News for hackers. Stuff that matters.

New_California

Re:How does Turbo listen for Turbo Streams on Sunday October 19, 2025 @02:45AM (#45632442)
Exactly that.
Re:O3 may output your PI into the source code on Wednesday April 16, 2025 @05:31PM (#43710614)
Even w/ memory feature OFF. It's easy to miss PI leaked into piles of HTML/CSS/JS code.

Ask HN: Light ML Laptop for Linux 2025?

Brief: looking for a well balanced laptop well running Linux and easily capable of local inference for small LLM-s, image generation, etc. It will be used standalone so quality display and keyboard are important.The perfect laptop (I do not expect to find one): * Light and quite 14"-15" * Gaming laptop midrange+ (because that's best for AI) * Great Linux support * NVIDIA GeForce w/ ~8GB VRAM *...

( Read More... | 3 comments | on HN )

Ask HN: Python Meta-Client for OpenAI, Anthropic, Gemini and other LLM API-s?

I am looking for a Python library that generalizes over popular LLM API-s without sacrificing full control (need to use tools, control temp, seed, top-p etc).Simon Willison llm seems close but seems to focus much on the CLI and not the library, also doesn't seem to give precise control for non-OpenAI models (might be wrong here).

( Read More... | 2 comments | on HN )
Re:Tech Independence on Sunday September 17, 2023 @04:10PM (#37548846)
But there is: https://umbrel.com/ (except for hosting email which is not realistic anymore).
Re:Building GitHub with Ruby on Rails on Friday April 07, 2023 @03:36AM (#35479140)
They keep saying "monolith".
Re:Building GitHub with Ruby on Rails on Friday April 07, 2023 @03:36AM (#35479135)
On the contrary, it is so much easier to upgrade a small app every week or so because you have little to test and probability of breaking changes affecting you is minimal.

You should be upgrading all the time since day one, adding necessary infrastructure gradually as your app grows.

Re:EU countries approve 2035 phaseout of CO2-emitting cars on Tuesday March 28, 2023 @09:37AM (#35340837)
Not saying it necessarily will fall apart but there is a possibility. UK is already out. Hungary and Poland are very nasty EU members for lack of a better term. The tension is growing.

Personally I would much prefer EU to *reset* to bare free-trade, free-movement, no-borders ethos, removing entirety of central regulations and central governance.

Re:EU countries approve 2035 phaseout of CO2-emitting cars on Tuesday March 28, 2023 @08:32AM (#35339822)
Anything else individual to be replaced by communal? Like, maybe, everything? Or just the cars? There are countries already implementing what you believe is right. Please emigrate over there and enjoy.
Re:EU countries approve 2035 phaseout of CO2-emitting cars on Tuesday March 28, 2023 @08:28AM (#35339764)
The governments will certainly exempt themselves the very first moment they get affected by the ban.
Re:EU countries approve 2035 phaseout of CO2-emitting cars on Tuesday March 28, 2023 @08:24AM (#35339692)
But that would be a regress in civilization, not progress! The civilization progress is to have much more private transport and also somewhat more public transport.
Re:EU countries approve 2035 phaseout of CO2-emitting cars on Tuesday March 28, 2023 @08:22AM (#35339657)
Poland and Hungary were also strictly against the ban on selling gasoline cars. So no, it's not only Germany.

Whether EU itself will be still a thing in 12+ years to enforce this bullshit ban is another question.

Ask HN: Curated list of top AI models per function?

I'm looking for a daily updated, opinionated catalog of best AI models, segregated per function. Both SaaS and open models. I mean functions like speech-to-text (Whisper etc), conversational (GPT-4 and open alternatives), text-to-speech, image generation, etc.I prefer it to only list a tiny number of top-performing models per category. I do not want a bag of hundreds of models to research...

( Read More... | no comments | on HN )
Re:Just use Postgres for everything on Saturday December 10, 2022 @12:45PM (#33934705)
> Just use the right tool for the job at hand. Period.

