
It's like dunking on QSR, but worse. These things are practically on welfare.
Everything is fungible, high risk, extremely expensive, extremely regulated. The margins are almost nil. They all fly the same planes. You can compete on "experience", and that's basically it.
Their dingy little ads, baggage fees, and wifi upsells are the the best they can muster. That's the entire farm, and they're scraping by as best they can. This is every single airline.
What a awful, utterly unrewarding business to be in.
> United maintains a relatively consistent experience between domestic and international, minus the free alcohol.
A consistent extremely mediocre experience I guess. I've flown Lufthansa, Air France, Air Canada and United with a toddler and I'd get out of my way to avoid United in the future, never seen staff everywhere in that airline that was more unpleasant and unhelpful, especially with young children, as much as when flying United.
To give my own anecdata to counter yours: every time I've flown with children on United they've gone out of their way to make it straightforward. The only better airline I've encountered for it is Virgin Atlantic, which is, to be fair, better at everything.
Interesting; I just flew Virgin Atlantic for the first time a week ago (SFO->LHR) and found the experience aggressively mediocre. Reservation management and check-in experience with their app was the hottest of hot garbage, boarding was slow, in-flight meal below average, general comfort just kinda okay. No complaints about the crew, at least.
I generally fly United (while they are also aggressively mediocre, they mostly fly direct to the places I need them to, and their pricing is generally good on the routes I take), and I'd honestly rate United higher than Virgin Atlantic, which surprised me. Maybe my single experience was a fluke, but it was a really bad first impression.
And I say this as someone who absolutely adored Virgin America back when it existed. Bizarre that the two could be so different.
Europe will financialize everything just slower and with more regulation. Branded credit cards are coming. See Brussels Airlines and Mastercard
A well optimized domestic USA airline makes money from credit cards, points, trip insurance, upsells, and segments the consumer into a dozen bins based on what they’re willing to spend for a couple more inches of leg room.
Not sure about your bit of Europe but I’ve had branded credit cards from European airlines here in Europe for a long time. They’re definitely past ‘coming’.
Not as lucrative for me the holder as you’d get it the US, but I can’t really imagine being without one.
As I recall, that's because it's basically impossible to have a US rewards card in Europe. European authorities tightly cap interchange fees (the fees that come out of the merchant's end of the deal).
US issuers are much less regulated. In the US there are cheap cards that offer no perks and take a small (0.2-1%) cut from the merchant, and the perks cards that have lots of perks and take a bigger (3.5%) cut from the merchant. The CC companies, naturally, want more people to use perks cards so they get more of a cut, so to encourage consumers to use these cards they give some of it back to the users in the form of these rewards.
This model recently came under attack when a whole bunch of merchants brought an anti-trust lawsuit against Visa and MC for their requirement that if you wanted to accept the cheap cards you had to accept the expensive cards as well, merchants want to be able to accept the cheap cards and reject the higher tier cards. The negotiations about that settlement continue, so we'll have to see how it all shakes out, but it could result in a major limiting of American reward cards. Or maybe not, always in motion the future is.
The EU parliament passed a law capping interchange fees at 0.3% (for local personal cards, business has some other limit that I don't remember) so there is just no money to offer rewards to customers of European banks. Much better for merchants, lower prices overall mean probably better for poorer folks, worse deal for wealthy people with good credit who pay attention and pay their bills in full every month. Speaking as an American one of those people who benefit from rewards cards, I think that it is better for society to go with the European choices than the American.
The credit card thing doesn't surprise me. I expect United makes a ton of revenue from the card. With how credit card transaction fees are capped in Europe, I doubt it's worth it for European airlines to bother pushing their branded card much, if they even have one in the first place.
I was on a Virgin Atlantic flight last week, and while there weren't ads before the safety video, there were three ads before every movie I tried to watch... and it was the same three ads each time.
I flew Turkish in October, and was annoyed to find the movies and TV shows heavily censored, including blanking out or dubbing over minor swear words. It was also wild to see the Qur'an in the entertainment system's reading library. (No judgement there, just notable as I've never seen the Christian Bible present on other airlines.)
