HNHacker News
TopNewBestAskShowJobs

koyote

2,431 karma · joined March 21, 2012

submissionscomments
koyote··on Lenovo’s new ThinkPads score 10/10 for repairability
What annoys me the most is that the information has become much less dense. There's a lot of unnecessary repetition. I feel like I need to feed every article through an LLM just to get a summary of it.
koyote··on Setting up phones is a nightmare
I still have one of those lying around in the draw. It's the backup phone and every time I or my partner needs to use it I am surprised at how well it still works.
koyote··on Layoffs at Block
I found it completely unreadable, similar to reading code without syntax highlighting.

Maybe he should have had AI fix up the grammar/spelling for him...

koyote··on I built Timeframe, our family e-paper dashboard
Which is weird because it's pretty straightforward to work out if you need sunscreen or not:

  * Is it any month other than May-August?  
  
  * Is it after 10am or before 4pm?  
  
Probably need some sun screen.

If you have very light skin you might want to increase the timeframe by an hour.

And if you really want to optimise your sunscreen usage and not use it if you don't have to, the real-time UV index from ARPANSA is the way to go (https://www.arpansa.gov.au/our-services/monitoring/ultraviol...).

All other apps simply display the expected UV index given the time of the day and the day of the year.

koyote··on Microsoft forced me to switch to Linux
I have the same specs in my work machine.

Task manager takes 10 seconds to load the list of processes. Right-click on the desktop takes about 1.5-2 seconds to show the 'new' context menu. Start menu is actually fast to start drawing but has a stupid animation that takes about half a second to fully load.

I sort of understand how the anti-consumer 'features' (ads) get added to a piece of software. But I have no idea how they manage to continuously degrade the experience of existing parts of the system for seemingly no one's benefit.

koyote··on Rust’s Standard Library on the GPU
I am not assuming it to be performant, but if you use this in earnest and the implementation is naive, you'll quickly have a bad time from all the data being copied back and forth.

In the end, people program for GPUs not because it's more fun (it's not!), but because they can get more performance out of it for their specific task.

koyote··on Rust’s Standard Library on the GPU
Are there any details around how the round-trip and exchange of data (CPU<->GPU) is implemented in order to not be a big (partially-hidden) performance hit?

e.g. this code seems like it would entirely run on the CPU?

    print!("Enter your name: ");
    let _ = std::io::stdout().flush();
    let mut name = String::new();
    std::io::stdin().read_line(&mut name).unwrap();
But what if we concatenated a number to the string that was calculated on the GPU or if we take a number:

    print!("Enter a number: ");
    [...] // string number has to be converted to a float and sent to the GPU
    // Some calculations with that number performed on the GPU
    print!("The result is: " + &the_result.to_string()); // Number needs to be sent back to the CPU

Or maybe I am misunderstanding how this is supposed to work?
koyote··on ThinkNext Design
I had the T431s for over a decade as well.

Replaced the keyboard (orange juice spillage), screen (upgraded to a higher resolution panel), hdd (to an SSD of course), RAM, Wifi adapter (Wifi 6) as well as the battery.

I now use an X1 Nano and while it's nice (and very light!) I am sad that the upgradability is nowhere near as good.

koyote··on Inlining – The Ultimate Optimisation
I ran into a similarish issue in C++ (MSVC++) where a small change that improved an error message led to a 10% slowdown.

The function was something like:

  int ReturnAnswerIfCharIsValid(char* c)
  {
    if(c == nullptr)
      throw std::exception("ERROR!");

    return 42;
  }
The exception line was changed to something like:

  throw std::exception("Char is not valid, please fix it!"); // String is now longer
The performance of this hot-path function went down the drain.

I fixed it by replacing the exception call with yet another function call:

  if(c == nullptr)
     ThrowException();
Other fixes might have included something like __forceinline in the function signature.
koyote··on Postal Arbitrage
I have done this specifically with the second item in the list in the OP.

Not only did I do it to get free shipping, I got it to get free international shipping.

For extra bonus CO2 points, the other item was coming from a different country. So I basically paid $0.42 to have a single packet of kool-aid shipped across the pacific ocean.

(I'd never had kool-aid before and I must say I was disappointed.)

koyote··on Xfce is great
> The only part that irritates me is having to interact with the GTK file chooser (file open dialog). Someday I might be annoyed enough to replace it.

That's probably my only annoyance as well. Is there an easy way to replace it? Not being able to see the path as a string is very "un-linux".

koyote··on Xfce is great
I've been using XFCE for several years on 4k screens and I agree that it's not great out of the box.

Once you've set it up it works pretty well though.

Now if only I could remember what I did to get it working nicely...(luckily I've had the same installation of XFCE on my machine for the past 5 years so haven't had to fiddle with that in a while)

koyote··on I dumped Windows 11 for Linux, and you should too
> Desktop Windows still has rough edges. Desktop MacOS still has rough edges. Desktop Linux still has rough edges. Pick your poison.

I think this sentiment is often overlooked as people are used to their 'poison'.

