petra

9 ms·

121.
by petra 6mo ago
Good perspective. Let's say Anthropic fails to pay it's debt, can Google take those TPU's back and make money from them?
122.
by petra 6mo ago
If someone feels aspartame makes them feel bad, they can try something else, and see how it makes them feel. That was the point behind my comment.
123.
by petra 6mo ago
For those who want diet cola without aspartame, there's an alternative: https://sodastream.com/products/diet-cola-4-pack
124.
by petra 6mo ago
Maybe machines alone, or with little help from human feedback can estimate possible brand damage.
125.
by petra 6mo ago
>> pretty much every manufacturer decided the trade offs are not worth the benefit. Isn't worth the benefit for who? the manufacturers? sure. Let's say a single manufacturer decides to offer some phones with a changeable ba
126.
by petra 6mo ago
Are there goals for an app design? can they be measured? specified? constrained? For example, in the world of e-commerce, one goal is improving conversion rate, as long as we get that and the design looks nice, that's OK.
127.
by petra 6mo ago
Even if chatbot LLM's stop at their current capability, There's a whole ecosystem of scientific language models(in drug discovery, chemistry, materials design, etc), and engineering language models(software, chip design, etc) that
128.
by petra 6mo ago
Nvidia has more money than God. Worst case they'll buy the competition.
129.
by petra 6mo ago
The question is whether robotics will look like a some number of platforms with little development to adapt to different scenarios, or a million types of machines that are highly fit for purpose. Because the first situation won't creat
130.
by petra 6mo ago
Is there a possibility to create a wireless point-to-point link ?
131.
by petra 7mo ago
This is interesting, for sure. Thinking about this, I think what happens if we take it to the extreme - a system that lets a single user enter his route to work - and compare how various transportation systems, including innovative ones, wo
132.
by petra 8mo ago
About integrated memory: Gemini says that between 85%-90% of PCs and laptops never upgrade their memory. On the other hand, multi-die integration can greatly increase memory bandwidth. So it's a reasonable move to integrate memory.
133.
by petra 8mo ago
It's probably possible to develop analog adsl chips in 1990 semi tech. But pretty difficult.
134.
by petra 9mo ago
As someone that is sensitive to displays, one of the best features of XFCE, unlike others desktops, is that it doesn't cause eye strain, probably because it doesn't play tricks - a pixel at a certain color is stable, and not dithe
135.
DARPA Backs Effort to Convert C to Rust(cs.wisc.edu)
2 points by petra 10mo ago|1 comments
136.
by petra 10mo ago
On that same note,of reducing the barriers for Linux usage. Would it be possible to create a Zorin OS USB drive that after inserting it into the USB drive of a laptop: The user would get a running Linux, with the UX they know(win 10/11
137.
by petra 11mo ago
There was a podcast with Mark Andreesen, the VC, and he said that Elon has deep understanding and involvement in the technical side in his companies.
138.
by petra 1y ago
I agree. The Arduino brand isn't for professionals. But let's say tomorrow they come together with bundle/partnerships to create a new, great dev environment, very easy, that a mechanical engineer can prototype a great robot
139.
by petra 1y ago
Zorin OS is very similar to windows, regular users just use it, without learning anything new. But still, marketing this must be brutal.
140.
by petra 1y ago
//And the ability to speak English natively is already in high demand throughout most the world, meaning if you ever get tired of online work and want some people time, you can have a job in like 5 minutes, particularly if you loo
141.
by petra 1y ago
Can an LLM detect a lack of precision and point it to you ?
142.
by petra 1y ago
A proper solution ? I have ideas, but their value is mostly in making me sound smart. In reality I don't have realistic, implementable ideas on how to get out of this mess. Just wishes.
143.
by petra 1y ago
There's a potential for 100x+ lower cost of chips/energy for inference with compute-in-memory technology. So they'll probably find a reasonable cost/value ratio.
144.
by petra 1y ago
Interesting link, thank you!. One other possible immune system link is the relationahip between the parasite toximoplasis gondii to schizophrenia. If I'm not mistaken that's the paper about that: https://dbc.wroc.pl&#x2
145.
by petra 1y ago
Intel/Altera still has this: https://www.intel.com/content/www/us/en/products/details/eas...
146.
by petra 1y ago
//If an application takes off, it is eventually cheaper and more performant to switch to ASICs, That's part of the FPGA business model - they have an automated way to take an FPGA design and turn it into a validated semi-cust
147.
by petra 1y ago
Z-Library has a keyword search. Personally i didn't find it too useful, especially given Google Books exists. It's not easy to create a quality book search engine.
148.
by petra 1y ago
That's very impressive, especially at low cost.
149.
by petra 1y ago
What was the kill rate of this system ?
150.
by petra 1y ago
The speed limits are because of the latency of the wires and the issue of heat

More ›