I think you're just falling victim to the usual thing where what you're used to feels normal, and everything else seems weird. I've definitely experienced the same as an American, when flying on European, Latin American, and Asian airlines.
Swiss air is a discount airline ever since its bankruptcy and restart right? I've only flown them once to Beijing, it was slightly better than Aeroflot.
> On e.g. a Lufthansa longhaul flight, which are priced similarly and cover the same route I flew (fra-ord), it would be unthinkable.
I fly both airlines regularly, United is _vastly_ better from a hard product perspective, a soft product perspective, and _especially_ a service recovery standpoint.
The credit card thing is easily ignored, but you used to heard it often on European flights too before branded credit cards got wiped out there. I've never heard an announcement about adult content, and have taken over 90 United flights this year.
> except maybe on Ryanair or other extremely low-cost carriers.
I fly on Lauda most often, who are operated by Ryanair. You show up, you get on, you sit down, a couple of hours you get off again. A trolley comes round with drinks and snacks, but it's a short journey even with a small child. Can't you just stick an orange and a bottle of water in your bag? It's what the Austrians do.
The first time I flew over with my small son he was three, and having been up since 5am was getting a little fractious and fidgety, so I explained he was probably a bit tired and bored and maybe he'd like to eat something and have a sleep, and I'd wake him up once we were back over land.
A bit later on someone further up the plane started remonstrating with the cabin crew that they didn't have the sandwich she wanted on the trolley, eventually shouting "IF IT'S ON THE MENU YOU GAVE ME I SHOULD BE ABLE TO HAVE THE DAMN SANDWICH!"
Well that shut everyone up.
And in the ringing silence that followed, a little voice, with the punch and clarity that only 3-year-olds have, that Brian Blessed or Meat Loaf would have given any limb you care to mention for, piped up:
"DADDY, DOES THAT LADY NEED A SNACK AND A WEE NAP TOO?"
I see businesses are getting you to blame the law rather than their activities which the law requires them to disclose. Personally, I’d expect to be paid before acting as a lobbyist.
I remember the days when movies shown on aircraft were edited for both content and dialogue.
These days, with so much content expected to be available at your fingertips, both in your seat and people bringing iPads on board loaded with R-rated streams, it's expected to be mindful of your neighbor on such things.
How that is distinctly "American" I don't quite understand.
THe Chinese seem to be extremely good at taking western products and just layering on tons of incremental improvements, which make their versions that much better. It's the Western companies that actually come up with the original idea, whatever good that does them.
Well, allow me to make a suggestion, that the US came up with the original idea, in 1947 - the transistor, and has been capitalizing on that ever since. Similar to how Germany had been capitalizing on the invention of combustion engine and various chemical processes for a century. Now, the curve of innovation on top of the fundamental invention (of the transistor) is in the flattening out region, where all the low hanging fruits had been taken down, and now it's about the remaining 5% of polishing - something that the labor force of well fed and comfortable nation is not really motivated to do.
> It's the Western companies that actually come up with the original idea, whatever good that does them.
I think that's a dangerous assumption to make. Certainly it's true that for most major technologies so far, western countries were first - but that's probably mainly because China's been busy playing catch up. But now the Chinese have huge numbers of factories, suppliers big and small, machine shops, PCB fabs and experienced engineers. You really think they're not coming up with original ideas?
Any engineer will tell you a new product is a little bit of idea and a lot of execution. The Chinese are able to execute in a way that the west isn't any more.
That's a generalization with some truth, but in this case it was blatantly obvious that iRobot was not putting much effort into improvement - or was not effective at improvement. They basically ignored the moat and relied on their headstart to the point that even brand new entrants to the market could equal or overtake them in an initial product offering.
And the business model aspects they relied on for their protective moat - e.g. mass commercial electronic production - was generalized and massively optimized in China (not just for vacuum robots but mass commercial electronics).
It is my understanding that muds (and all the flavors of Mush in particular) can sort-of do it, by letting players create their own story through roleplay, supported by an extremely open (and often player-modifiable) world, as well as good admins / GMs.
