bubaumba

7 ms·

31.
by bubaumba 2y ago
> But if there was an obvious path forward Even obvious can be risky. First it's nice to share the risk, second more investments come with more connections. As for LLMs boom. I think finally we'll realize that LLM with algorith
32.
by bubaumba 2y ago
> The annoying part, a lot of money could be funneled into these unworkable businesses in the process, crypto being a good example There was a thread here about why ycombinator invests into several competing startups. The answer is succe
33.
by bubaumba 2y ago
No, as long as it works they don't care. I.e. even if Amazon stops development but continues to support existing infrastructure most customers will stay. It's expensive and may take many months to move back on premises or another
34.
by bubaumba 2y ago
Unlike Google Amazon was never known for its great culture. "decay" looks to be a global trend when managers finally "win" and start optimizing for short sited money making to please investors right now. Usually this ha
35.
by bubaumba 2y ago
Similar case from soviet era: On 4 July 1989, a pilotless MiG-23 jet fighter of the Soviet Air Forces crashed into a house in Bellegem, near Kortrijk, Belgium, killing one person. The pilot had ejected over an hour earlier near Kołobrzeg, P
36.
by bubaumba 2y ago
Don't expect much. Even screwing a nut on a bolt is a huge problem for generic robot. Which means you will need a 'robot friendly' lab. Were all things can be done by a primitive robotic hand. Other options are making more ca
37.
by bubaumba 2y ago
Sounds idiotic if you ask me.
38.
by bubaumba 2y ago
Musk should by them to accelerate Tesla's robots development. Then we may have them on Mars by 2030
39.
by bubaumba 2y ago
half of the population has IQ below average. what 'idea' are you talking about? Actually that's getting interesting. For non-physical jobs we already got significant boost from LLMs. Robots will be another wave when they get
40.
by bubaumba 2y ago
well, well, I'm not the best driver. almost hit one idiot in the dark. take it as a warning..
41.
by bubaumba 2y ago
yeh, but the main problem is the quality. with algorithm bug can be fixed. with llm it's more complicated. in practice they do some mistakes consistently, and in some cases cannot recover even with assistance. (don't take me wrong
42.
by bubaumba 2y ago
I learned touch typing long after the beginning of my pro developer career. That made a huge difference. The main benefit is it make typing easy. Which is important when writing docs and comments. And with that I can handle much more becaus
43.
by bubaumba 2y ago
How about GUI and API are just interfaces, the rest is the same. They can coexist in the same model or setup. Same functionality can be implemented usually by one or another. In advanced products by both. But model's thinking is probab
44.
by bubaumba 2y ago
great minds think alike. remote consulting looks within the reach now.
45.
by bubaumba 2y ago
you can't expect all at once. just one step forward. note how fast everything moves since 2020, and accelerating. finally 'it's' coming...
46.
by bubaumba 2y ago
not only AI, we have one 'guru' who sounds like he is reading copilot on remote audio only meetings.
47.
by bubaumba 2y ago
I'm thinking the right way of doing it is to embed in a viewer. So that the original stays untouched but visible quality gets better with new technologies.
48.
by bubaumba 2y ago
sure, how about chess? bad news, even champions show no exceptional skills outside of chess world. the same goes for tetris players. in other word specialized training does not result in general abilities by default. I'm not saying is
49.
by bubaumba 2y ago
I haven't seen any research on this matter. How modern kids compare to their (grand) parents. Likely with free food for everyone and legal drugs more are choosing jobless path. One way is sort of a job. For example 'influencer&#x2
50.
by bubaumba 2y ago
sure, just don't forget, that will be the kid who chooses the shelter...
51.
by bubaumba 2y ago
Don't worry, 20 year back most people had no clue how to use the second core. They believed in GHz and when it became obvious the theoretical limit is close thought that's the end of evolution.
52.
by bubaumba 2y ago
NVidia should be afraid. Same GB/s on CPU are much more than on GPU. The problem here will be the price of the whole system. Anyway it's a step in the right direction if we want something like Copilot on premises or at home.
53.
by bubaumba 2y ago
>> I just do not see why people seem to use "unsafe" so much >Because it’s impossible to implement any non-trivial data structures in safe Rust. Even Vec has unsafe code Hmm.. wasn't memory safety the main selling po
54.
by bubaumba 2y ago
They shouldn't die, just become robotic with optional remote control. Comparing to copters helis have significant advantages and disadvantages too. I.e. different enough to have different roles
55.
by bubaumba 2y ago
That's how it always was. People were eating only the best parts of the animal and dumping the rest. More over, the best is converted mostly to sh*t and dumped too. Fish do the same.
56.
by bubaumba 2y ago
> I think the most damning thing about this whole saga for all of AI is how much energy and attention people are giving it. That's because there is nothing better today, and nothing like it in the history.
57.
by bubaumba 2y ago
> How often do people expect to compile open-source code and get _exactly_ the same binary _Always_, with the right options. And that's the key point. If distributed code is different it means it may be infected or altered in other
58.
by bubaumba 2y ago
You are missing key points here. "reproduce" means produce the same. Not just train similar model. I can simplify the task, can you convincingly explain how the same model can be produced from this dataset? We can start simple, ho
59.
by bubaumba 2y ago
No, it's not open source till someone can actually reproduce it. That's the hardest part. For now it's open weights open dataset. Which is not the same.
60.
by bubaumba 2y ago
They are not that fragile if you don't drop hard. Never used case for A50, it still looks almost new after years of active use. But for iphone I got a thin case, to protect popping out cameras and to make it look cool. Main concern is

More ›