Hacker Newsnew | past | comments | ask | show | jobs | submit | obenn's commentslogin

The AWS equivalent to Cloud Run and Container Apps is called Fargate, https://aws.amazon.com/fargate/


Neither Fargate nor App Runner scale to zero (unless something changed). So there is always a baseline cost of a few dollars.


Lambda container would be the closest equivalent with that functionality, https://docs.aws.amazon.com/lambda/latest/dg/images-create.h...


Agreed. I ran a small GPT chatbot on Lambda through API Gateway with a dynamodb storage backend and don't recall incurring any cost (or if I did, was just pennies per month).


But if you’re constantly pinging the container (as suggested above), it will never scale to zero.


It "scales to zero" as soon as the request stops as far as billing is concerned.

However, the image remains "warm" and incurs zero cost once the last request ends. So I usually have a `/heartbeat` endpoint for this purpose and point a Cloud Scheduler job at it.

I haven't read the docs to figure out the exact heuristics of when it becomes "cold" again.


One example would be anybody who deals with lots of data transfer, like video editing. Accelerating that with a better cable can really add up.


The cable can't accelerate data transfers. Either it works, or it doesn't¹. If it works, it works at full speed.

The bit specific to this Apple cable is that it does TB4 at 3m. That's not possible without active electronics to regenerate the signal in the cable; the maximum length for a passive cable is 2m. Passive 2m cables can be had for about $30 (noname) to $70 (non-Apple quality brand).

Fun fact: if you integrate optical transceivers, the cable can essentially be arbitrary length. (Though at some point you run into problems due to neither hardware nor software being built to work with the built-up latency.)

[¹ - for the sake of completeness, yes, these cables can partially fail with either sporadic bit errors or breaking one of the lanes. Both of these failure modes are extremely rare.]


Was ready to buy this until I saw it is a subscription, not stand-alone.


Price seems fair, honestly. The full blown app is like $70, plus upgrades when those come.


I’d gladly pay 70$ for this, for myself and maybe a couple of friends, if it were a one-time purchase. As a subscription, I just closed the page.

Several apps on the Mac App Store have both subscription and lifetime pricing. I’ve bought several.


The non-Mini version of Little Snitch is a one-time purchase.


The non-Mini version was such a hassle to maintain¹ that I stopped using it². This looked like a great middle ground, until the subscription part.

¹ I don’t want to become a networking expert and babysit everything. I already practice good security regarding what I install and run, I just want a little more control.

² And if it was a bother for me, there’s zero chance any of my non-technical friends would last an hour.


I forget what it's called but there's a mode you can put Little Snitch into that it allows whatever goes through and you can monitor it by opening and reviewing when you want. I think that's fairly similar to what mini does. This turns off all the nagging.

This is actually what I did last time I used Little Snitch, it helped a lot. When I installed a new app I'd use it for a bit, then review in LS and see what it was doing. From there I could allow or disallow whatever I needed.

It's sort of the reverse of "block everything until allowed"


You still need a Mac with Xcode


Are you sure this removes the ability to perform ANY software update? As far as I can interpret from this email this just seems to remove the ability to perform the update remotely.


Maybe? The remote firmware update is non-disruptive.

In the best case, I risk bumping cables while futzing with a USB stick. In the worst case, it wants a power cycle to update via USB.


In Ontario it is regulation to be able to obtain any of your car parts that got replaced, I wonder if it’s possible to request the old battery…


The $15k is after a "core charge" of ~$10k (iirc)

(aka at least in the US they can basically charge whatever they want extra if you demand your old parts)


That's because the old parts are worth money. Check how much extra they want if you request your old tires after getting new ones. Hey, they took money off! Because folks might want your old batteries but they sure as hell don't want your old tires.


It generally is, but in this case the insurance paid for it. I'd think they would be the ones that could do that.


In this case, not enough people know/care how to cook.


This is an integration with Apple Wallet and Apple Pay, so it does indeed go beyond a regular app.


The issue is in the combination of data. Could you imagine the value of knowing if your heart rate went up after reading a certain message, or seeing a certain ad?


I don't care about the value.

What's the cost?


We can’t know the cost without knowing the value. Nothing will happen to the data unless someone buys it.

To identify the costs, we’ll have to think about who might want to purchase this data and what they could do with it.

Ask yourself this: why would anyone be willing to pay for your heart rate?


>Could you imagine the value of knowing if your heart rate went up after reading a certain message, or seeing a certain ad?

No, I'm guessing it's worth a couple of cents ?


Perhaps. But this kind of psychographic profiling is used in labs to craft better adverts. Video games are another example.

The Left 4 Dead director was trained on player emotions and designed to improve the game, which was a huge success.

Having access to it in the field has real applications.


Isn’t 1Password secrets the 1Password for developer secrets? https://1password.com/secrets/


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

Search: