lambdaone

Absolutely! Can this probability be quantified?

This is actually an extremely interesting question. 'Weak' files that are more easily compressable than others certainly exist, but with low probability.

For example, the all-zeros file is a member of the set of all random 3 megabyte files, and it would certainly be possible to compress that, if by great good fortune you were lucky enough to receive it - albeit something that is unlikely to ever happen in the possible lifetime of the universe, given current physical theories.

Is it possible to quantify this idea of a 'weak' file more accurately?

You're really going to hate it when you learn about JSON Web Tokens, which exist exactly to hack past this sort of problem.

Let's Encrypt is a massive achievement, and is now essential infrastructure.

Basing it on an open protocol, so it doesn't become a single point of failure, was a clever idea that allows the idea to survive the demise of any single organization.

May there be many more such anniversaries.

Have you considered the possibility that the ad is deliberately intended to stir up controversy, putting the brand in people's minds and priming people to pay more attention when they actually launch the new car range?

I think the idea here is that somebody else runs an underlying IP layer, and this rides on top as an overlay network.

You could, of course, run a local wireless IP layer and use this to route, but peer-to-peer wireless has well-known scaling problems.

Still, it looks like a very interesting and reasonably well thoughout out idea.

Ladybird is indeed very interesting. But it's too soon to tell whether it will be successful, and how, if it is, it will get the money to maintain its development in the long term.

The real value of Firefox is to have a serious browser vendor that is not wholly owned by a FAANG company, and it still to some extent focused on the user's needs rather than its parent company's agenda. Having a third browser engine which is neither WebKit or Blink is a win too.

Implementation diversity is absolutely essential to an open-standards based Web, and right now the Web is pretty much the only open channel for pretty much all of human discourse.

Having said which, Mozilla is in real trouble, and desparately need to sort out their business model.

It really, really struggles to move the pieces. More development is needed.

Hmm. I got ChatGPT-4o to write some code for me today. The results, while very impressive looking, simply didn't work. By the time I'd finished debugging it, I probably spent 80% of the time I would have spent writing it from scratch.

None of which is to discount the furture potential of LLMs, or the amazing ability they have right now - I've solved other simpler problems almost entirely with LLMs. But they are not a panacea.

Yet.

He's part right. There's certainly a law of diminishing returns in terms of model size, compute time, dataset size etc. if all that is to be done is to do the same as we are currently doing, only more so.

But what Marcus seems to be assuming is the impossibility of any fundamental theoretical improvements in the field. I see the reverse; the insights being gained from brute-force models have resulted in a lot of promising research.

Transformers are not the be-all and end-all of models, nor are current training methods the best that can ever be achieved. Discounting any possibility of further theoretical developments seems a bold position to take.

Evolution wasn't trying to do anything. But learning (in some sense) to walk was exactly something that evolution did, albeit very indirectly.

I'm surprised that the authors seemingly can't see that their argument hoists them by their own petard by showing, using exactly the same reasoning, that human intelligence is equally impossible to create (unless, perhaps, you are willing to use vitalist/supernatural arguments to claim is is different in kind from computational AI).

Surely the referees must have raised this at the review stage?

This is astoundingly good work; even though, as other commenters have said, it could still be improved on, the fact that it achieves such a level of realism with such a simple and elegant framework is truly impressive.

Wait until someone uses "AI printing" for medical purposes and it adjusts a dose value.

You might want to try the UK national archives:

https://www.nationalarchives.gov.uk/

or get in touch with the enthusiast community, people like

https://www.nathanmediaservices.co.uk/ceefax/

You should ensure whoever you give them to can give you a binding commitment to use for legitimate archival purposes, and not just acquire them and sit on them for copyright reasons.

OpenSCAD feels awesome to start with, but is ultimately a disappointing experience because it lacks parametric solver capacity; data flow is one way, from the script, to shape generation, to CSG operations, and you cannot do quite reasonable things like extracting dimensions from a CSG-constructed shape and feeding them back into the script.

A better system would enable true constraint-based parametric design, but I think doing so would involve an order of magnitude more complexity in the implementation, and quite probably infeasable compute times.

