I think the owner wanted 100 Gbps of scan traffic or had set a specific scan-rate target, which determined that bit rate, so the LLM (correctly) predicted it needed all of those to hit the target.

That's appalling! At that stage, it's not on-call, it's just another badly paid shift, which you're probably forced to do in addition to your 9-5. Good on you for leaving.

You're right, but the blog author also seemed to be in the same position.

Great point, and even beyond that I think (based on the paths) it was just a command line invocation, with something like NFS handling all the networking.

Wow! I hadn't seen this, thanks. Do you think they are doing it with relatively innocent motives?

I have no insight, but I assume they are doing it because they can use AI to make a few variations of a video and then automatically A/B test them to see which ones get more engagement, and then use that to make videos that are more engaging than what the author actually uploaded.

This is "innocent" if you accept that the author's goal is simplify to maximize engagement and YouTube is helping them do that. It's not if you assume the author wants users to see exactly what they authored.

When do you see this? For me, I just go to System Settings → Privacy & Security. Scroll down to Security and look for the message about the blocked app, click Allow Anyway, and then reopen the app.

If I install Xournall++[1], just opening the application will fail and MacOS will ask me to move it into trash. I am required to execute `xattr -c /Applications/Xournal++.app` to "remove quarantine".

I just reinstalled and can confirm that I don't see anything in the System Settings as you say.

[1] https://github.com/xournalpp/xournalpp/releases/tag/v1.2.8

Not that oldness is much of a metric for quality, but Postgres does go quite a lot further back: https://en.wikipedia.org/wiki/PostgreSQL#Ingres_and_Universi...

They can do so today. Works great for my parents.

This may be an obvious point, but I didn't see it mentioned in the (otherwise excellent) article: I would have been interested in the cost saving in just implementing the 'delete on read' with S3 that they ended up using with the home-made in-memory cache solution. I can't see this on the S3 billing page, but if the usage is billed per-second, as with some other AWS services, then the savings may be significant.

The solution they document also matches the S3 'reduced redundancy' storage option, so I hope they had this enabled from day one.

Yes, they mention this as a 'fix' for connection-related memory usage:

> Disable keep-alive: close the connection immediately after each upload completes.

Very odd idea.

Possibly missing session resumption support compounding the problem.