lambdaone

The numbers are used as labels or indicators, not for their numerical values, so I think the title is still correct.

AV2's mission is to nip VVC in the bud. They seem to be more or less at parity, and given that, why would anyone want to use a royalty-based codec when they could could use an essentially equivalent free one? There's no massive hurry to implement either - we have existing codecs that are largely good enough for now - this is technology which take 5 to 10 years to fully deploy, as has been seen with other codecs.

This makes the hair rise up on the back of my neck; it reminds me the sub-prime crisis - "they can't all default at once!"

This is really impressive work. I'm not sure how this gets fixed, but it needs to be fixed, and soon.

...or taking them to somewhere where you are legitimately going to use them for a legal purpose, or taking them back from there, or, indeed, doing anything else with them for which there is a legal justification. Provided the knife is securely packaged (in a toolbox, for example) and not available for use as a weapon, it's all good.

Note that all this refers to big knives, not pocket knives with short blades, which are still completely legal to carry on you without needing any justification.

Seems very reasonable to me. As a result, contrary to the myths, and in spite of guns not being available for attacks, we actually have a fairly low rate of actual knife murder; slightly less than that of the US, I believe.

We never see Kyoko speak in the film, which is explained away by Nathan as being because she does not speak English. The implication is that she is an earlier model than Ava, with less intelligence. There is certainly sexism and orientalism in this, but it's Nathan's, and the film is explicitly critiquing it, not endorsing it.

I think this ship has already sailed, with a lot of comments on social media already being AI-generated and posted by bots. Things are only going to get worse as time goes on.

I think the next battleground is going to be over steering the opinions and advice generatd by LLMs and other models by poisoning the training set.

This brings back memories of Adult Swim's "Smart Pipe" spoof infomercial.

You could use a similar argument to argue that "brains do not think, understand, reason, reflect, comprehend and they never shall." After all, there's nothing in there but neurons, synapses and other biological gunk. If you look at it that way.

Which you shouldn't.

What you are trying to minimize here is the error rate of the composite system, not the error rate of the individual modules. You take it as a given that all the teams are doing their human best to eliminate mistakes from their design. The idea of this is to make it likely that the mistakes that remain are different mistakes from those made by the other teams.

Providing errors are independent, it's better to have three subsystems with 99% reliability in a voting arrangement than one system with 99.9% reliability.

It's baffling from our perspective, but perhaps not so much if you try to look at it the mindset of its proponents.

It's been sold as "for the children". A very substantial proportion of the population are natural authoritarians, and this is red meat for them. Never mind that "the children" that they profess to be protecting are going to grow up living in an increasingly authoritarian surveillance state, this is what authoritarians want for our future, and they see it as not only morally good, but any opposition to it as indefensible.

The big problem is the Red Queen's Race nature of development in rapidly-evolving software ecosystems, where everyone has to keep pushing versions forward to deal with their dependencies' changes, as well as any actual software developments of their own. Combine that with the poor design decisions found in rapidly-evolving ecosystems, where everyone assumes anyting can be fixed in the next release, and you have a recipe for disaster.

I find this fascinating, as it raises the possibility of a single framework that can unify neural and symbolic computation by "defuzzing" activations into what are effectively symbols. Has anyone looked at the possibility of going the other way, by fuzzifying logical computation?

Unity is absolutely being squeezed between the two. I can't really see how it can compete with Godot at the low end; it's hard to compete with free, and most of the goodness in low end games is the gameplay logic, not graphics or animation. And Godot can only get better; look at how Blender ate the CGI tools market. This leaves Unity having to either compete with Unreal at the high end - a very high bar - or somehow finding a new business model. The switcheroo they tried to pull on their customer base can best be viewed in that light.

RL would reasonably be expected to work if the market had some sort of discoverable static behavior.

The reason why RL by backtesting cannot work is that the real market is continuously changing, as all the agents within it, both human and automated, are constantly updating their opinions and strategies.

Monophasic waveforms are generally considered less safe than biphasic waveforms. That's why many TENS units have an output stage based around a pulse transformer, so they can deliver two pulses, one in each direction, shortly after each other. Leaving this out seems to me to be a false economy when you've gone to all the effort of building the rest of the system.

What makes trading such a special case is that as you use new technology to increase the capability of your trading system, other market participants you are trading against will be doing the same; it's a never-ending arms race.

Even more than 20 years on, people are finding new ways to improve kernel internals. This is the sort of subtle, intelligent work that distinguishes excellent software from the rest.

There is absolutely nothing to prevent anyone from generating arbirary DOM content from XML using JS; indeed, there's nothing stopping them from creating a complete XSLT implementation. There's just no need to have it in the core of the browser.

I found the whole 'activation' model deeply confusing and annoying. The underlying mechanism of venv is simple and elegant, but the original interface is terrible piece of design.

A microgrid lets you power your neighbors, or receive power from them.

This paper cites several examples of peer-to-peer electricity trading: https://www.sciencedirect.com/science/article/pii/S030626191...

If you want the job done properly, this already exists: https://en.wikipedia.org/wiki/Decimal128_floating-point_form...

This really is a Black-Mirror-esque hellscape, and the fact that this is marketed as something to be desired is absolutely astonishing.

No, they're not electrically powered, but given that you can buy an electrically-powered 5-horn kit from eBay for $20, it's certainly the cheapest way to go for a typical HN-reading hobbyist/hacker. Getting 14 horns is then $60, and the rest is relays a microcontroller, and a bit of 3D printing for the horns.

Which is an interesting nerd-sniping exercise in itself: https://www.grc.com/acoustics/an-introduction-to-horn-theory...

There's also nothing stopping African communities from building peer-to-peer microgrids if they want to.

It turned out laws were only for little people, after all.

This is fantastic. The one-handed organ keyboard is a work of genius, and it all feels like very simple organic technology: sets of tuned car horns are really cheap, and the rest is just a matter of wiring.

I'm severely tempted to buy a bunch of car horn parts and build my own MIDI-controllable one - every bit of this I can either buy on Amazon or 3D print.

This is exactly what makes this such an interesting development. Standardization is part of the process of the CPU industry becoming a mature industry not dependent on the whims of individual companies. Boring, yes, but also stable.

Sadly, this is not true of a (admittedly very small) number of individuals.

RISC-V is well on its way to be a challenger to both, with (eventually) full support for Rust.