As someone who uses Linux a their main personal machine (with dual boot to Windows every now and then) as well as W11 for work, it's amazing what you get used to.

I was almost agreeing with OP, remembering bluetooth issues I had with Linux just last month when one of my headphones couldn't connect properly and I had to spend 10-15 minutes messing about with bluetooth stacks to get it working again.

But reading your comment I just realised that my current work machine doesn't even detect my bluetooth headphone's microphone and I have not found a fix yet. That machine also does not go to sleep properly (a common, real, complaint from many linux users) and I have to hibernate it manually via command line as the option does not exist in my power menu due to corporate's rules and regulations.

I also get Windows blue screens far more often than I get Linux kernel panics.

You're just so used to the issues and inconveniences that you don't even recognise them as such anymore. Issue and inconveniences from a new piece of software you're trialing stick out like sore thumb though...

koyote··on Opus 4.5 is not the normal AI agent experience that I have had thus far
I might give that a go in the future, but in this case it would've been faster for me to just do the work than to coach it for each file.

Also as this was an architectural change there are no tests to run until it's done. Everything would just fail. It's only done when the whole thing is done. I think that might be one of the reasons it got stuck: it was trying to solve issues that it did not prove existed yet. If it had just finished the job and run the tests it would've probably gotten further or even completed it.

It's a bit like stopping half way through renaming a function and then trying to run the tests and finding out the build does not compile because it can't find 'old_function'. You have to actually finish and know you've finished before you can verify your changes worked.

I still haven't actually addressed this tech debt item (it's not that important :)). But I might try again and either see if it succeeds this time (with plan in an md) or just do the work myself and get Opus to fix the unit tests (the most tedious part).

koyote··on Opus 4.5 is not the normal AI agent experience that I have had thus far
A greenfield project is definitely 'easy mode' for an LLM; especially if the problem area is well understood (and documented).

Opus is great and definitely speeds up development even in larger code bases and is reasonably good at matching coding style/standard to that of of the existing code base.

In my opinion, the big issue is the relatively small context that quickly overwhelms the models when given a larger task on a large codebase.

For example, I have a largish enterprise grade code base with nice enterprise grade OO patterns and class hierarchies. There was a simple tech debt item that required refactoring about 30-40 classes to adhere to a slightly different class hierarchy. The work is not difficult, just tedious, especially as unit tests need to be fixed up.

I threw Opus at it with very precise instructions as to what I wanted it to do and how I wanted it to do it. It started off well but then disintegrated once it got overwhelmed at the sheer number of files it had to change. At some point it got stuck in some kind of an error loop where one change it made contradicted with another change and it just couldn't work itself out. I tried stopping it and helping it out but at this point the context was so polluted that it just couldn't see a way out. I'd say that once an LLM can handle more 'context' than a senior dev with good knowledge of a large codebase, LLM will be viable in a whole new realm of development tasks on existing code bases. That 'too hard to refactor this/make this work with that' task will suddenly become viable.

koyote··on Why is the Gmail app 700 MB?
I recently made a tiny utlity app with a single screen, one dropdown and a few lines of text (it makes a single API call to populate the text).

According to my phone it's 25MB.

There's no assets, not even an app icon.

I have no idea what would be taking up more than a few hundred kb, let alone 25MB.

koyote··on It's hard to justify Tahoe icons
Uber very much falls into the category of "useful despite the awful app" for me.

It's slow, bloated, buggy and ugly. Probably one of the worst apps running on my phone.

koyote··on Kidnapped by Deutsche Bahn
The irony is that OPs train, the RE5, was actually not run by DB but by 'the competition': a private company called National Express.

https://en.wikipedia.org/wiki/National_Express_Germany

koyote··on A production bug that made me care about undefined behavior
> Hopefully future versions of C++ will mandate default initialization for all cases that are UB today and we can be free of this class of bug.

I have production code where we specifically do not initialise some data in order to be more performant (it gets set before it is read, but not at declaration as we do not have the values at that point).

I do agree that this (and numerous other footguns) make C++ a pain to work with. I also think it's too late to fix.

Ideally, all values would be initialised by default and instead you could forcefully construct something that is not initialised (e.g. something like `no_init double x[100];`). Instead, we have the bug-prone default and twenty different ways to initialise something, each with their own caveats.

C++ would be a much better language if most if not all defaults were reversed.

koyote··on Show HN: Ez FFmpeg – Video editing in plain English
You're getting a lot of flak due to how you started off your comment, but I mostly agree with you.

In my opinion there are two kinds of users: 1. Users who use FFmpeg regularly enough to know/understand the parameters. 2. Users who only use FFmpeg once in a while to do something specific.

This wrapper is superfluous for users in group number 1. But group number 2 does not really get much out of it either, for the reasons you've mentioned.

