pitched
pitched on Hacker News
Completely anecdotal but I did find the better engineers tended to take to the AI tools quickest. At the beginning, tokens was probably a good way to pick them out. “Good” being entirely subjective around a product focus/mindset. If you ever find a good metric like that, it is critically important to never tell anyone, else it will quickly become not good.
Flattened might be an exaggeration but they lost many a customer with boneheaded decisions like blocking subscription access to OpenCode. I don’t think they’re past the point in their growth curve where tinkerers don’t matter anymore. Again, definitely not flattened but not as parabolic as they could be.
I subscribed to Spotify so my kids could have screen-free content. They can listen to music whenever they want.
Spotify now has podcasts and videos and it’s all a lot less that screen-free, just-music space I wanted my kids to have.
Now I can either vibe-code a replacement and start pirating or just deal with it? What a terrible deal.
I still use the old form and don’t mind too much if anyone gets the wrong idea. It’s a good way to find other like-minded people.
Hacking in the way presented in the article is vandalism. A door being unlocked isn’t permission to enter. A lock with a flaw doesn’t mean it can be removed.
This has to be either activists or trolls and my bet is on trolls. Aiming to cause the most damage for lols.
The centaur / reverse-centaur is another interesting way to describe this split.
The title is relevant to the article. The first line walking back the statement was a bit cowardly but understandable given how easily angered it could make readers and that fact makes it funny and the reason I kept reading. The rest of the article backs up the title.
Also Aaron is on the Rails Core team and probably colleague of the author at Shopify. Sam is also a colleague of the author at Shopify. The target audience of this article is people deep in that same culture and definitely not you.
> GCC then compiles that together with a runtime that stands in for the rest of the console
A change of language in this new generation. I would call this an emulator that does translation ahead of time instead of the normal JIT. Ahead-of-time translation is now called recompilation, apparently.
Canada too, for something closer to you. Doors are rarely locked, even in big cities.
The dev in /dev/null is device, not developer. So you’re not a device but you can still be a developer.
GitHub is a social network already. You’re logging into one social network with another social network. And I’m complaining about that fact on a different social network. What has the world come to?
Cracked APKs are a thing that have a large market.
Convoluted encryption algorithms are something the games industry has been playing at forever and they haven’t made something uncrackable yet.
Orion browser on iOS supports this very seamlessly, both pip and background.
Hell no but the days of saas products are gone. Why pay for that when I can have agent build an equivalent in 10 minutes?
Not saying this counts specifically but I love seeing devs who hate Jira build something like Jira to solve managing a team of agents. It’s a beautifully ironic thing to watch.
The skill required to get an AI to cooperate is not the same one as the craft of coding. Typing what it says out yourself will never get you there.
It’s like calling yourself a mechanic because you brought your car to a shop and paid for a mechanic to do an oil change.
Ads can still be served to pirates, especially if they’re product placements baked into the assets shipped.
From the amount of activity and how much this all smells like heavy AI usage, I have zero doubt that something will be working shortly after launch. The agents won’t sleep until it boots.
This project is the first hint I’ve seen of the complete destruction of all copy protection schemes. Scary times for any industry that relies on those!
> cannot do precise things like coding software since humans will never be able to use natural language to specify their requirements.
To your point, this example. The issue expressed here is with humans, not AI. We are still pretty terrible at writing specs. TBF, the AIs are too but that wasn’t being voted on.
FTA, this is a fault-tolerant server where all hardware is redundant and can be hot swapped while running. The servers you’re thinking about are redundant at the software level so rebooting one server won’t cause the service to go down.
What is remarkable is that apparently that VOS thing has never crashed. I wonder if they also do software redundancy under the hood to keep uptime going during reboots. If the CPU is hotswappable, it must have something.
A kid can follow a basic “todo list” tutorial and open a port on your home router then get friends to chat on that. In that case, how is the government enforcing a ban? We’re expecting some agency to do constant port scans on all its citizens to look for anything allowing writes?
The responsibility is on parents. We can try to push some aspects onto government but that will not reduce any responsibility of parents.
The issue I find with this is that the frontier models still outperform the small finetuned model on its specific task. So much so that the ROI on doing fine tunes is likely negative. I would love to hear some specific example where it did provide value though, if any has any. That would be helpful to start being able to find similar cases.
70 years after the peak of assembly, that value probably comes more from supply than demand.
Keeping the comparison to compilers, people in the 60s refusing the transition weren’t crushed, their skills were devalued. The open models work well enough at coding to keep up without buying OpenAI or Anthropic products.
Rust is used a lot. Elixir/Phoenix comes up often. You’re looking for something statically typed with string guarantees and as FP as possible to make unit testing easier.
AI is not as big of a productivity boost as compilers were but it is not zero. Back then, SQL was created and marketed as something “managers could use”, another parallel to the current cycle. This whole hype cycle is something we’ve seen before and IMO, we should know better how to handle it this time around. It’s a bit disappointing to hear the same arguments every time.
We’re sharing a world with cultures that do not respect or enforce patents. The internet has helped us stay competitive by making information a lot cheaper to share. AI is the same thing but more potent. It will help us stay competitive even longer.
There were some hidden downsides to the internet and they are worse with AI. Upsides are better, and downsides are worse. Still largely the same deal we already accepted though. “Information wants to be free” never said information wasn’t harmful.
The original creators of Mario Kart are still involved in the latest entries (a 34-year run!) Thats not common outside Nintendo.