mg

Not sure.

I would certainly try out Linux on one of the new 13" iPad Pros with anti-glare display. With an external keyboard, it could make for a great mobile development machine.

And many kids would certainly have dual booted their iPhone into Android when Fortnight stopped working on iOS.

Just two use cases. Who knows how many there are.

I think a simpler and more logical law would have been to disallow the sale of products which are locked down beyond their owners wishes. It's a net negative for society when we add additional production steps to decrease the usefulness of the things we produce.

So any computer (phone, tablet, laptop, desktop, watch) would have to allow installing whatever software the user wants. On the hardware level.

That would mean alternative OSes like Linux could run on any computer.

It would still be allowed to ask or even warn the user "Do you REALLY want to install an alternative OS?" but not to ultimately deny it.

I wish there was a way to sort these by date.

The terminal is such a wonderful environment. It makes many tasks joyful.

I think it stems from its simplicity and logic.

You have a simple grid of characters. You type the name of a program, hit enter, and the computer executes it. The output of the program is written below your prompt. Below the output, you can execute the next program. You can chain output and input of programs, allowing all kinds of data flows. The programs can also take over the whole grid and thereby allow all kinds of interfaces.

I can't think of a way to describe a window manager or a browser in similar, simple terms, which would explain the basis of all the usefulness they bring.

I like this approach you do with UsefulChecklists - building very generic software projects and then applying them to different topics. So my feature request is a way to follow your journey with this project. An RSS, ActivityPub, Nostr, Bluesky or Twitter feed.

I would expect that they have access to the cameras of all Teslas on the road, which should be about 5 million:

https://www.statista.com/statistics/502208/tesla-quarterly-v...

And I would expect that those generate more data than they can crunch, so that the bottleneck is the number of GPUs they have.

They recently reported that they have the equivalent of 40k H100 GPUs:

https://twitter.com/marekgibney/status/1784133742983860307

Unless there is a change in how autonomous driving software is trained, it seems Tesla is destined to dominate that market, right?

Tesla has the most training data consisting of video input + corresponding driver behavior. And they have the most compute power.

Which twist of fate could happen, so that Tesla falls behind in the future?

I could imagine that there is an invention which lets AI learn about the world by just observing and thinking. Without the need for "training" in the form of "if you see this, do this". But as far as I know, this is not on the horizon yet.

Awesome, added:

https://www.gnod.com/search/ai?q=Python%3A%20How%20do%20I%20...

Have you thought about accepting the query as a get request?

The 3 engines you mention (Perplexity, You.com and Phind) all do that. So do Google, Bing and DuckDuckGo. It makes it easier to link to results and build custom links.

Also, I could add you to Gnod Search then:

https://www.gnod.com/search/ai

We see the prisoners dilemma in action.

What would be a typical coding question which AI would not be able to answer in the near future without having access to Stack Overflow?

I find it hard to imagine that AI will need humans to teach it technologies like programming languages and APIs for long.

We don't need humans to teach computers how to play chess anymore.

Would it be possible to base a device like the R1 on a Raspberry Pi and Linux instead of Android?

If not, what would be missing?

Currently, LLMs start from scratch for each output token, right?

Lets say you ask an LLM

    What makes bananas yellow?

And it replies

    Bananas are yellow due to a pigment called bromelain.

I would think that the concept of "pigment" and "bromelain" are already somehow activated in the neural net when it outputs "a". Because now it can't change its mind anymore and follow up with "an optical illusion that makes humans perceive every bent object as yellow". So it seems to have already planned ahead to talk about the pigment called bromelain.

Would it be possible to capitalize on the work that has already been done when the LLM outputs "a"? Could the state of the neural net be somehow preserved for the next answer?

That could be a good idea.

How would you like to use it? To show a set of monitors to a friend, or to bookmark it for yourself to see how the selection for a certain set of parameters evolves over time?

For example, should the monitors which you have hidden with the "hide" button also be hidden when you use the bookmarked URL? Should the ones you marked with "compare" still be marked?

Great, added!

It's a bit tricky to add captions. It depends on the number of items on the page and the size of the screen, whether there is enough space.

A submit form might be a good idea, true.

Yes, I should add a ratio filter.

Probably a dropdown with the different options, like the Panel Type.

That will need a bit of tinkering, as the ratio is not a readily available parameter, but need to be calculated on the fly and then put into the right bins like 3:2, 16:9, 16:10 etc.

Nice one! Added it now.

So a 38" monitor with 3840x2160 pixels for less than $500 or so?

Hmm.. yes, I'm indeed not aware of one.

At 42.5", there is the ViewSonic VX4381. But I don't know one at 38".

Happy to hear that.

What would make a monitor a perfect match that the FV43U and the XG43UQ do not offer?

I maintain this comparison chart of currently available large monitors:

https://www.productchart.com/monitors/large

Maybe it helps. If a specific large monitor is missing, let me know.

The prompt I always try first:

    Two cars have a 100 mile race. Car A drives 10
    miles per hour. Car B drives 5 miles per hour,
    but gets a 10 hour headstart. Who wins?

Tried it on Qwen1.5-110B three times. It got it wrong 2 times and correct 1 time.

Europe could put quite some weight behind open social media if they would put official accounts like x.com/bundeskanzler on bundestag.de/@bundeskanzler instead.

Similar to https://www.threads.net/@potus but truly open by tying it to a government controlled domain.

I think ActivityPub supports putting a "look over there for the actual content" file on a domain, so that technically all you have to do is put a static file on the domain. And then you can use an external service to publish your content.

So is "id" stored as a property or as an attribute?

The gist of it seems to be that attributes sometimes behave like properties.

Setting 3 attributes on an html element:

    <input id=good x=morning value=hn>

Then logging 3 properties with the same names:

    <script>
        e = document.querySelector('input');
        console.log(e.id); 
        console.log(e.x);
        console.log(e.value);
    </script>

Will output

    good
    undefined
    hn

I'm not sure why. The article says it's because "Element has an id getter & setter that 'reflects' the id attribute". Which sounds logical. But MDN does not mention a getter function for "value" for example:

https://developer.mozilla.org/en-US/docs/Web/HTML/Element/in...

It even calls "value" a property AND an attribute: "...access the respective HTMLInputElement object's value property. The value attribute is always..."

And HTMLInputElement lists "value" under "properties":

https://developer.mozilla.org/en-US/docs/Web/API/HTMLInputEl...

So maybe it's not as easy. Maybe <element some=thing> sometimes sets a property and sometimes sets an attribute.

If one wants to do a "Masonry" layout, what's wrong with calculating it via JavaScript?

What makes houses so expensive?

Is it the land or the house itself?

I live in a city of 2 million in Europe. With a car, it takes 20 minutes from the center to places where absolutely nobody lives.

If it is cheap to build a house - why not do it there?

If it is expensive to build a house - could it be optimized, 3d-printed, automated in some way?

I wish there was something like a second hand store where you can just drop your stuff for 10% of its price on eBay.

I don't want to deal with anything. Not wait for someone to pick it up. Not wrap it in a parcel and bring it to the post office. Not sign up for some service. Nothing. Just drop my stuff, get a little bit of money and am done with it.

I think there is a pretty secure approach of doing it.

1) When loading a file, nothing needs to be different from the already available upload functionality. The user picks the file to load.

2) When saving a new file, nothing needs to be different from the already available download functionality. The user picks the name and the place to save it.

Only after 1 or 2 have already happened can the page request to update the file. With the same name and location. In this case, the user gets a prompt that explains the risks and asks the them to confirm the update.

The one browser API I wish every browser would support is the File System Access API.

Because with it, we can offer users to hold their data natively on their devices. Instead of storing everything in the cloud.

So far, I think only Chrome on the desktop supports it. Here is a nice demo:

https://googlechromelabs.github.io/text-editor/

A text editor that works just like a native application.