Hacker Newsnew | past | comments | ask | show | jobs | submit | 127's commentslogin

Job and slavery are two very different things. I love free markets too but not to the level of autism (for lack of better word) and ignoring how things actually work in the physical world. Shallow economic theory doesn't create happy and functional families. Economic theory that is mostly rigged to serve the rich, entrenched, and powerful anyways.

> Job and slavery are two very different things.

Having to work to survive is not slavery.

Slavery is a really bad thing. Equating it to having to get a job is really gross and downplays the severity of actual slavery.


Is there alternative syntax? I don't want to be typing "end" everywhere. Stack based, declarative, sexp, indentation, anything?

Overtone [0] is a precursor to Sonic Pi that uses Clojure. Sam used to work on it before he created Sonic Pi. I remember having great fun playing with it at a Clojure meetup in about 2011. Overtone appears to still be going if you want to check it out.

[0] https://github.com/overtone/overtone


There are alternative frameworks that also drive supercollider:

https://tidalcycles.org is built in Haskell

https://strudel.cc is a Javascript REPL


I am working on live coding WebDAW, written with ClojureScript (so a Lisp dialect): https://github.com/tmythicator/tritoncha

Live: https://tmythicator.github.io/tritoncha/

Right now it is based on mix of Three.js (for visuals) and Tone.js (for audio). I playing with the idea to rewrite the synth/sequencing part on WASM. But the live music scripting language will remain to be ClojureScript. As I see the Lisp language family to be much superior in expressiveness. Less syntax rules, more code as data


There are many languages like sonic pi now, search for "livecoding". The most prominent currently is strudel

And you're claiming Anthropic and OpenAI are not doing the same thing?


For the price of can of Coke, you can do a week of work. For most, that is not a bottleneck.


The whole point of turning intelligence into a commodity is to drive its price down, not up

They are hoarding HW at massive scale, they make it harder and more expensive to own

Just because you are fine with the new price doesn't mean it's not a problem

Perhaps it's time to pop this bubble


DSV4 Flash is in a tier of its own, until at least the price change arrives.


DSV4 Flash 0731 already runs on RTX 4090 24GB + 128GB system RAM at a usable tok/s and quantization.


You personally? Just curious. Context window is also a factor and ram isn’t really cheap. Sparks are assembled units which I like.


For the same price as a DGX Spark here (A$8499) I can buy roughly 544GB of DDR5-5200MHz from retail; which on a quad channel platform would deliver ~160gb/s real world; and ~320gb/s with octa channels (Xeon, Threadripper Pro).

If you can afford it or somehow find a used unit, you can go Epyc for 12 channels.

8/12 channel DDR5 will beat DGX Spark in inference/decode even without a GPU of any kind, as it’s memory bandwidth bound, and the Spark tops out at ~240gb/s real world.

With some optimisation and maths, it’s entirely plausible to ach

You are paying an extraordinary amount of money for the convenience of a super small unit, with still mediocre software support, but at least a community. Expect to be crawling through forum posts regularly, as SM121/Spark has many quirks and ecosystem issues still.

Please don’t pay another 70-80% gross margins on top of already inflated DRAM prices unless you need. The Spark IS really nice if you want to test out ConnectX or if you really need something small and compact and quiet.

Also consider: used Adas or even Ampere NVIDIA workstation GPUs can come with a lot of VRAM and be “reasonable”, with CUDA.


Been investigating these multichannel AMD based platforms last year and seem like none of them can in real scenarios utilize anywhere close to their theoretical bandwidth.


I've run it with a large context window on 256GB ram + 4090. It wasn't super fast, but it was manageable and it completed the tasks I gave it well.


How many tps and size of ctx?


IoT works just fine with Zigbee that can be 100% local.


Most of which you don't need IMO. When a bunch of stuff had to be redone after a kitchen fire, I told my electrician that I wanted everything to be dumb.


Did IoT cause the fire?


No. A 15-year old (so not ancient) dumb microwave.

Not a luddite but, aside from a couple of Alexas, just no need for any IoT.


How did you start a fire with a microwave? Plain old electrical fire, rats eating wire insulation?


Apparently a short circuit in the control panel in the middle of the night. At least that's what the investigator said. I wasn't using it and was asleep. So I didn't start it. Was spontaneous. Didn't seem to be related to any external wiring.


We had a dishwasher let out the smoke recently. No fire, luckily, just literal smoke and melted plastic. The melted part was the mechanical power switch. Yes, dead simple non-IOT holy century old technology. The parts that fail are almost always the moving parts and almost never the smarts.


My washing machine was intermittently staining stuff for a while. The it started smoking one day. Should have replaced it at least a couple years prior. Bet it was a good 50 years old.


Didn't know this existed. Have to say that's some great looking art.


Similar market correction happened in 2008 and the the US has definitely not gone more free-market-let-the-bad-businesses-fail direction since then. Rich a-holes always find ways to build systemic risk where they can reap the rewards and socialize the losses.


OpenAI and Anthropic valuation crashing would crash the entire AI story, which would crash the market.


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

Search: