avidphantasm

Brilliant, 996 yourself out of a job.

This, and no one will understand the software that is created. Then you are beholden to AI companies who can charge you whatever they want to maintain the AI code. Will this be cheaper than paying software engineers? Maybe, but I could also see it costing much more.

For more advanced terrain analysis, see GRASS GIS (https://grass.osgeo.org/). It’s a but clunky, but accessible from QGIS via a plugin. Whitebox also has some nice terrain analysis tools (https://www.whiteboxgeo.com/).

You know that river was bad if they closed it in 1925.

Oh, absolutely.

This title is misleading. It makes it seem like the robots did this autonomously, when in reality hundreds if workers were involved. The “robots” were “smart jacks” I would say. Humans couldn’t have done this without hydraulic jacks, they used fancy hydraulic jacks.

And both missionaries and mercenaries are responsible for the abuse and obliteration of many millions. Neither are forces for good.

Windows NT had the ability do away with drive letters since maybe the beginning; I was certainly running NT 4 and Windows 2000 like this. The problem I think is that every legacy app would instantly break if they couldn’t use drive letters and Microsoft doesn’t care enough to prioritize migrating away from drive letters. This, I believe, is a great example of MS having no taste (as Steve Jobs pointed out).

I don’t care if the probability of failure is lower in a helicopter than automobile. The failure modes for helicopters are all worse than that of an automobile. Chances of dying if something goes wrong is much higher.

The only way I will ride in a chopper is if myself or an immediate family member is being flown to a hospital.

The "one IP per container" approach (instead of shared IPs) is similar to how kubernetes pods work.

Reminds me of when they added more transparency to the UI around Mac OS X 10.9 where they argued that it "helps you focus on what's important". Huh? By showing me what's behind what I'm trying to look at? The first thing I do when I setup a new machine is to go to accessibility settings and turn on "reduce transparency". Hoping there is a way to do something similar with this.

Yeah, I would imagine they maybe started with off-the-shelf tools that were then gradually replaced as the system grew and matured.

Yeah, I am constantly curious about how the sausage that is cloud services like AWS is made. It seems generally slick on the surface, but what’s holding it all together? I imagine it as a tangled ball of tools like Puppet, Chef, etc. and custom glue.

I just use pytest-docker-compose, then I don’t need to bother with NPM. I usually don’t like “magic”, but pytest’s fixtures are so powerful I’m okay with a little bit of “magic”.

Came here to say this. Well said. Also language skills and reading have a complicated relationship. My daughter was saying sentences like “no suppository for me!” when she was two (she had constipation issues). But is now seven and while she still has excellent verbal skills, reading is coming slowly for her, though she can read and is making progress, it’s just slower than her classmates. Her school district has high standards (students on average read at two years above grade level). However, I wonder if this push to have kids read at younger and younger ages is not appropriate and harmful at some point. We were warned by our OT upon entering kindergarten that the curriculum will have the kids do things that are not developmentally appropriate. Also, my daughter has a mid spring birthday, so she’s 6 months younger than many of her classmates, which at this age still matters I think.

I think the theory is that it’s okay if the bucket overflows, at least it’s full. However, I worry that pushing kids to do too much too early can make it hard for them to build confidence and enthusiasm for reading and learning in general.

Exactly. The power of MMT is that it credibly explains what taxes actually are (not a source of government funding, but control rods for inflation and a constant source of domestic demand for the currency that helps maintain monetary sovereignty) and how to make sure spending is productive and not inflationary (e.g. through the currency issuer being the employer of last resort, etc.). You might not like it, but if you want to have a vaguely capitalistic system that has long term stability, you need to employ mechanisms like this.

Thank you for elucidating the fact that US “borrows” in its own currency and so as long as we have monetary sovereignty (which we do to a great degree due to USD being the global reserve currency) we have no deficit problems (as long as we keep inflation in check).

I implore everyone reading this thread to read up on Modern Monetary Theory (MMT). Stephanie Kelton’s “The Deficit Myth” is a good place to start. Until more people have a better understanding for how the system works, they’ll continue to be confused and easily led astray by politicians who don’t have the best interests of most people in mind, but instead wish to rob us blind in a vain attempt to eliminate trade/budget deficits.

(Note: I’m very much for restoring domestic manufacturing as much as possible, including the use of highly targeted tariffs, industrial policy, R&D, workforce development, etc. but this is not what the current administration is doing, not even close).

HFTs gonna hook up to LHC and do femtosecond gold futures arb. plays.

Spotlight search on macOS is in general kinda…spotty. Now that we have super fast SSDs it should be instantaneous very reliable. How hard can this be? BeOS seem to have figured it out 30 years ago. Apple missed a chance to fix this once and for all when APFS was developed, but they are fat and happy, no fire in their guts. Craig Ferengi must go.

You cannot put LLMs in control loops.

It would be more impressive if the AI could make it as a real 2600 ROM, since the programming model is much different than what it was likely trained on, and requires more knowledge of how the hardware works (and is just much harder to do in general).

My lay understanding is that the US can’t have USD be the global reserve currency without running trade deficits. Countries need dollars because it’s what oil and other commodities are exchanged for. At some point, the only way to get dollars is by selling things to the US. If they turn around and buy an equal amount of things from the US, they wouldn’t have the dollars they need to buy things like oil.

I have been using Python for 20 years, and have been an intermediate to advanced user of it for last 5-7 years. I use it mostly for scientific computing (so lots of Numpy, SciPy, etc.), IoT data processing, and also for some microservices that don’t need to be super fast. I publish and maintain a few packages in PyPI and conda (though I almost never use conda myself), including a C++ library with Python bindings generated by SWIG (SWIG wouldn’t be my first choice, but I inherited it).

In what I’ve done, I’ve never found things like pipenv, let alone uv, to be necessary. Am I missing something? What would uv get?

Assuming we have anything close to reliable data coming out of BLS going forward.

There is a through line from Goldwater to Nixon to Reagan to Gingrich to Trump. Connective tissue has been provided by the likes of Roger Stone. This has been a 60-plus year project to overturn the New Deal.

For the millionth time, we don’t pay the government. The US federal government is not funded by taxes. The US government spends money into existence. It taxes money out of existence to help control the supply of money, inflation, etc. Stop using the deficit as a smokescreen for redistribution of wealth away from the 99% to the 1%.

I’ve found that self-tapping screws and an impact driver give me the same effect.

I was recently experimenting with local-only LLM coding assistant in JetBrains products. They did speed things up a bit, but I quickly realized that they were essentially automating the creating a copy-paste errors, resulting in time lost to debugging errors I never would have introduced myself, so I stopped using them.

The notion of no longer training radiologists because computer vision algorithms and deep learning are good at detecting cancers in imagery strikes me troublingly naïve. Who fucking trained the AIs? Will the AIs magically be able to detect yet to be discovered maladies?