agarren

That is a fantastic presentation. It reminds me of the annotated source/site that backbonejs used to have, but this one is interactive. Really great job!

Elderly gen-x are screen addicts because they had access to gaming consoles and computers? Maybe I’m misunderstanding your comment. Not every gen-xer was strapped to a console growing up.

My boomer parents and their friends are all staring to their phones wa-a-ay more often than I’d consider healthy. At least as much as my millenial/xennial friends.

Social media and attention stealing algos are addictive and unhealthy, regardless of the age group. If anything I’d say that gen-x is uniquely positioned - old enough to have experienced the world without the internet, young enough to see the consequences of it.

but from time to time, doing this does require doing arithmetic correctly (to correctly add two exponents or whatever). so it would be nice to be able to trust that.

It sounds weird, but try writing your problem in LaTeX - I don’t know why, I’ve found a couple models to be incredibly capable at solving mathematical problems if you write them in LaTeX.

I’d add that the potential to support other architectures is also a benefit. At the moment the framework 13 chassis supports risc-v [0] [1] with rumors about an arm variant.

Beyond practical repairability and sustainability, I appreciate the possibility of swapping out a mainboard for another with a completely different arch

[0] https://frame.work/products/deep-computing-risc-v-mainboard

[1] https://store.deepcomputing.io/products/dc-roma-risc-v-mainb...

That jibes with what Nadella said in an interview not too long ago. Essentially, SaaS apps disappear entirely as LLMs interface directly with the underlying data store. The unspoken implication being that software as we understand it goes away as people interface with LLMs directly rather than ~~computers~~ software at all.

I kind of expect that from someone heading a company that appears to have sold-the-farm in an AI gamble. It’s interesting to see a similar viewpoint here (all biases considered)

The criticisms in the cnn article are already out date in many instances.

Which ones, specifically? I’m genuinely curious. The ones about “[an] unfalsifiable disease-free utopia”? The one from a labor economist basically equating Amodei’s high-unemployment/strong economy claims to pure fantasy? The fact that nothing Amodei said was cited or is substantiated in any meaningful way? Maybe the one where she points out that Amodei is fundamentally a sales guy, and that Anthropic is making the rounds saying scary stuff just after they released a new model - a techbro marketing push?

I like anthropic. They make a great product. Shame about their CEO - just another techbro pumping his scheme.

Ditto - no spam at all. I do get reminders that it’s time to file, and conveniently, if nothing has changed since the previous year (address, employer, marital status, etc.) then they can pre-populate the form.

I’ve used them for probably the better part of a decade. I’ve never had any issues, they’re pretty inexpensive, they e-file by default and you get a notification when state and fed are accepted. The wizard is easy to follow and explanations are available. They have bells and whistles like audit protection, that seem reasonable but that I’ve largely avoided.

The name might seem a little sketchy, but it’s a legit service. I don’t know about “free” - I’ve always ended up paying to file at least one of state or federal, I don’t recall, but not pricy.

(Edit - looking closer, I’ve been using them since 2007. Wow, time flies. Great service, no complaints.)

That doesn’t seem very likely from my reading. It sounds like it’s not unheard of for prion-diseases like Creutzfeldt-Jakob to take a while to destroy a person. The poor lady fta apparently had a genetic component that contributed to the latency. Brutal. Inducing dormancy in this and similar cases would effectively present a treatment, if not a cure, to an hereto untreatable, uncurable disease.

… the woman first visited doctors with tremors and trouble balancing. As is often the case, once symptoms started, her condition rapidly worsened. She was hospitalized … and several days into her stay, she fell into a coma that she would never awaken from.

In other news, apparently hgh used to be extracted from cadavers. Wtlf.

Out of curiosity, what kind of writing? Journaling and getting your thoughts out, or fiction, something else?

A “complex” cfg is still a cfg, and, giving credence to Chomsky’s hierarchy, remains computationally less complex than natural, context sensitive, grammars. Even a complex cfg can be parsed by a relatively simple program in ways that context-sensitive grammars cannot.

My understanding is that context sensitive grammars _can_ allow for recursive structures that are beyond cfgs, which is precisely why they sit below csgs in terms of computational complexity.

I don’t agree or disagree that LLMs might be, or are, capable of parsing (i.e., perception in Chomsky’s terms, or, arguably, “understanding” in any sense). But that they can learn the grammar of a “complex cfg” isn’t a convincing argument for the reasons you indicate.

[Harper Reed] cautioned against being overly precious about the value of deeply understanding one’s code, which is no longer necessary to ensure that it works.