My hope is still that common humanity and decency could be shown by him to save the life of this unfortunate man. I didn't say I thought it was likely.

The Easter Island story of ecocide through decadence has become part of the accepted truth of modern culture. Now it looks like it may never happened, and that the moral to be drawn is not one of human frailty and vanity, but one of human resilience.

Scandalous. It's now up to the governor to act to stop this happening; let's hope common humanity prevails over politics.

I would have thought the major problem after a technological collapse would be the production of computer hardware, even up to the level of something like ENIAC, as the technology stack needed to create that represents centuries of technological development.

Software development, on the other hand, took merely decades, and would be much faster to reboot, as computer language and compiler technology can be easily reinvented given information in printed books. If anything, I'd suggest an emphasis on having paper copies of books on classic programming languages and computer history widely distributed in paper libraries, and making sure that paper libraries are not displaced entirely by electronic ones.

I'd consider archiving full documentation of pre-existing basic technology, rather than rolling your own; a widely distributed complete set of printed books, circuit diagrams, manuals, source code and assembly code listings capable of reproducing something like BCPL on PDP-11 might be a better choice; creating something like a PDP-11 class computer with valves or relays would be eminently feasible.

You have it exactly backwards. The original goal of Linux was to create a Unix-like operating system on Linus Torvald's own Intel 80386 PC. Once the original Linux had been created, it was then ported to other CPUs. The joy of a portable operating system is that you can run it on any CPU, including Intel CPUs.

Enshittification has finally reached peripherals. I thought this story was a joke at first, but it evidently isn't. I can't imagine what extra services a mouse might offer, except in the role of acting as a dongle for other software. Presumably Logitech will enlighten us before launching this product.

The Perun video is really very good.

Adding autonomy makes drones potentially a much more serious asymmetric weapon, as jamming command and control stops being an effective countermeasure.

Missiles, bombs and artillery are one thing, but personalized silent assassination out of nowhere, at any time, without a traceable attacker you can pin down or counterattack, is quite a different threat.

Alas, doing this with quadcopters and winged drones is really not difficult to do, for anyone with a reasonable computer science education - or come to that, a particularly bright teenager somewhere in the world. NVIDIA even market their Jetson boards as being suitable for "smart drones": https://developer.nvidia.com/buy-jetson

Of course, there's a difference between a DIY demo and a fieldable weapon, but it's one of productization, not the Manhattan Project.

I'm quite surprised that at least one of the participants in the various current wars haven't done this already. They've certainly got sufficient incentive to do it.

I'm not as worried about the Black Mirror robot dogs, current robot dogs seem far less practical to weaponize compared to suicide/surveillance drones, which are already cheap and being used as weapons already sans the autonomy.

They've been this fragile for some considerable time. We've just been lucky, over and over again, that nothing of this seriousness has happened before.

Absolutely. Global variables are always going to need some sort of arbitration mechanism, and this is doing it in the innermost loop. Even if there's only one thread, something needs to find that out. I'm surprised it's only half the speed.

It was inevitable that something like this would happen sooner or later, either through security complacency or sheer bad luck. We were just fortunate it only happened to Windows.

What gets me is that much of the OSS/Linux ecosystem consists of thousandas of lashed together piles of code written by independent and only very loosely coordinated groups, much of it code and lashed together by amateurs for free, and it is still more robust than software created by multi-billion dollar corporations.

Perhaps one reason is that OSS system programmers are washing their dirty linen in public; not a matter of "many eyes make bugs shallow", but that "any eyes make bad code embarassing".

Just for example, I'm planning to make one of my commercial projects open source, and I am going to have to do a lot of fixing up before I'm willing to show the source code in public. It's not terrible code, and it works perfectly well, but it's not the sort of code I'd be willing to show to the world in general. Better documentation, TODO and FIXME fixing, checking comments still reflect the code, etc. etc.

But for all my sense of shame for this (perfectly good and working) software, I've seen the insides of several closed-source commercial code bases and seen far, far worse. I would imagine most "enterprise" software is written to a similar standard.

Zig is amazing. What started as a one-man project is becoming a mature software ecosystem capable of standing on its own as a peer with the major languages.