inventor7777

As far as I understand it, yes, you have to be connected locally.

The question is, will it work over Tailscale/WireGuard? I'd assume so, but I've honestly never tested local pushes as I simply haven't needed them :)

How does the Neo getting to 100°C make it crap? By that logic, aren't all older Intel/x86 chips crap? If anything, I find it impressive that a small laptop CPU can do 100°C without a problem...my i7-7700T M710qs hit 75°C and throttle within a minute if I use a tool like y-cruncher or stress-ng. To be fair, totally different purpose.

Can't say I've ever thought of a word like "developery", but now that I've seen it I like it a lot :-)

While I have slight worries about what it means for users if Apple and Google notification services go down/censoring, I do appreciate the features that they provide to me as an end user.

So many apps use annoying and questionable marketing notifications that I'd say I have about 70% of app notifications disabled globally (because the app itself does not allow disabling notifications / has no granular control).

However, it seems that SOME self hosted services can directly notify you without APNS / FCM. As an example, see https://companion.home-assistant.io/docs/notifications/notif...

True, but the use case that Flipper shows off is a chatbot style AI, which is different from the ML type stuff you're talking about. I find small AI chatbots to just not be very helpful for technical applications, thus my comment.

Of course, if you had a custom trained one that's different but then that brings me back to how are you gonna type with it easily?

I really, really, really love this concept. I think there is SOME feature creep, but it does seem more or less scoped well to IP-type protocols.

However, I don't think they need to be prioritizing the local AI features, which are cool...but models get far smarter when you run them on a proper Mac/external GPU vs a small battery powered Flipper device. I think it might be helpful on the go, in the field, etc, but the usability with no dedicated keyboard will be rather poor.

However, I think they should keep focusing on the Zero for a possible Zero 2 to match the capabilities of this One device. I love my Zero, but I think it is missing key features like full support for garage door and RFID rolling codes, and some other protocols. The WiFi dev board is very limited, and there is no simple way to capture/playback BLE remotes IIRC. Of course, it depends on whether you consider BLE to be layer 0 or layer 1.

I'd say it depends on what you are trying to do. If it is simple device control and media playing and other stuff, then all you need is to update [dashboards](https://www.home-assistant.io/dashboards/) when you add/move devices and the users will find it pretty easy and straightforward. My parents are not extremely tech savvy but they find Home Assistant easy to use when I make the dashboards thoughtfully.

Making automations and scripts is getting easier every update, but it has a small learning curve as the logic can get complex and you sometimes need to know details like entity IDs or raw states. And there are some simple missing features that some people are very used to. Home Assistant is improving that sort of thing constantly, but sometimes the device APIs do not allow all functionality without the OEM apps.

For example, the two biggest camera-related things that are missing in my opinion is that the camera viewer does not allow zoom or two way talk. It uses the native browser media player, and on both a Samsung tablet and all iOS devices, this means that you cannot zoom and pan around the image. This is obviously not an issue if you embed a dashboard such as Frigate into the HA UI, which IIRC supports both two way talk and zoom. But YMMV.

I've got the Reolink PoE doorbell and it works great!

The very first thing I thought when I read this is "Hmm, wonder how long this one will last before Google kills it."

Well, I am still waiting for the price. If it is $450 or higher, I'd just get a MacBook Neo at that point.

That seems _awfully_ extreme LOL

I've used UTM before to install Ubuntu on my Mac Studio (M4) and it even supported my 4K70 monitor.

To be fair there is some config and tweaking required, but for a free tool it seems pretty good. Parallels has a better EXPERIENCE but I don't use VMs often; when I need raw Linux I just use one of my homelab servers.

Have you heard of this?

According to their GitHub, this should solve the issues you mentioned with Linux development on macOS. Note: I have not used it myself as I find macOS+Brew sufficient for my tasks.

https://github.com/apple/container

I agree somewhat, but I do still think there is a decently sized separation between true vibe coding (the typical "make me an app...fix this bug") and actual AI assisted development. I personally think that if you are a dev and you simply trust the AI's output, that is still vibe coding.

