Hacker News new | past | comments | ask | show | jobs | submit | NortySpock's comments login

Ah, that must be the research paper that is being referred to by this short video by the paper author: https://youtu.be/azDaPm13CT8?si=Z11qpbjNIwuhxdnB

I think there will still be room for "debugging AI slop-code" and "performance-turning AI slop-code" and "cranking up the strictness of the linter (or type-checker for dynamically-typed languages) to chase out silly bugs" , not to mention the need for better languages / runtime that give better guarantees about correctness.

It's the front-end of the hype cycle. The tech-debt problems will come home to roost in a year or two.


> It's the front-end of the hype cycle. The tech-debt problems will come home to roost in a year or two.

The market can remain irrational longer than you can remain solvent.


> not to mention the need for better languages / runtime that give better guarantees about correctness.

Use LLM to write Haskell. Problem solved?


> I think there will still be room for "debugging AI slop-code" and "performance-turning AI slop-code"

Ah yes, maintenance, the most fun and satisfying part of the job. /s


Congrats, you’ve been promoted to be the cost center. And sloppers will get to the top by cranking out features you will need to maintain.

> slopper

new 2025 slang just dropped


That's sloppy programming. You are promoted.

Just wait till 'slopper' starts getting classified as a slur

"Slopper" maybe not, but "slophead"...

A pre-existing problem, but it's true LLMs will make it worse.

You work in the slop mines now.

youtube link if you want to watch the interaction: https://www.youtube.com/watch?v=roEsJcQYjd8

"minimum 10% tariff on all imported goods and materials" certainly sounds like a way to increase costs.

Increasing costs seems generally worse for an economy than decreasing costs.

I feel like most people could follow this logical chain of reasoning to a conclusion of "thus you have a elevated risk for an economic recession compared to the state of affairs before the tariffs."


For Beyond All Reason, it seems the Spring/Recoil engine will eventually decide to "close the action window", so that if one player is lagging hard, they simply submitted no actions for that "round" and the rest of the players keep going.

I know because I've gotten to the point in the late game where my computer can't simulate at realtime, and I can no longer control my units, but everyone else keeps playing.

Conveniently you can even still sort of chat in this state, and ask a teammate to assume control of your army on your behalf.


Yes, for nearly any restaurant this is the unspoken recommendation, and sometimes enforced automatically if your group is larger than 6-8. Source: I am an American.


30 is hot.

20 is nice.

10 is cold.

0 is ice.

Not hard to remember for converting Celsius to how humans typically perceive temperature.


My problem isn't remembering the scale, it's that Fahrenheit offers me double the effective resolution and descriptive accuracy without awkward decimal points in the numbers used.

I like my room at 73F, not 72F or 74F, and I can feel the difference. That's 22.77C. :-/


My main observation in temperature scale and imperial lengths discussions om the Internet is that Americans seem to have a strange aversion against fractional parts of numbers, as if those were irrational.

(On the other hand a lot of Americans consume fava beans.)


Au contraire base 12 measurement is _all about the fractions_. 12 can be divided evenly by 2, 3, 4, and 6. Metric gets 5 and 2. By that measure it's y'all that are afraid of the small numbers.

US units are pinned to Metric standards anyway. We're just using the most creative ratios. :)


American, I think we use fractions all the time: 7/8 inch hardware tools, 3/4 measuring cup in cooking, etc. Especially awkward when you have human distances, because you have to mix feet + inches: 3 feet 4 and 1/4 inches.

It was a dream once I got a metric tape measurer and realized that using centimeters eliminates the need to do annoying conversions.


I found this take from 'Torque Test Channel' (tests battery tools) humorous: https://www.youtube.com/watch?v=2QUum9NymZY


Brilliant and hilarious.


Most temperature sensors are accurate to 0.1C. Most weather forecast is 0.5C resolution. So yeah only explanation to Americans behavior is an aversion against fractional numbers as you said.


> I like my room at 73F, not 72F or 74F, and I can feel the difference.

I suspect you wouldn't notice if it changed by a degree.


I like my room exactly at 21C. Any F number is Greek to me.

Remembering is more about what you are used to here.


Perceiving temperature involves dew point and relative humidity, not just temperature.


Okay. Or, hear me out.

On a scale from 0-100, you have very cold and very hot.

Or you’ve got from 0-45. Where 0 is “meh” cold and 45 is incredibly warm.

So you’ve got a nice little 0-100 scale that all humans are going to experience just living that goes from very cold to very hot.

Or you’ve got a useless 0-100 scale that the bottom just means freezing, and ignores every pain point of being really cold below that, and anything really greater than 50C only has practical applications in cooking.