That just strikes me as an odd thing to say. I’m convinced that this is the dividing line between today’s software engineers and tomorrow’s AI engineers (in whatever form that takes - prompt, vibe, etc.) Reed’s statement feels very much like a justification of “if it compiles, ship it!”

“It would be crazy if in an auto factory people were measuring to make sure every angle is correct,” he said, since machines now do the work. “It’s not as important as when it was group of ten people pounding out the metal.”

Except that the machines doing that work aren’t regularly hallucinating angles, spurious welding joints, etc.

Why? I can understand the argument that you don’t want an ISP or a middlebox injecting ads or scripts (valid I think even if I’ve never encountered it to my knowledge), but otherwise you’re publishing content intended for the world. There’s presumably nothing especially sensitive that you need to hide on the wire.

I’m seeing a lot about price hikes, new service charges, and bad customer service since the acquisition. I noticed the price hike with one of my domains through them, I paid it but was close to reconsidering. Thinking I’ll try transferring it now.

https://techrights.org/n/2023/12/14/_Video_Lessons_to_be_Lea...

https://blog.cogitactive.com/website/gandi-outrageous-price/

How is it winner-take-all? Training models is expensive, yeah, but big companies are willing to do it, and if deepseek is any lesson, it’s not as expensive as we used to think. You might be building a castle in someone else’s kingdom, but you’ve got a few kingdom’s to choose from. None have a moat and none seem to be going away any time soon.

It’s cheaper to move from one model to another than it is to train a general purpose model yourself (to say nothing of domain-specific smaller models or anything open source.)

I’m not sure about problems turning into black boxes, LLMs are pretty explicit in my experience when producing a solution (good or bad.) _How_ they came about that solution _is_ a black box, but that’s not a new problem.

Temperature? Definitely tweaks the results…but I don’t know if “deterministic” is a term you can use in any way, shape, or form, in the context of LLMs.

https://www.ibm.com/think/topics/llm-temperature

It feels disingenuous but even companies I know and respect are adding questionable “agent” features that rarely work.

Word. I’m not a huge Microsoft fan, but it feels like they’re just shoving ChatGPT/copilot down your throat every chance they get. It’s integrated into everything - it’s useful when it’s useful, okay, but it generally isn’t. It’s one more Microsoft’ism that you have to learn to tolerate or simply ignore.

I can’t tell if Nadella’s really betting the farm on it all, or if he’s just trying to leave his mark.

Ditto! I haven’t found any hardware that’s daily-driver ready, but I keep looking.

https://store.deepcomputing.io/products/dc-roma-ai-pc-risc-v...

I especially like the idea of getting a framework version in this case I want to swap in a different mainboard. By their own admission, the risc-v board is targeting developers and not ready for prime time. Also coming from the US, not sure how the tariff thing will workout…

That sounds a bit unfair. All that code that we neither wrote nor understood, I think in the case of Rust, it’s either the borrow checker or the compiler itself doing something it does best - i.e., “defining memory safety away”. If that’s the case, then labeling such tooling and language-enforced memory safety mechanisms as “a bunch of additional code…you didn’t write and…don’t understand” appears somewhat inaccurate, no?

Monopolies can only be created by government dictate, like old AT&T or modern cable companies.

By virtually every definition I can find, a monopoly is a an entity that functions as the sole, or effectively the primary, provider of a good or service in some market. That seems to perfectly describe Google’s position wrt web-based advertising. Do other ad-platforms exist? Absolutely. Do they exhibit the kind of market dominance or control that Google does? Nowhere close.

Google has no power to stop them.

Fact? I’d argue that Google’s sheer size and dominance means they don’t need to stop them. Potential competitors simply don’t stand a chance given Google’s size, number of resources, and reach. Explain how that’s not a significant factor into Google “power to stop” a potential rival?

AST had raccoons on one or two of his books, earlier editions of “ Operating Systems Design and Implementation” iirc. I always thought cartoon raccoons on the cover of a technical text book seemed like an odd choice. I wonder if there’s a connection?

I really enjoy Crafting Interpreters, but I’ve revisited it a couple times and haven’t yet made it through half the book. It’s very well written and there’s a lot of material to chew on. That said, Writing an Interpreter in Go, is short, sweet, and the Pratt Parser implementation and explanation are great! I haven’t cracked open his compiler book yet, but the interpreter book was a breeze.

He got Jeff’d. Or maybe Bill’d (or Satya’d). Regardless, any kind of non-gnu open-source collaboration with the Powers That Be in the tech world is just begging for punishment. Amazon will just blatantly rebrand something, but Microsoft seems especially comfortable wasting a dev’s time mining for details before stealing or copying their work entirely.