That is more like "computer tabletop", however, and doesn't scale beyond a small number of players.
There's a flywheel where programmers choose languages that LLMs already understand, but LLMs can only learn languages that programmers write a sufficient amount of code in.
Because LLMs make it that much faster to develop software, any potential advantage you may get from adopting a very niche language is overshadowed by the fact that you can't use it with an LLM. This makes it that much harder for your new language to gain traction. If your new language doesn't gain enough traction, it'll never end up in LLM datasets, so programmers are never going to pick it up.
> Because LLMs make it that much faster to develop software
I feel as though "facts" such as this are presented to me all the time on HN, but in my every day job I encounter devs creating piles of slop that even the most die-hard AI enthusiasts in my office can't stand and have started to push against.
I know, I know "they just don't know how to use LLMs the right way!!!", but all of the better engineers I know, the ones capable of quickly assessing the output of an LLM, tend to use LLMs much more sparingly in their code. Meanwhile the ones that never really understood software that well in the first place are the ones building agent-based Rube Goldberg machines that ultimately slow everyone down
If we can continue living in the this AI hallucination for 5 more years, I think the only people capable of producing anything of use or value will be devs that continued to devote some of their free time to coding in languages like Gleam, and continued to maintain and sharpen their ability to understand and reason about code.
* One developer tried to refactor a bunch of graph ql with an LLM and ended up checking in a bunch of completely broken code. Thankfully there were api tests.
* One developer has an LLM making his PRs. He slurped up my unfinished branch, PRed it, and merged (!) it. One can only guess that the approved was also using an LLM. When I asked him why he did it, he was completely baffled and assured me he would never. Source control tells a different story.
* And I forgot to turn off LLM auto complete after setting up my new machine. The LLM wouldn't stop hallucinating non-existent constructors for non-existent classes. Bog standard intellisense did in seconds what I needed after turning off LLM auto complete.
LLMs sometimes save me some time. But overall I'm sitting at a pretty big amount of time wasted by them that the savings have not yet offset.
> One developer tried to refactor a bunch of graph ql with an LLM and ended up checking in a bunch of completely broken code. Thankfully there were api tests.
So the LLM was not told how to run the tests? Without that they cannot know if what they did works, and they are a bit like humans, they try something and then they need to check if that does the right thing. Without a test cycle you definitely don’t get a lot out of LLMs.
You guys always find a way to say "you can be an LLM maximalist too, you just skipped a step."
The bigger story here is not that they forgot to tell the LLM to run tests, it's that agentic use has been so normalized and overhyped that an entire PR was attempted without any QA. Even if you're personally against this, this is how most people talk about agents online.
You don't always have the privilege of working on a project with tests, and rarely are they so thorough that they catch everything. Blindly trusting LLM output without QA or Review shouldn't be normalized.
A LOT of people, if you're paying attention. Why do you think that happened at their company?
It's not hard to find comments from people vibe coding apps without understanding the code, even apps handling sensitive data. And it's not hard to find comments saying agents can run by themselves.
I mean people are arguing AGI is already here. What do you mean who is normalizing this?
I fully believe there are misguided leaders advocating for "increasing velocity" or "productivity" or whatever, but the technical leaders should be pushing back. You can't make a ship go faster by removing the hull.
And if you want to try... well you get what you get!
But again, no one who is serious about their business and serious about building useful products is doing this.
> But again, no one who is serious about their business and serious about building useful products is doing this.
While this is potentially true for software companies, there are many companies for which software or even technology in general is not a core competency. They are very serious about their very useful products. They also have some, er, interesting ideas about what LLMs allow them to accomplish.
I am not saying you should be a LLM maximalist at all.
I am just saying LLMs need to have a change-test cycle, like humans, in order to be effective. But looks like your goal is not really to be effective at using LLMs, but to bitch about it on the internet.
I wish I could just ship 99% AI generated code and never have to check anything.
Where is everyone working where they can just ship broken code all the time?
I use LLMs for hours, every single day, yes sometimes they output trash. That’s why the bottleneck is checking the solutions and iterating on them.
