Hacker News new | past | comments | ask | show | jobs | submit | Allezxandre's comments login

Is it?

My experience with Arc (which I would assume would be heavier than Chrome) has been that it adds a lot of scary entries in the Resource Monitor App, but doesn't impact my day-to-day use on my 16 Gb M1 MBP.

With Safari, on the other hand, I've experienced audio stutter and delays when using resource intensive wep-apps like YouTube, Figma, Netflix and others.


YouTube and Netflix are resource intensive webapps? What is so resource intensive about them? A <video> element?


That would be the case if the Netflix UI was a search box where you type in the name of a movie and then it starts playing.

The actual UI is a lot more complex. On the front page there are dozens of dynamic categories that are filled out with high-res thumbnail images, and the content starts autoplaying within a few seconds when you stop on an item.

Now, I’m not saying that’s necessarily a good interface (personally I hate overactive autoplay). But from an engineering POV, clearly there’s a bunch of prefetching and cache management that needs to happen to make that UI look seamless, without loading pauses or missing content at any step. And media caches do qualify as “resource intensive.”


You can turn that autoplay off now in the settings.


Widevine DRM for Netflix. AV1 decoding and the SPA for YouTube.


On my Mac with Safari, Youtube uses VP9, even with 4k HDR material. Apple Silicon Macs have VP9 hardware decoding, so the load is very small.

I have never experienced audio stutter and delays like the upthread commenter claims.


Apple's M3 has AV1 hardware decoding, so I imagine that shouldn't be a massive issue in the future.


Not for YouTube, but Netflix DRM possibly?


You think DRM is implemented in software? :D


Wikipedia [1] has a problem statement that I find explains the game quite simply:

> the busy beaver game aims at finding a terminating program of a given size that produces the most output possible.

The Busy Beaver Challenge website [2] also has an explainer page with interactive Turing Machines.

[1] https://en.wikipedia.org/wiki/Busy_beaver

[2] https://bbchallenge.org/story


I understand where Level is coming from with this response, but cybersecurity has taught us better.

Yes, less than 10% of attacks use this exploit, but now that it’s public that you’re vulnerable to it, you become a target for 100% of attacks involving this exploit.

So for instance, if the delivery address of Level customers leak (and given the security of their physical locks, I would say them having bad cyber-security practices is not something too far-fetched), a malicious attacker would essentially have a database of home addresses with users that have bought a $300 locker that he can break into with the help of any YouTube tutorial


Spoiler: the majority of typical residence locks are no better, and in reality there are far less malicious attackers than you think.


I don't expect a $5 to try to do much beyond preventing someone from accidentally opening the door. I do expect a $329 lock to not skimp on basic features to cut a dollar off the BOM.


I remember looking at this lock. I really wondered how strong it was because the battery is in the deadbolt. How easy would it be to smash the door open?

I never even considered a bump key might work. And raking? It’s like they’re not even trying.

You’re right with the $5. This now sounds like a $5 lock with automation around it. I’m not saying it needs to be the best lock on the market but come on.

And Level’s response is just sad. It’s very clear they got caught flat footed (shouldn’t happen) and are trying a bad move-the-goalposts to save face.

By their argument you don’t even need a key. Just let someone insert an Allen wrench to turn the bolt directly! No one breaks into houses that way. It’s secure!

Wow.


So long as the battery is on the inside piece, it really isn't much different from a manual deadbolt. There's some extra gears and the motor, but they're going to be behind plates of metal- if you are taking a crowbar or hammer to one, you'll make enough noise that you might as well just break a window.

On the other hand, a super distinctive deadbolt on your front door that is known to be easy to open with a bobby pin is basically anti-security as you are advertising that your home can be entered without leaving a trace.


If you’re gonna have your house robbed anyway you might as well save on the lock. Even better if you don’t add your name to a database of people with weak locks.


Yup.. only takes one data leak


A locked door is not going to stop someone entering your house if they really want to get in. Doesn't matter what lock.


The key is if they're trying to get into your house or if they're just casing the neighborhood looking for appropriate targets.

It's the "only have to run faster than you" joke.

And since this lock is a joke, having one is a nice little advertisement.


That's a chicken-little argument of false equivocation. There is often to always a level of security sufficient to make forced entry too costly.


On iOS, just hold the button you would use to turn the phone off, but hold longer, it will loudly count down to 0 and call emergency services. You can also press the side button 5 times and it will call without you having to hold the button pressed. Works on iPhone and Apple Watch.

More info here: https://support.apple.com/en-us/HT208076


And as well, if the FaceID doesn't work & the PIN is prompted for, "Emergency" is available in the bottom left.


Press and hold doesn’t work if you have Siri enabled.


It does, just hold the buttons you would hold to turn the phone off, not the button you would use to trigger Siri.

On FaceID phones that means holding both the Lock and Volume Up buttons.


But the point is that Apple Maps also has the commute suggestions without the privacy cost


As long as you're using a cellular network, you're trackable, no? https://www.youtube.com/watch?v=VFns39RXPrU


Maybe, I don’t know, but it doesn’t mean we should just give up on protecting our data everywhere else.

A GPS or Bluetooth signal would be far more precise as a mass surveillance tool, so let’s not give that up


Not being tracked in addition by an advertising company is worth something to me.



Because it advertises a shorter time to arrival?

This has been an issue with Google Maps for some time: being more optimistic than e.g. Apple Maps just so you use the service, as suggested by this article: https://arturgrabow.ski/2018/02/19/navigation-apps/


> But let's say you're a technically savvy person that is actually interested in cookie privacy; I'm really surprised you're not using "cookie autodelete"-style plugins already.

I feel this is like saying "we don't need safety legislations at work because businesses take a hit trying to stay compliant, if you don't want to be hurt at work, I'm surprised you're not wearing a helmet already."

Thing is, I wouldn't need "cookie autodelete" style plugins in the first place if companies cared about data privacy.


You shouldn't wear a helmet because other people tell you need to wear it. Instead, you should wear it in order to be safe.

The problem with law is that only good guys abide the law. And you don't need to defend yourself from the good guys, only from the bad guys. And bad guys will violate the law anyway.


Proof-of-work really is a nice system


Hey it's slightly off topic, but if you work with Datasets, a struggle I often have is sharing them, so I built a service for this that you can find here: https://Joule.Host


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: