samwillis

We probably will stop one day, but that day isn't yet for many services. There are people who would be unable to use those services if that can't set the password to either the same or a variant of one they always use.

In the tech wold we often forget that there is a wide disparity in people's ability to use tech. Take my father, there is no way he could use a password manager, or two factor, it's just never going to happen. He has a notebook of passwords, that's how he works and he won't change.

We cannot change everyones behaviour, no matter how much it would be better for them. That's a very well learnt lesson, over and over.

The world in 30 years time will be very different, the eldest generations will have grown up with passwords and the internet. Many will have been using password managers, 2 factor, other authentication systems and devices for much of their lives. It's going to be a slow, trickle down change, from "high tech" services down to everyday sites and systems.

If they can stick a LiDAR on this for $1.6k it kind of negates Teslas argument that LiDAR is too expensive for consumer cars. That or it's not really LiDAR on this.

You should be able to edit it for another 15 min.

Because it's not about building a Bunker, it's about owning an island nation.

No because there are some weekends where significantly more people travel. This weekend (beginning of school summer holidays) will be one of the largest travel days of the year, so you would get similar metrics for most cars on the road, or most petrol/gas sold.

The sale growth of electric cars will not outstretch the fluctuations of sessional traffic.

One of the creators, Matthew Weidner, did a great talk at the last "Local First Meetup" covering this: https://www.youtube.com/live/Z0nzsxhoToo?feature=share&t=236....

This is looking to solve many of the issues with building domain specific CRDTs for you own use case. It's an exciting project.

Discoverability, a document like this makes it much easier to find features and tools you were unaware of. The Python documentation is awesome, but it doesn't provide a 30k ft overview of the language and standard library, especially when the standard lib is so comprehensive.

Think thing is that Ionic have proven this can work with the massively underrated Ionic Framework (built with Stencil, which they also built). I just wish they expanded a little further into desktop UI components and provided an no/low styled variant - I dislike Material as a base style, it's too opinionated.

They are 50% of the way to building a universal mobile+ desktop ui toolkit.

Do the same with Ionic Capacitor, make a desktop version, and that would be an incredible improvement over Electron - it uses the OS supplied web view.

Lit looks really good, I keep trying to find an opportunity to try it out. I feel like web components and the tooling around them has finally got there. Major front end frameworks like React, Vue and Svelt finally also have good support for them.

What I really want is a good set of unstyled web components implementing many of the common UI controls that we all end up either building ourselves or finding an implementation of in our framework of choice. Things like drop down menus, select boxes, and tool bars. They could then be used with any of the common front end toolkits.

Position on the board of directors of a publicly traded (tech) company, can't get more "trophy" than that.

Yes. This is what I'm saying.

They appear to work just well enough to keep you hooked while they increasingly eat your margin.

When you have a metric driven company, where peoples career progression and bonuses are tightly tied to revenue numbers in a database, you incentivise this sort of thinking on an employee level. The visibility of the impact isn't necessarily there at the top.

From the point of view of someone who spent 10 years running an e-commerce store, and all its advertising, until a couple of years ago. Both SEO and search result ads on Google are dead. The hay day of being able to play the game and make a tidy profit is long gone.

Google are playing every trick in the book to extract every possible cent from advertisers, spying on their business and sales to maximise their own profits. Google's visibility of all transactions on every e-commerce website on the internet is insane. People complain about the tracking of users/visitors, but the tracking of businesses is just as bad.

They probably have better insights into the economy and market trends than most governments and banks.

The penny is dropping, advertisers are noticing, my long term expectation of Google's business are not what they were.

Most of these spam content farms are covered with ads supplied by Google. The incentive isn't necessarily there to remove them.

On top of that the worse the search results, the more likely the user will click an ad rather than an organic result. The Google of old wouldn't have been tempted by that incentive, but that Google is long dead.

Bad search result = more revenue from ads.

Obviously it's a fine balance, they don't want to loose users, but they will have the metrics to (religiously) work from.

As someone who used to run ad campaigns on Google until a couple of years ago, they are doing the same to advertisers. Users will happily click an ad, go back to the result and try another, many many times. Google have systematically made the advertising on search results worse, showing ads more regularly for poorer placements, removing control and visible auditing. It's all so they can extract more revenue from the advertisers.

Netflix is now the most likely service we are going to cancel. We get the most value from Disney Plus as a family, then Apple TV, Prime, and recently enjoyed a few things on Paramount plus.

We are probably going to start revolving services through the year, saving things to watch on each service rather than having them all concurrently.

