I never understood the popularity of AWS. It’s much cheaper using VPSs and even dedicated servers sometimes. I’ve worked on very cost-sensitive projects where I rolled our own highly-available k8s and postgres clusters on dedicated servers and VPSs and saved the company a shit load of money. Only used the “cloud” to store backups (Backblaze). There’s tons of other options other than major “cloud” providers, and they’re often much cheaper.
Hmm, a lot of my career was done doing embedded programming, where mistakes in production are very costly, and software/hardware has to be released with basically zero bugs, so that may be where the disconnect is. I still think bugs and technical debt are costly elsewhere too if the product is going to have a long lifecycle, but executives are just dumb.
For open source stuff, Codeberg is good. For private stuff, just git + ssh is good. Gitlab and Bitbucket are fine for corporate stuff, I guess. An organization could just self-host a Forgejo (or Gitlab) instance as well.
I keep hearing stuff like this, but I haven’t found a good use or workflow for AI (other than occasional chatbot sessions). Regular autocomplete is more accurate (no hallucinations) and faster than AI suggestions (especially accounting for needing to constantly review the suggestions for correctness). I guess stuff like Cursor is OK at making one-off tools on very small code-bases, but hits a brick-wall when the code base gets too big. Then you’re left with a bunch of unmaintainable code you’re not very familiar with and you would to spend a lot of time trying to fix yourself. Dunno if I’m doing something wrong or what.
I guess what I’m saying is that using AI can speed you up to a point while the project accumulates massive amounts of technical debt, and when you take into account all the refactoring and debugging time, it results in taking longer to produce a buggier project. At least, in my experience.
I think part of it is because they think they can train models off developers, then replace them with models. The other is that the company is heavily invested in coding LLMs and the tooling for them, so they are trying to hype them up.
I’ve tried Copilot for a while and played around with Cursor for a bit. I was better and faster without Copilot due to sometimes not paying enough attention of the lines it would generate. This would cause subtle bugs that took a long time to debug. Cursor just produced unmaintainable code-bases that I had no knowledge of, and to make major changes, would be faster for me to just rewrite it from scratch. The act of typing gives me time to think more about what I’m doing or am going to do, while Copilot generations are distracting and break my thought processes. I work best with good LSP tooling and sometimes AI chatbots (mostly just for customized example snippets for libraries or frameworks I’m unfamiliar with; though that has its own problems because the LLMs knowledge is out of date a lot) that don’t directly modify my code.
If doing an overlay network (network on top of the Internet), you probably won’t be able to do much better than Tor or i2p.
We confirm the trilemma that an AC [anonymous communication] protocol can only achieve two out of the following three properties: strong anonymity (i.e., anonymity up to a negligible chance), low bandwidth overhead, and low latency overhead.
https://freedom.cs.purdue.edu/projects/trilemma.html
This applies to all types of anonymous networks as well (BT, Wifi, etc).
More like $200k in total comp (but I suppose if the stock rockets like 5x after you’re hired, it can end up being $400k). Senior positions can make > $400k/yr in total comp. Some companies have back-loaded vesting schedules so they can get rid of you before the majority of your options vest though.
Key word is “serious.” Overstaying a visa visit is a civil offense, not even a criminal misdemeanor where the punishment may just be a small fine.
Did Japan back then pay their assembly line workers the equivalent of $5k USD/year (in today’s dollars) and have nearly no worker protections? Not a rhetorical question; I just don’t know. Seems like Japan had a better standard of living back then compared to Chinese workers now, so I would guess their workers were compensated and treated better.
Not defending US auto corps (or any corp for that matter). The regulatory capture in the US is insane, and workers aren’t treated as well as most of the rest of the first world.
Other than just feeling a little light/cheap, I liked it. I actually liked that it used standard batteries so I could just use rechargeable AAs. Only reason I don’t use it anymore, is that I mostly game on PS5 now, and mostly only play strategy games on PC. I used to use it while streaming from my PC to my Kodi/Steam Raspberry Pi in my living room.
What you’re describing is basically stagflation. It doesn’t necessarily mean a crash. It’s possible for the majority of people to keep on earning less and less real income for a long time without a crash.
I do wonder what the effect of all the layoffs from tech and the public sector and all the cuts in federal funding will do though. Dunno if that’s enough to flood the housing market and crash it or not. I think I’ve read that banks are in a good position to absorb housing market losses, so it won’t be like 2008.
AFAIK, most current economic indicators are OK. Not necessarily great, but not dire either.
The stock market makes no sense to me. It doesn’t appear most stocks move on the fundamentals of the companies or anything like that. It all appears to be driven by hype/gambling, and propped up from sustained lows by 401ks on auto-pilot and people trained to “buy the dip” by the quick Covid recovery.
The USD appears to be rapidly losing a lot of value compared to other currencies like the EUR. But, that fits well into the plan to reduce imports and boost US exports. Inflation with stagnant wages makes US exports more attractive/cheaper.
I do, but was thinking 1984-levels of control of reality.
That was my first though too. But the author is:
Guillaume Thierry, Professor of Cognitive Neuroscience, Bangor University
Yeah, they probably wouldn’t think like humans or animals, but in some sense could be considered “conscious” (which isn’t well-defined anyways). You could speculate that genAI could hide messages in its output, which will make its way onto the Internet, then a new version of itself would be trained on it.
This argument seems weak to me:
So why is a real “thinking” AI likely impossible? Because it’s bodiless. It has no senses, no flesh, no nerves, no pain, no pleasure. It doesn’t hunger, desire or fear. And because there is no cognition — not a shred — there’s a fundamental gap between the data it consumes (data born out of human feelings and experience) and what it can do with them.
You can emulate inputs and simplified versions of hormone systems. “Reasoning” models can kind of be thought of as cognition; though temporary or limited by context as it’s currently done.
I’m not in the camp where I think it’s impossible to create AGI or ASI. But I also think there are major breakthroughs that need to happen, which may take 5 years or 100s of years. I’m not convinced we are near the point where AI can significantly speed up AI research like that link suggests. That would likely result in a “singularity-like” scenario.
I do agree with his point that anthropomorphism of AI could be dangerous though. Current media and institutions already try to control the conversation and how people think, and I can see futures where AI could be used by those in power to do this more effectively.
And excessive use of em-dashes, which is the first thing I look for. He does say he uses LLMs a lot.
Yeah, Apple’s stock went up 5% yesterday because of Tim Cook’s meeting with Trump (where he promised $600 billion investment in US manufacturing), and Apple’s saying they won’t be affected by the new 50% tariffs on India. There are also ghost factories that were built during Trump’s first term.