I am not a developer and have very basic code knowledge. I recently built a small and lightweight Docker container using Codex 5.5/5.4 that ingests logs with rsyslog and has a nice web UI and an organized log storage structure. I did not write any code manually.

Even without writing code, I still had to use common sense in order to get it in a place I was happy with. If i truly knew nothing, the AI would have made some very poor decisions. Examples: it would have kept everything in main.go, it would have hardcoded the timezone, the settings were all hardcoded in the Go code, the crash handling was non existent, and a missing config would have prevented start. And that is on a ~3000 line app. I cannot imagine unleashing an AI on a large, complex. codebase without some decent knowledge and reviewing.

This is a neat idea.

Personally, I would not find it useful as they seem relatively fragile and if the receiver is not bulletproof and VERY sensitive then it could miss state changes.

However, it would be interesting to see how it would work if you were able to have different ones on different frequencies with one main receiver. That would make a cost effective way to track cabinets, doors, etc in the same room.

In my case at least, (for support calls) it's not a "racist filter", it's that I sometimes simply cannot understand what they are saying.

Unless you want third party WebViews... (on normal Android)

(Technically besides the point, but that is a broad statement)

On Apple Silicon Macs, the RAM is shared. So while maybe not up to raw GPU VRAM speeds, it still manages over 450GB/s real world on M4 Pro/Max series, to any place that it is needed.

They all do have a limitation from the SSD, but the Apple SSDs can do over 17GB/s (on high end models, the more normal ones are around 8GB/s)

Ahh, that actually makes more sense now. (As you can tell, I just skimmed through the READMEs and starred "for later".)

My Mac can fit almost 70B (Q3_K_M) in memory at once, so I really need to try this out soon at maybe Q5-ish.

Sure!

Note: these were just two that I starred when I saw them posted here. I have not looked seriously at it at the moment,

https://github.com/danveloper/flash-moe

https://github.com/t8/hypura

Wow, that is neat! I had never heard of that before. However, they still have a long way to go to even equal the capabilities of normal phones made by huge companies for a typical consumer. Whether that is normal app support or even just simple stuff like camera quality.

Although...that was likely never the main goal for that phone anyway.

Weren't there some frameworks recently released to allow Macs to stream weights from fast SSDs and thus fit way more parameters than what would normally fit in RAM?

I have never tried one yet but I am considering trying that for a medium sized model.

I remember reading about some new frameworks have been coming out to allow Macs to stream weights of huge models live from fast SSDs and produce quality output, albeit slowly. Apart from that...good luck finding that much available VRAM haha

Hmm yes I should have been more specific. I am aware that it does not block trackers...my Pi-hole logs are still full of sketchy/tracking domains.

I simply meant that the unique ID can't be used to track me anymore, at least across different third party application companies. I would edit my comment but it is too late now.

Really? Do you have a link? That sounds very interesting and very frustrating.

Regardless, because of such things I'm guessing the only ways to disable such tracking in the foreseeable future will still be 3rd party non-affiliated DNS/extensions or browsers such as Brave and Safari (to some extent).

Yes, that is very true. Unfortunately, without a true non-Apple or non-Google OS, the ATT toggle is likely about as good as you can get. And it does stop the apps themselves getting raw access to your data, which is not something to sniff at.

When Apple first released App Tracking Transparency, I immediately used it to block the trackers and I have not even thought about it since because it is so simple and useful.

What a contrast to modern websites which require all sorts of weird clicking gymnastics to disable similar tracking.

That's relatively straightforward for more technical people, but I _really_ can't picture farmers going through all that trouble unless someone gets the foundation ready to go.

I'm pretty sure Tahoe added that behavior natively. I personally use Magnet on Sequoia, however, so I am not 100% certain.

Ha! Reading this comment made me curious, so I went back and looked at the article and there does seem to be a full sized HDMI connector. I wonder if it is enabled, or just for Tesla internal testing?

Hey, this comment sounds very AI generated. I could be wrong, but if so, you might want to read the newest HN rules.