@skyblond Faster?
@skyblond *We don't know what we are doing but we didn't consult the nerds.*
"2023 showcased the power and perils of #OpenSource"
https://techcrunch.com/2023/12/26/the-eternal-struggle-between-open-source-and-proprietary-software/
"Software Developers must continually Learn in order to keep up with the fast-paced changes in the field"
#Linux D-Bus overview
@tasket @caligari @lupyuen It's not difficult to follow a signal. Ransomware attacks aren't going to be far behind as some of those systems use multiple wireless sensors.
They are difficult to open without using a Dremel. I've cut myself a few times with Xacto blades trying to get through that thick sonic welded plastic shell. At least the app wasn't too difficult to peek at. White listing devices and some other things that might possibly be going on.
There has to be a way to get in without bleeding everywhere. ![]()
@skyblond That's an interesting result.
Tried the "Dolphin 2.5 Mixtral 8X7B" LLM. It's an uncensored model, so it can tell you how to cook meth or make a nuclear bomb. But I don't have enough knowledge to let it tell me how to do it precisely.
Anyway, I noticed the LLM have different performance on different languages. Multi lang support is fairly normal for LLM nowadays, but English is significantly better than expensive languages like Chinese and Japanese. Not only on speed (one token can be one English word, or one Chinese character), but also the result. The model is fairly shy and doesn't spill out a lot in Chinese. If you force it to generate more details, it's just repeat the same thing. However in English, it gives a more detailed answer by default.
@skyblond So the performance is inadequate? Those two examples should be possible for college students.
Some might say that it would be harmful to society to have such capabilities. I remember learning about the designs of fission weapons in elementary school. The fission fusion chain was a bit more advanced. The knowledge is beneficial and the chance of someone achieving anything more than a fizzle is incredibly low.
One shouldn't need the instructions as better quality instructions and actual science are located in a library.
With that out of the way, can it make malware?
@freemo If you could find a way to give workers diplomatic immunity, they could be retained indefinitely.
Make sure to feed your phone sound from UHF band, and readings of latin.
@thendrix The DSP capabilities of modern phones should be capable of overcoming even that. It's absurd how much power these devices have and how many sensors are integrated.
Good find.
@lupyuen
Some people might think that Go is a useless language. It works really well for malware.
"in view of its ability to infect MIPS and ARM systems, it also poses a threat to #IoT devices"
@lupyuen It's beautiful to see China doing well. Through the USSR, Russia and US restrictions on the export of knowledge, China has succeeded.
The Article points out Alibaba and SiFive as manufacturers of RISC-V while it would be T-Head that makes a physical product. SiFive is fabless which means they hoard any information and only take from the community. T-Head posts Verilog of their creations for the world to see and use. SiFive doesn't deserve to be on the same list.
RISC-V was indirectly and likely unwittingly helped by DARPA. It's odd that the US is so anti freedom that it forced RISC-V out of the country. So perhaps a better summary is that the US was incompetent and China did something smart. That move was apparently not allowed because the US government allows corporations to own thoughts and ideas.
As an American I find it humorous but I'm still waiting for the punchline of the CCP messing it up like the RISC-X talks. Leave it to governments to take credit for something they had no part in and make it political. At least the surveillance systems in China will be less expensive.
Hopefully the inexpensive SoCs will be functional and have accurate documentation. RISC-V has always had challenges with all three and I suspect that the first few generations are going to be ewaste. That should sum up my commentary on the article. The hints of propaganda were appreciated.
Getting started with #LLM in Bash
"over 300 companies in China developing products using #RISCV"
https://jamestown.org/program/examining-chinas-grand-strategy-for-risc-v/
Toughbook fan, Mathematician and Locksmith with limited success in other areas.
Political stance is far right and far left. Proponent of First Aid Kits and PPE. Easily disheartened by big tech. Partially hinged personality and stubborn enough to not write this in the First Person.
Distrust of Psychology and a fan of satire. I love a good joke and contradict myself. Somewhat serious but easily distracted.