As a member of group 2, I usually want to do something very specific (e.g. remove an audio track, convert only the video, remux to a different container, etc.). A simple English wrapper does not help me here because it is not powerful enough; the defaults are usually not what I want. What I need is a tool that will take a more detailed English statement of what I want to achieve and spit out the FFmpeg command with explanations for what each parameter does and how it achieves my goal. We have this today: AI; and it mostly works (once you've gone through several iterations of it hallucinating options that do not exist...).

koyote··on The EU made Apple adopt new Wi-Fi standards, and now Android can support AirDrop
The Nexus 5x (and all models after that) had a USB-C connector in 2015, so did the OnePlus 2 (and all models after that).
koyote··on The EU made Apple adopt new Wi-Fi standards, and now Android can support AirDrop
> In Apple land, everyone already had a bidirectional, reliable connector

Wait, I thought the Apple 30-pin connector was not reversible?

USB-C has been out for over a decade now. There was only a small window of about two years where iphones had lightning and other phones did not yet have usb-c.

koyote··on .NET 10
> I can't see myself ever adopting it for any real work

Why not?

I see a lot of people saying they don't like it or won't use it but few of them list any reasons to, the ones that do raise issues from 20 years ago that have since been resolved.

Maybe you should give it a try, you'd be surprised how productive the language is and how comparatively unproductive all other languages and ecosystems are.

koyote··on Laptops with Stickers
I wonder if stickers in general are more common in the US just like bumper stickers are.

As someone based outside of the US, I rarely see people with stickers on their laptop. The very vast majority of cars also do not have bumper stickers here (and those that do might have one at most).

koyote··on Vacuum bricked after user blocks data collection – user mods it to run anyway
Maybe our threshold for cleanliness is vastly different, or you somehow managed to produce babies don't throw half of their food on the floor after every meal (3x a day)?

But even with a magic baby and magical dogs, you mentioned only spending 10 minutes a week vacuuming. I have no idea how that is possible with babies and dogs unless your threshold for when something requires cleaning is extremely high.

Before having a robot vacuum/mop I would have to go and pick up every piece of food and wipe the floor after every meal. Sure, the whole kitchen didn't technically need a mop, but there's usually also food in other places simply through the action of cooking. We cook every meal for the baby and most meals for ourselves.

Do you just leave the food and crumbs on the floor until your weekly 10 minute vacuum? In which case, yes, the notion of a robot vacuum must feel idiotic to you. The notion of a vacuum would also feel idiotic to me in that scenario as you can surely just use a broom and a dustpan for such a small amount of cleaning.

koyote··on Vacuum bricked after user blocks data collection – user mods it to run anyway
> Buy a normal vacuum cleaner for half the price, spend 10 minutes a week vacuuming your apartment

You obviously don't have a pet or a baby.

Make that 15 minutes of vacuuming AND mopping 3 times a day for a baby. Suddenly it seems very attractive to have a clean house while not having to find the time during the baby's sleep and nap time to do it manually.

You could argue the same for a dishwasher: I used to only use a single fork, glass and pot (eat out of the pot). A dishwasher seemed like the most idiotic device anyone could own if that's all you need to rinse every day. Until of course you add more people to that equation...(and maybe cook more than just pasta)

koyote··on Apple reports fourth quarter results
I could go into detail how being able to have a dozen tabs open almost always gives a better result than simply picking the first flight on google flights. But let's assume there's no difference:

Do you really want to use a phone's on-screen keyboard to type in your family's passport details, address, then credit card numbers, then review all of this to ensure your $2000 purchase doesn't have any typos or mistakes? If you have the choice to use a real computer for this, then why not? It's not like booking a big trip is something you do while sitting on a bus.

Then of course there's accommodation, itineraries, visas, trip research...

koyote··on Show HN: In a single HTML file, an app to encourage my children to invest
That is very cool! Do you have any details on how it was built or is the source code available?

Also, what happens if one of the daily missions is not completed? Is there a passive income from those?

koyote··on Comparing the power consumption of a 30 year old refrigerator to a new one
A lot of fridges in Europe are integrated into the kitchen cabinets, similar to dishwashers.

So I'd say they face the same kind of issues.

koyote··on Retiring Windows 10 and Microsoft's move towards a surveillance state
Some things that any semi-power user will notice and get angry at:

* Needing internet and a microsoft account to install the OS

* Start menu now requiring two clicks to get to programs list

* Right-click requiring two clicks to get to the options you most likely want to use (e.g. 7z unzip or opening in a specific program)

* Task manager being slow and laggy

* Random ads asking you to install a game pop up in the notification area

* ...

And then there's little bugs everywhere that just grind away at you on a daily basis:

* A tab in explorer will sometimes randomly stop accepting clicks (keyboard select works). So I have to close the tab and re-open

* The keyboard layout setting gets corrupted and there's no proper way to reset it (nevermind the fact tha this setting is now burried twenty levels deep in the new settings app)

* The settings app search does not work

* ...

It is by far the worst Windows version (beating Vista and ME to that title) in my opinion. I use linux as my daily but am forced to use Windows at work and they have of course been forced to upgrade us to Windows 11...

← PreviousPage 3 of 25Next →