Boo Microsoft. Winget still sucks.

I read recently that Microsoft is adopting rust more and more. I think that’s a step in the right direction for an OS with such a huge marketshare. That said, I’m just waiting for Rust.Net or Managed Rust to get excreted in a thinly veiled attempt to split the community, steal mindshare, and take over the project.

My favorite that I’ve heard a couple times is “solve math” and/or “solve physics”

Altman’s claimed LLMs will figure out climate change. Solid stuff.

I think we have a pretty good idea that we are not stochastic parrots - sophisticated or not. Anyone suggesting that we’re running billion parameter models in order to bang out a snarky comment is probably trying to sell you something (and crypto’s likely involved.)

I think you’re right, LLMs have demonstrated that relatively sophisticated mathematics involving billions of params and an internet full of training data is capable of some truly, truly, remarkable things. But as Penrose is saying, there are provable limits to computation. If we’re going to assume that intelligence as we experience it is computable, then Gödel’s theorem (and, frankly, the field of mathematics) seems to present a problem.

Gödel’s incompleteness theorem, and, say, the halting problem seem to fall squarely into the bucket of “basic computability theory” in precisely the way that “we think, that is a fact”, does not (D.A. hat tip)

You’re arguing that we know artificial reasoning exists because we are capable of reasoning. This presupposes that reasoning is computable and that we ourselves reason by computation. But that’s exactly what Penrose is saying isn’t the case - you’re saying we’re walking Turing machines, we’re intelligent, so we must be able to effectively create copies of that intelligence. Penrose is saying that intelligence is poorly defined, that it requires consciousness which is poorly understood, and that we are not meat-based computers.

Your last question misses the point completely. “If we are something else, then out CoT won’t be computable…” it’s like you’re almost there but you can’t let go of “we are meat-machines, everything boils down to computation, we can cook up clones”. Except, “basic computability theory” says that’s not even wrong.

It was (is) an ABI, more stable than something available through a C/C++ compiler. It’s been years since I’ve lived in a Windows world, but as I understand it, it’s still very much alive and well. It didn’t fail because it was complicated, although it was, but it continued to evolve and was eventually superseded by .net. I worked with win32 & mfc quite a bit and had to interface with Com/activex a little, usually through wrappers of some kind, but it was always under the covers. I also wouldn’t say that it failed because not enough apps did it. I can’t speak to drivers or system components, but if you were distributing windows apps prior to and early into .net, you dealt with com directly or indirectly.

I doubt he’d think much of it. I’m about 2/3rds through it. It’s all more of the same and not terribly convincing.

Fr. Andrew Dalton is a Sindonologist(?), somebody who studies the shroud, but not necessarily from a scientific perspective. Sindonology isn’t exactly embraced by Shermer in his article, or, apparently, by the scientific community in general. I hadn’t encountered the term before, but at a glance it seems to have all the scientific rigor of ufology or cryptozoology.

What I’ve heard so far in the video isn’t new evidence, but enthusiastic re-hashing of previous topics that have already been discussed to death. The video offers precisely the kinds of suspicious arguments that Shermer talks about. Exactly, in the case of C14 dating, “French invisible weaves”, and corner-holding-contamination.

I hadn’t encountered the AZ 1 & 2 C14 differences though and I’d be curious to get more details. He only briefly mentions them and doesn’t provide anything further. Fr. Dalton seems to point people to shroud.com which he calls “the most scientific of websites’ - that statement seems dubious at best, and browsing the site doesn’t change my impression.

I do appreciate that towards the end of the video that Fr. Dalton completely acknowledges that there is no evidence dating the shroud to the 1st century. All the other discussion around the “historical facts” surrounding JC, such as the kind of crown we supposedly wore, the kind of cross he supposedly carried, etc. is very eye-brow raising. The discussion of Eucharistic miracles and the prevalence of the AB blood type is similarly…interesting.

At the end of the day the authenticity of the shroud is a matter of faith. Evidence doesn’t seem like it should be important.

The Pray codex is listed under “fringe theories” in the Wikipedia Shroud entry. That certainly make it wrong, but it does make it a little more suspect than the c14 dating criticism. The c14 contamination arguments seem more than adequately addressed in the entire last 3rd of the article.

Fastmail allows for aliasing too - username@domain.tld -> bank@username.domain.tld, retailer@username.domain.tld, etc. Pretty convenient. I use that feature pretty often and I can only recall one instance which seemed to indicate my address was sold to spammers. It’s more useful for organizing incoming mail, like plus-aliasing in gmail.