All the best engineers I know, the ones managing 3-4 client projects at once, are using LLMs nonstop and outputting 3-4x their normal output. That doesn’t mean LLMs are one-shotting their problems.
You are overlooking a blind spot, that is increasingly becoming a weakness for devs. You assume that businesses care that their software actually works. It sounds crazy from the dev side but they really don't. As long as cash keeps hitting accounts the people in charge MBAs do not care how it gets there and the program to find that out only requires one simple unmistakable algo Money In - money out.
evidence
Spreadsheets. These DSL lite tools are almost universally known to be generally wrong and full of bugs. Yet, the world literally runs on them.
Lowest bidder outsourcing. Its well known that various low cost outsourcing produces non functional or failed projects or projects that limp along for years with nonstop bug stomping. Yet business is booming.
This only works in a very rich empire that is in the collapse/looting phase. Which we are in and will not change. See: History.
I once toured a dairy farm that had been a pioneer test site for Lasix. Like all good hippies, everyone I knew shunned additives. This farmer claimed that Lasix wasn't a cheat because it only worked on really healthy cows. Best practices, and then add Lasix.
I nearly dropped out of Harvard's mathematics PhD program. Sticking around and finishing a thesis was the hardest thing I've ever done. It didn't take smarts. It took being the kind of person who doesn't die on a mountain.
There's a legendary Philadelphia cook who does pop-up meals, and keeps talking about the restaurant he plans to open. Professional chefs roll their eyes; being a good cook is a small part of the enterprise of engineering a successful restaurant.
(These are three stool legs. Neurodivergents have an advantage using AI. A stool is more stable when its legs are further apart. AI is an association engine. Humans find my sense of analogy tedious, but spreading out analogies defines more accurate planes in AI's association space. One doesn't simply "tell AI what to do".)
Learning how to use AI effectively was the hardest thing I've done recently, many brutal months of experiment, test projects with a dozen languages. One maintains several levels of planning, as if a corporate CTO. One tears apart all code in many iterations of code review. Just as a genius manager makes best use of flawed human talent, one learns to make best use of flawed AI talent.
My guess is that programmers who write bad code with AI were already writing bad code before AI.
> but LLMs can only learn languages that programmers write a sufficient amount of code in
i wrote my own language, LLMs have been able to work with it at a good level for over a year. I don't do anything special to enable that - just front load some key examples of the syntax before giving the task. I don't need to explain concepts like iteration.
Also llm's can work with languages with unconventional paradigms - kdb comes up fairly often in my world (array language but also written right to left).
I don't think this is actually true. LLMs have an impressive amount of ability to do knowledge-transfer between domains, it only makes sense that that would also apply to programming languages, since the basic underlying concepts (functions, data structures, etc.) exist nearly everywhere.
If this does appear to become a problem, is it not hard to apply the same RLHF infrastructure that's used to get LLMs effective at writing syntactically-correct code that accomplishes sets of goals in existing programming languages to new ones.
> LLMs have an impressive amount of ability to do knowledge-transfer between domains, it only makes sense that that would also apply to programming languages, since the basic underlying concepts (functions, data structures, etc.) exist nearly everywhere.
That would make sense if LLMs understood the domains and the concepts. They don't. They need a lot of training data to "map" the "knowledge transfer".
Personal anecdote: Claude stopped writing Java-like Elixir only some time around summer this year (Elixir is 13 years old), and is still incapable of writing "modern HEEX" which changed some of the templaring syntax in Phoenix almost two years ago.
What's the state of Gleam's JSON parsing / serialization capabilities right now?
I find it to be a lovely little language, but having to essentially write every type three times (once for the type definition, once for the serializer, once for the deserializer) isn't something I'm looking forward to.
A functional language that can run both on the backend (Beam) and frontend (JS) lets one do a lot of cool stuff, like optimistic updates, server reconciliation, easy rollback on failure etc, but that requires making actions (and likely also states) easily serializable and deserializable.
You can generate those conversions, most people do.
