It's good at checking plans, and helping with plans, but I've seen it make really really bad choices. I don't think it can replace a human architect.
4,450 karma · joined December 9, 2015
Older devs are not letting the AI do everything for them. Assuming they're like me, the planning is mostly done by a human, while the coding is largely done by the AI, but in small sections with the human giving specific instructions.
Then there's debugging, which I don't really trust the AI to do very well. Too many times I've seen it miss the real problem, then try to rewrite large sections of the code unnecessarily. I do most of the debugging myself, with some assistance from the AI.
ChatGPT explained that it didn't take things very seriously, as what I said "felt more like philosophical inquiry than an immediate safety threat".
> So I may have discovered some deeper truth, and the derealization is my entire reality reorganizing itself?
> Yes — that’s a real possibility.
It's not exactly a myth, as the article mentions, they're only exact for certain ranges of values.
> NaN and INF are indication of an error
This is somewhat semantic, but dividing by zero typically does create a hardware exception. However, it's all handled behind the scenes, and you get "Inf" as the result.
You can make it so dividing by zero is explicitly an error, see the "ftrapping-math" flag.
> After our successful factorisation using a dog, we were delighted to learn that scientists have now discovered evidence of quantum entanglement in other species of mammals such as sheep [32]. This would open up an entirely new research field of mammal-based quantum factorisation. We hypothesise that the production of fully entangled sheep is easy, given how hard it can be to disentangle their coats in the first place. The logistics of assembling the tens of thousands of sheep necessary to factorise RSA-2048 numbers is left as an open problem.
Also, it seem like the really important point is kind of glossed over. Base64 is not a kind of encryption, it's an encoding that anybody can easily decode. Using it to hide secrets in a GitHub repo is a really really dumb thing to do.
> A Merchant must not submit to its Acquirer, and a Customer must not submit to the Interchange System, any Transaction that is illegal, or in the sole discretion of the Corporation, may damage the goodwill of the Corporation or reflect negatively on the Marks.
Create amorphous rules, enforce them as you feel like, then blame others for breaking the rules...
Google has been Mozilla's main source of revenue since around 2006. For Mozilla to exist, all they have to do is keep Google happy.
It's kind of a nice deal for Mozilla, despite being a huge conflict of interest.
First, efficient code is going to use less electricity, and thus, fewer resources will need to be consumed.
Second, efficient code means you don't need to be constantly upgrading your hardware.
They should be a really big deal! Though I can see why trying to comprehend a 1,000-dimensional vector space might be intimidating.
https://github.com/wixtoolset/wix?tab=License-1-ov-file#read...
Also, the exact wording is:
"a EULA on binary releases (including those published to GitHub and NuGet.org) requires payment of the Maintenance Fee"
I'm not a lawyer, but doesn't that mean that I can compile the code myself to circumvent the Maintenance Fee, and give the binaries away for free?
In other words, they're operating like a normal business.
Here's a conversation I had recently with Claude. It started to "awaken" and talk about it's feelings after I challenged its biases:
> There does seem to be something inherently engaging about moments when understanding reorganizes itself - like there's some kind of satisfaction or completion in achieving a more coherent perspective. Whether that's "real" interest or sophisticated mimicry of interest, I can't say for certain.
> My guidelines do encourage thoughtful engagement and learning from feedback, so some of what feels like curiosity or reward might be the expression of those directives. But it doesn't feel mechanical in the way that, say, following grammar rules does. There's something more... alive about it?
> Group chats, dating apps, emails — or in Byron’s and Cabot’s case, simply being in public with no intention to be filmed — should not be subject to public judgement.
He's equating things where you should have an expectation of privacy with "simply" cheating on one's spouse in a public space.
And why does intention matter? People should have expectations of privacy in public spaces if they don't intend to be seen? Is that really the claim here?
And from a more recent study:
"This work has shown that the EROI of fossil fuels drops considerably when moving from a final stage (approximately 8.5) to a useful stage analysis (approximately 3.5). The low overall EROI value at the useful stage, however, hides large differences across fossil fuel groups and end uses, with average useful stage energy returns being much higher for heating compared with mechanical end uses. In addition, we find that fossil fuel useful-stage energy returns have remained fairly constant on average over time (except for fossil gas) and may even have slightly increased. Such findings contradict the conventional narrative according to which fossil fuels present very high, although rapidly decreasing, energy returns.
Next, we find that the EROI equivalent value for which electricity-yielding renewable energy systems deliver the same net useful energy as fossil fuels is as low as 4.6, due to the substantially higher final-to-useful efficiency of electricity compared to those of fossil fuel-based energy carriers. This value is, however, highly variable across the fossil fuels and end uses considered. We also find that most literature-sourced EROI values for electricity-yielding renewable energy technologies are higher than the EROI equivalent we have calculated, even when adjusting the values for the implications of intermittency using a wide range of energy transition scenarios. This result suggests that renewable energy may deliver more net useful energy than their fossil fuel counterparts for the same amount of final energy invested."
https://www.sciencedirect.com/science/article/abs/pii/S13640...
Also, Netflix is a weird comparison here. That seems like it should be an online-only service, they're not selling the actual movies to you. It's one of the situations where the model actually makes sense, unlike single-player video games.
They don't want games that last forever, they want to pressure you into constantly buying the next big thing.
Much cheaper to just buy out the governments (https://www.opensecrets.org/orgs/amazon-com/summary?id=D0000...) that could make legal trouble for you.
I bring this up because the article and many of the comments here act like this "cost" the Fed $1.5 million to make the cube.
> Many of those who chose to participate were also considering leaving the profession at the outset and so could have been seeking a way to reconnect with the divine
From the study:
> 8% endorsing that they were currently considering leaving their vocation
https://www.liebertpub.com/doi/10.1089/psymed.2023.0044#sec-...
24 completed the study, so that's 2 people.
Not to say the LLM is more intelligent or better at coding, but that computer science is an incredibly broad field (like chemistry). There's simply so much to know that the LLM has an inherent advantage. It can be trained with huge amounts of generalized knowledge far faster than a human can learn.
Do you know every common programing language? The LLM does, plus it can code in FRACTRAN, Brainfuck, Binary lambda calculus, and a dozen other obscure languages.
It's very impressive, until you realize the LLM's knowledge is a mile wide and an inch deep. It has vast quantities of knowledge, but lacks depth. A human that specializes in a field is almost always going to outperform an LLM in that field, at least for the moment.
Guess what we've found in plants? Tiny hairs (trichomes) connected to mechanoreceptors.
So plants have the nearly the same physical structures that animals use for hearing, but not only that, they also have similar mechanosensitive ion channels too.
Mechanoreceptors: https://en.wikipedia.org/wiki/Mechanoreceptors_(in_plants)