That's Postgres.

Re:Tor Browser 12.0 on Wednesday December 07, 2022 @04:36PM (#33900633)
I struggle with the very same problem. Even connecting a real physical SIM won't protect accounts from suspension. This is crazy because it isn't based on behavior on the platform (like posting too much or liking too much) but merely on the IP and browser fingerprint.

Unfortunately, no solutions in sight.

Re:Bitcoin Is Digital Scarcity on Monday October 03, 2022 @08:18AM (#33066428)
Because the 21m hard cap and change resistance are the definition of Bitcoin.

If these are taken away for any reason, the result would be indistinguishable from shitcoins.

Re:Bitcoin Is Digital Scarcity on Monday October 03, 2022 @08:13AM (#33066388)
> Whther it's likely or not is beside the point. It's possible in theory, therefore (and this is my main point here) the digital scarcity of bitcoin is not some emergent mathematical property of the universe, it's a social phenomenon.

Correct.

> Was it? Did you get to vote on that? Or was it decided by a handful of pools and large players?

Yes, I got to vote on that with the economic full nodes I ran.

Re:Bitcoin Is Digital Scarcity on Monday October 03, 2022 @08:10AM (#33066348)
Fractional reserve is already a thing and will only increase from here. Bitcoin's scarcity relates to the base money supply and obviously cannot coerce the higher economic layers.

That said bitcoiners are somewhat resistant to extreme fractional reserve because of the strong "not your keys, not your coins" ethos.

Re:Bitcoin Is Digital Scarcity on Monday October 03, 2022 @08:05AM (#33066302)
It's much more than exchanges and miners as exemplified by the 2017 block size war, won by the users.

The scenario you are describing - while possible in principle - is unlikely. Stakeholders understand that scarcity and change-resistance is Bitcoin's very definition. Take that away and there is no value left, Bitcoin becoming another malleable shitcoin.

You are incorrect. The miner does not intermediate transactions in any way, shape or form.

Firstly, miners do not technically intermediate transactions. Transactions are fixed and sealed by the parties involved and can't be modified. Transferred funds are never in a possession of a miner, not even briefly. Miner can't take them, can't redirect them, can't change the amount, etc.

Secondly, it's not a specific miner that confirms the transaction. It is the ever-growing group of miners. Transaction finality is ever only probabilistic. If you think picking by law the "first" miner then this is completely arbitrary and 1-block chain tips are routinely orphaned in all cryptocurrencies.

Finally, many persons and businesses accept unconfirmed (not mined) transactions, for relatively low-risk goods and services.

If you really want to regulate miners (which is making a crime running certain algorithms on your personal computer - a terrible idea), then at least please come up with a new name and new framework, because "money transmitter" or "financial intermediary" is simply incorrect.

Re:Manhattan rents cross $5k threshold for first time on Thursday July 14, 2022 @12:15PM (#32097976)
I already had communism and state-owned housing in my country, so no thanks! Never again. You are welcome to move to a state-managed economies though.
You nailed it.
Re:Pen and paper exercises in machine learning (2021) on Wednesday June 29, 2022 @05:27AM (#31918013)
No, this is not a hands-on introduction to ML.
> Do you think this cycle the whole crypto space goes down the drain never to be heard of again?

No. The crypto/scam space will be back because people love to gamble and love the concept of easy money.

> Do you think the whole space is a fad or do you think there are patches of values underneath the froth?

There is crypto/scam space and there is Bitcoin/Monero space. The actual decentralized cryptocurrencies are widely used where fiat system fails, like on DNMs.

No, company board and owners are public and readily available data in virtually all countries, US being an exception to the rule.
No, we can't. Bitcoin *is* anonymous against average Joe which is what the OP needs here.

Pretty irrelevant here but for your fucking information, with significant expert effort, Bitcoin *can* be also anonymous against big actors.

Older »