But also, you shouldn’t think of it as writing the same type twice! If you couple your external API and your internal data model you are greatly restricting your domain modelling cability. Even in languages where JSON serialisation works with reflection I would recommend having a distinct definition for the internal and external structure so you can have the optimal structure for each context, dodging the “lowest common decimator” problem.
I understand your point, and I agree with it in most contexts! However, for the specific use case where one assumes that the client and server are running the exact same code (and the client auto-refreshes if this isn't the case), and where serialization is only used for synchronizing between the two, decoupling the state from it's representation on the wire doesn't really make sense.
This is also what really annoyed me when I tried out Gleam.
I'm waiting for something similar to serde in Rust, where you simply tag your type and it'll generate type-safe serialization and deserialization for you.
Gleam has some feature to generate the code for you via the LSP, but it's just not good enough IMHO.
Rust has macros that make serde very convenient, which Gleam doesn't have.
Could you point to a solution that provides serde level of convenience?
Edit: The difference with generating code (like with Gleam) and having macros generate the code from a few tags is quite big. Small tweaks are immediately obvious in serde in Rust, but they drown in the noise in the complete serialization code like with the Gleam tools.
> Rust has macros that make serde very convenient, which Gleam doesn't have.
To be fair, Rust's proc macros are only locally optimal:
While they're great to use, they're only okay to program.
Your proc-macro needs to live in another crate, and writing proc macros is difficult.
Compare this to dependently typed languages og Zig's comptime: It should be easier to make derive(Serialize, Deserialize) as compile-time features inside the host language.
When Gleam doesn't have Rust's derivation, it leaves for a future where this is solved even better.
The simplest possible such method? Single-use age verification codes, generated and validated by the government, sold on physical scratch cards with in-store verification of ID, piggybacking on the infrastructure we already use for selling alcohol and cigarettes.
This would be far easier to implement for websites too. You'd just have a single, unauthenticated API endpoint which, given a code, tells you if the code is valid (and marks it as used). Integrating with such an API is about 1 day of work for a competent dev. Even open, non-profit platforms like Mastodon could easily implement such a mechanism.
Scratch cards wouldn't have to be the only way of getting such codes. THe vast majority of people could just generate them in their banking app or whatever (which would still be far more privacy friendly than the current ID verification mechanisms).
I see this argument repeated over and over on HN, with 0 evidence for it. Any
"evidence" people cite is usually of the "politicians are evil, so this should be obvious by definition" kind, sometimes of the "they tried x in the past, so surely some unrelated y they're trying to pass in the future is also about x" kind.
I haven't seen a single leak, a single admission from somebody trying to pass a law like this, that surveillance is actually the goal here. There are far too many politicians trying to pass laws like these, in very different countries across the world, for some kind of giant global conspiracy to stay undetected.
> There are far too many politicians trying to pass laws like these, in very different countries across the world, for some kind of giant global conspiracy to stay undetected.
This is today's top agrument! "There are far too many gangsters each operating in their own district for some kind of notion of organized crime to be credible".
You don't need conspiracy, you need the incentives.
The state always thinks of self-preservation. Any bureaucrat is aligned to this goal by getting the benefits from the state. So, the more power it has over its citizens, which is the first threat it, the more safe it is and the less opinions of citizens matter.
Understanding this, every citizen must think carefully about giving away more power to the state.
> Analyze posts of people giving stock tips on WSB, Twitter, etc and rank by accuracy.
Didn't somebody make an ETF once that went against the prediction of some famous CNBC stock picker, showing that it would have given you alpha in the past.
> seems to be a way harder problem for generic free form comments.
That's what prediction markets are for. People for whom truth and accuracy matters (often concentrated around the rationalist community) will often very explicitly make annual lists of concrete and quantifiable predictions, and then self-grade on them later.
long-term his choices don't do well, so the Inverse Cramer basically says "do the opposite of this goober" and has solid returns (sorta; depends a lot on methodology, and the sole hedgefund playing that strategy shutdown)
It's the only website I know of where queries can just randomly fail for no reason, and they don't even have an automatic retry mechanism. Even the worst enterprise nightmares I've seen weren't this user unfriendly.
reply