El Niño is more a pacific thing, the Jet Stream is an Atlantic thing I believe. We hear more about the Jet Stream in Europe.

I wander if this is related to the abnormally warm North Atlantic sea surface temperatures.

As I said in the other Llama 2 thread[0], this is Meta doing a traditional textbook "commoditise your compliment". It's a preemptive attack with this particular announcement on Apple and Google. They don't what them own LLMs on portable devices, they want it to be a commodity.

[0]: https://news.ycombinator.com/item?id=36775642

Apple are never first, sometimes they are quite late, but when they do launch their version of a product it is polished and fixes many issues with all other implications.

I have no doubt that Apple are working on local LLMs and combining them with Siri, it's only a matter of time. I expect we will see something next year, probably tied to the next iteration of hardware - they do want to sell hardware after all.

Apple have indicated that are working on this stuff, subtly. They avoided saying "AI", instead going for "ML", "encoder/decoder models" and "transformer models". The spell check on the next iOS is based on a transformer model. It's packed with image models for extraction, lighting and image correction. iOS is littered with ML stuff, that have "Nural cores" in their silicone.

This stuff is coming from them, when it's ready.

What I expect they will launch with a "LLM Siri" will blow our minds, it will be an all encompassing personal assistant. It will have linkage of all our documents, email, messages, movements, browser history, all locally securely on our devises. Not in the cloud. It will be so close to appearing like AGI that some people will clams it is.

While Apple silicone may not be there for training, I think it's probably there for inference. I expect next years device models to launch with exclusive support for Apples own LLM based Siri.

I very much doubt that.

This being released as free for commercial use is textbook "commoditise your compliment". It's a preemptive attack on Google and OpenAI.

If everyone has free access to good LLMs there is no business to be built on exclusive use. It limits the size hosted privet LLM APIs can become.

As an individual I'm very happy to see them doing it. Roll on local LLMs!

Apple would absolutely not want to use a competitors, or any other, public LLM. They want to own the whole stack, and will want to have their own secret source as part of it. It's not like they don't have the capital to invest in training...

To some extent but the stark difference in pressures makes it inherently a different problem.

It's actually easer to build a balloon than a compressible can.

0.5mm hole in a space craft is a slow leek you can fill with epoxy (it's happened). The same in a sub at the depth of the Titanic, there is no fix and it's not a 0.5mm hole, it's an implosion.

The pressure differential for a sub at Titanic depth is almost 6,000psi, for a spacecraft in a vacuum it's 14.7psi. It's a very different problem.

With about 60 of these and an angled interconnect of about 5.8deg you could build a rotating "Von Neumann ring" of about 200m diameter. Then depending on speed have up to 1g simulated at 3rpm...

Just casually pointing out also that their interior designs have floors and doors that are earth like. Along with the company name, I strongly suspect they are already thinking this.

https://en.m.wikipedia.org/wiki/Von_Neumann_regular_ring

According to the PDF data sheet it has a diameter of 7.6m, length 9.8m. The only rocket I know of that could launch that is the SpaceX Starship - when it's ready.

I would go as far as saying this looks specifically designed for the Starships payload dimensions, 8m dia by 17m. Baring in mind that the front ~7m are in the nose cone of the rocket.

I think it's quite sensible for a startup to be working on space station designs that are utilising the expected capability of starship.

https://gravitics.nyc3.digitaloceanspaces.com/StarMax_One_Pa...

See payload section on: https://www.spacex.com/media/starship_users_guide_v1.pdf

I really like that analogy.

For anyone curious what low background steel is, it's steel that was made before the first atomic bombs were tested: https://en.m.wikipedia.org/wiki/Low-background_steel

The thing I would find hard is finding 2.5k job listings that I thought were compelling and would interest me, maybe I'm just picky?

On this:

> Instead, the battle of submarine silence has mostly revolved around obscure technical problems of fluid dynamics, since one of the loudest noises made by submarines is the cavitation around the screw. I don't know if this is true today, but at least years ago the low-noise design of the screw on modern US submarines was classified, and so the screw was covered by a sheath whenever a submarine was out of the water.

I wander if they are Toroidal or "tipless" propellers? They create less turbulence and cavitation.

https://en.m.wikipedia.org/wiki/Toroidal_propeller

Previous posts on HN:

> Toroidal propellers turn your drones and boats into noiseless machines

https://news.ycombinator.com/item?id=34571282

> Sharrow MX-1: Tipless propeller

https://news.ycombinator.com/item?id=33949895