0 isn't the bottom of the celcius temperature scale, there are negative numbers


Cool yeah let's compress the entire scale to 0-35 and waste the other 65 up to 100 lol yeah what a great scale


The "entire scale" has no maximum. So your waste of the Celsius scale from 35-100 is Fahrenheits waste of the scale from 0-32 or whatever you're trying to base your comments on.


Just a point here, "good code" is sometimes subjective, and depends on understanding the context of what the code is doing. What you think is good code might be overly verbose to another person, or overly terse, or have poorly named variables, or not have sufficiently conservative guard clauses, or throw insufficiently-granular exceptions. What you think is confusing code might lack context for where it is in the stack and what problems it needs to solve at that layer of the stack.

You can also read critical reviews of someone else's work, compare them with the work in question, and see if the critic's punches land or if they look like misses.

https://qntm.org/clean

^ This, I thought, was a good takedown of Clean Code, highlighting some cases where Bob Martin made too many overly thin functions that lacked meat and made it hard for the reader to gain context for what the function was trying to do. [1]

I would also say, reading "the same code" in different programming languages might get you a feel for if you prefer code to be more verbose or more terse, more explicit or more implicit. e.g. https://rosettacode.org/wiki/Globally_replace_text_in_severa...

[1] sometimes derisively referred to as "lasagna code" or "baklava code" -- https://www.johndcook.com/blog/2009/07/27/baklav-code/


As a data engineer, who is regularly fighting

- "these two databases have different SQL dialects"

- "did we miss a few rows due to poor transaction-isolation when trying to query recently changed rows on the upstream database"

- "is there some checksum of a region of cells that accepts any arrangement of rows and columns that doesn't require me to think about ordering?"

- "is there a subtle data format conversion problem silently mangling my data?"

...I've toying with trying to find a way to serialize everything consistently into something that can be XOR'd, then compare the output of XOR for two tables in two different databases that should be identical, without having to do some giant order-by comparison. Ideally, if necessary, this could be done by eye with two query windows pointing at two different databases...

Basically, Datafold's datadiff, but in a way that could plausibly be home-rolled for on-premise applications (especially quick spot-checks), and not be a total maintenance nightmare...

https://github.com/datafold/data-diff

Don't have anything working yet, but it just seems like one could at least xor a bunch of integers and get something useful... Somehow.


An algorithm I've heard of in multiplayer game programming, in which massive state must be transmitted to many clients ... xor the previous game state with the new game state on the server ... most of the result will be zeros because most of the new state is the same as the old state. Then compress it. All those zeros compress really really well which is great for network traffic. Then transmit the compressed xor'ed state, client decompresses and xors with current state to get new state. Don't remember where I first heard of this but it was probably 15 or 20 years ago.

Really clever. Never tried to implement such a thing though. Sounds fiddly and hard to debug.


In this framing MPEG video compression is just a more sophisticated XOR.


I noticed from the docker overlay filesystem that the container was spraying files all over the disk. (Ephemeral, destroyed on container shutdown, sure, but I wanted to reduce write-wear on my ssd...)

I tried setting it up with /tmp as a tmpfs (ramdisk) but it then refused to start...

Anyone know any broad-spectrum docker incantations to force all overlay writes to RAM, for a container?


> I wanted to reduce write-wear on my ssd

Modern SSDs are pretty good at things like wear levelling.

For example [1] reports that a bunch of 256 GB SSDs lasted to 2000+ terabytes written, and a handful up to 7000 terabytes written. So you could saturate a 100 megabit internet connection for 5 years before even a small SSD would wear out. And an SSD 4x the size has 4x the life.

If you're running on a raspberry pi with a microsd card for storage, feel free to keep worrying though :)

[1] https://www.reddit.com/r/chia/comments/mukiwz/are_we_overthi...


> the container was spraying files all over the disk

Right, that's basically the point...the Warrior downloads files, compresses them, and uploads them for archival. This necessarily requires staging the files somewhere between download and upload.

> Anyone know any broad-spectrum docker incantations to force all overlay writes to RAM, for a container?

Why would you want this? This sounds like a terrible footgun.


The Warrior doesn't resume old jobs after a power cycle, so what's the point of committing anything at all to non-volatile storage?


They say exactly why they want it... "I wanted to reduce write-wear on my ssd"


I think you'll just need to mount it at the right place, with right permissions.

Demonstrated here https://stackoverflow.com/questions/39193419/docker-in-memor...


You can put the entire docker directory in a ramdisk. Same as you would when trying to move it to a secondary harddisk. Risky though as a reboot would wipe everything


Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: