90-00-09

That's a good guess! Checked eBay and iCloud locked M2 MacBooks Air go for around $400.

Just FYI: manifest v3 extensions are already cross-browser. They are all based on the WebExtensions API. However, browsers have very specific constraints that require complex alternative solutions (e.g. Safari does not allow reading httpOnly cookies, so auth can't be handled easily).

Oh I don't argue about art. Most of these decisions are driven by financial constraints, not the artistic vision. And this is not going to change.

Christopher Nolan can afford extreme artistic choices like not showing ads or previews before Oppenheimer. That's such a rarity that I wouldn't make any extrapolations even in today's world. Betting that gen AI will result in an alternative type of art and not just replace multiple roles and production techniques in existing industries is not something I can see happening considering that professional art is business.

E.g. we didn't see CGI creating an alternative movie industry that exists in parallel to non-CGI movies. CGI is used by everyone, including artsy independent creators.

I will forever love Studio Ghibli animation. But if in the future the same level of expression can be achieved with CGI/gen AI - I just fail to see them not switching their tools (possibly in the post-Miyazaki world).

> What's hilarious is that the execs still don't fucking see it

Why would they? Very few people can notice and appreciate the level of detail you are describing. And I am not even sure we will never be able to replicate it computationally. In 10-20... 50 years? Don't see why not.

Do they mention anywhere what DALL-E version is being used? I have a suspicion it's not DALL-E 3 but happy to be proven wrong.

Apparently Mozilla is working on that for over a year now (announced in Feb '22).

Ah cool. Yes, that makes sense!

Would you allow people browse the list of existing communities and join them, or are individual communities invite only?

Sounds good. One problem that you will likely encounter in FF is that you can't open the native extension popup (the one that is attached to the action button) programmatically. That severely limits how you can present your UI, with the most obvious choice being injecting an iframe with the extension UI directly into the body of the page. The other choice is opening a new popup window, which is less convenient than seeing things side by side.

Here is evidence: large crowds of extras have already been replaced by CGI. This is a list of 10 movies with the most number of extras: https://www.tasteofcinema.com/2017/10-movies-with-the-most-n.... The most recent entry on the list is from 2003 (~20K extras). Why would anyone hire 20K people for a movie today?

I really like the UI:

1. using the new Chrome Sidebar API is a great choice!

2. selecting actions to show in the context menu is also a convenient customization idea

Are you considering porting it to FF? Other than the Sidebar API it should be a very straightforward process.

This is an interesting take. A couple of observations:

- Small groups of people who know each other IRL are very sensitive to breakups. When a couple breaks up, it frequently splits their friend group into 2 camps that "follow" one of the partners (strange behavior in my view but certainly observable and prevalent, especially in the US). Dealing with this complexity is going to be a challenging task.

- Small groups of people rarely have many already connected people that would consider dating each other but for some reason don't. And expanding the group dating pool might be awkward as users will start inviting people from completely different circles with a potentially lower and lower overlap in interests, attitudes, etc.

- Dating websites are very often used for short term dating or hookups, which, I guess, is not going to be the focus of your app. Taking that into account, reaching a meaningful number of profiles in a circle might become an unsolvable blocker (i.e. having 4 friend profiles is not a much better experience than just introducing friends via socials/IRL)

It's too early to say who is winning/will win, of course. But so far the UI and its accessibility have made a huge difference in how different gen AI models are being used.

For example, I struggle to see DALL-E winning over Firefly if Firefly is integrated into a very rich environment, whereas DALL-E is basically prompt UI only (while DALL-E 3 is a better model IMO).

This is quite an exaggeration. Integrating DALL-E into MS Paint has basically no similarities to the vast gen AI feature set announced by Adobe.

- Firefly Vector model

- Generated templates

- Res-up upscaling for videos and gifs

- Fast-fill for videos

- Distractor removal

- Individual item replacement

Too many to mention. Not to say OpenAI/MS are not capable of building similar features. Just that Adobe's tools are incomparably more advanced than integrating a prompt into MS Paint.

Personally I think that’s the best selling point of passkeys. Most non-tech people don’t use password managers and have to memorize passwords, reset frequently passwords they can’t remember, etc. Security is way harder to sell than convenience.

Saying that, I am struggling to understand what is the expectation for ordinary user behavior in terms of hardware-tied credentials. Eg so many people upgrade their iPhone every 1-2 years. If passkeys are not transferred to the new phone, what is the industry suggesting people do?

Here is Google discussing how passkeys are easier and simpler to use than passwords: https://security.googleblog.com/2023/05/making-authenticatio...

Here is 1Password discussing how passkeys are simpler to use than passwords: https://1password.com/product/passkeys

Here is Ars Technica declaring that passkeys are easier to use than passwords: https://arstechnica.com/information-technology/2023/05/passw...

Etc etc. If this is the first time you are seeing businesses and media refer to passkeys as being simpler to use than passwords you haven't been paying attention.

How does this address OP's concern? If you have a single device (e.g. an iPhone) and you store all your passkeys on it, then losing it means you lost all passkeys. Your post describes exactly that:

- "I can’t recover the keys if I lose the hardware"

- "That is a risk you’ll need to take if you’re using hardware authenticators"

Fantasizing that with this proposed simplification of the authentication process people will introduce complexities such as a password manager or a backup hardware device is naive to say the least.

It is a bit disappointing. This video did not show anything more "extreme" than various Boston Dynamics videos from years ago. And to be even more pedantic, this is hardly parkour at all. Jumping over a gap and climbing on a box has little resemblance of what we came to expect when we hear this term.

Looks like it slows down just a tiny bit (needs time to compute?) before making some of the jumps in this video. So not completely losing momentum. Wonder if this will look more fluid once it's able to reach a higher speed.

My understanding is that such robots do exist and have a different application. For example, balancing on longer hind legs by lifting the rest of the body up, then making a jump over a gap is likely easier than when using wheels.

In terms of landing on the back: if I remember correctly, one of these prototypes--from Boston Dynamics?--is able to rotate its legs 180 degrees to push off the ground when landing on the back.

Agreed! After blocking Twitter I either see nothing where an embed is supposed to be, or only see emojis rendered in the middle of an otherwise blank area.

What a wonderful experiment! My favorite type of posts on HN.

I always imagine large scale financial fraud as something requiring a great amount of financial expertise. Seeing this code snippet is surreal.

That's just moral defeatism. Fraud exists in many businesses and institutions. A healthier outlook on this is that we need to stop it whenever we can, not that we should give up because "everyone does it".

I typed this from memory, so probably not repeating LLMs word for word. If I change that prompt to "As an AI language model..." the detection works as expected. In a sense, a broader point my original comment demonstrates is that even for an absolutely obvious AI-generated text, auto-detection can't work reliably because it's trained on specific responses of specific LLMs that can be altered at any time.

To be clear: not attempting to discourage you. It's a very complex and interesting problem to tackle.

Even in the US there are enthusiasts who buy and restore super cars for cheap (relatively speaking). There are several crafty YT channels showing the process from start to finish.

Yup, agreed. In the US the incentives are against repairs, not for. I don't know enough about cars to have an opinion on whether Tesla-style cast bodies will result in more cars being written off with similar damages to other fairly expensive cars.

Still happy not using PHP :)

The fact that WP is so big doesn't make PHP ecosystem much more attractive. And I used to do WP dev full time.

Looks like they load all assets directly from the "wp-content/uploads" folder (unless they reverse proxy this path to a CDN, which is unlikely). Plus custom fonts.

Was caught off guard that it rates the following text at only "3.2% chance AI generated": "As a large language model, I am not able to answer this question."

Not unsafe necessarily. The cost of labor is so high in the US that fixing even minor damage might not be cost efficient. Also consider the possible increase in the insurance fees if you continue driving a car after repairs. Other countries have lower cost of labor overall, so it's a profitable (and, hopefully, safe) business for them.

TIL about region-based diffusion. Thanks for the context!