Unrelated, but in that link:

> Interestingly, the code continues to check the entire hash after a mismatch.

This is a standard practice in cryptography, but maybe not at the time.

> There we did see some file corruption creep in

Did you figure out what caused corruption? Was minio losing blocks or was juicefs corrupted even though minio was consistent?

It was definitely MinIO related, I probably should have made that clearer. We noticed that with zero fault tolerance, MinIO objects would randomly become corrupted, which MinIO would present as "you're making too many requests, please slow down". We were certainly not making too many requests.

What is your plan after MinIO enters maintenance mode?

We're looking at alternatives, I've made some previous comments on that front. Sadly MinIO was the only option with sufficient performance for this particular situation. Thankfully we're not using any MinIO-specific features, so at least the migration path away is clear.

Ceph. The answer is always Ceph.

How does she get around the restrictions? I'm not aware of a way to do that on iPhone

Change the region. Not time and date, that can be blocked by screentime. Change the region. Sometimes takes multiple attempts but it does work. I had to pay a bug bounty to my child to get them to tell me how they did it

uninstall app when limit runs out, reinstall used to work.

For anyone (like me) wondering what this is. Foamcore is foam with paper on each end (https://en.wikipedia.org/wiki/Foamcore)

Looks like people use it to organize board games: https://www.orderofgamers.com/abomination-foamcore-v1/

Disagree. "Just do it" works for some people. Not everyone.

In my experience, I hated looking at a gray screen, so I just stop using my phone when it's in grayscale.

But they didn’t say “just do it” :)

They said “spend effort ensuring other meaningful time-draws are around that are more fulfilling/less damaging”.

Relevant excerpt:

> the actual problem is finding meaningful activities to replace your empty screen time, which isn't necessarily easy.

That's true. I came off too dismissive. I like the advice, I was just disagreeing with:

> it’s clear to me why they rarely work: ...

and

> boring old discipline is much more sustainable approach

I like the idea of trying a more holistic approach, but grayscale doesn't have to contradict with that. Some people (myself included) can use grayscale as a tool alongside other things.

Approximately the instant when a single component (RAM) of a comparable product (Gaming PC) became more expensive than the entirety of said product.

I wonder what you'd think if bus tires exploded in price and started costing .25 London busses per tire.

> We can make it a bit more terse using list comprehensions:

    import functools as ft
    def fact(n):
        lst = range(1, n)
        return ft.reduce(lambda acc, x: acc*x, lst)

Amusing that there's not a list comprehension in sight.

Nextcloud isn't lighter and leaner outside of docker. It isn't faster either

On the other side, docker is just installation system, so why even care.

You wouldn’t deploy NextCloud if not for docker.

I actually manually ran both ownCloud and Nextcloud manually before switching to docker.

Edit: just checked because I remember upgrading to Owncloud 6... This was 12 years ago to be fair.

Strange, on my nixos seems to be possibel!

That isn't a bug; it's working as intended. I do wonder if that could be mocked up easily in html to see what it feels like though. Sounds like a cool idea

I'd believe it. Not sure when this is, but if it's a few years old and business software, they could probably asume everyone uses java, which doesn't even have unsigned integers.

Right but just because it's `int id` doesn't mean all code that uses it will still work when it's negative.

True, but it does seem like the best alternative here. If it's a SOAP API in 2005 for business customers, for example, then it sounds like the least bad option of the four (tell consumers to update, hold up the whole company's deployment, push negative ints, or push longs). I'm just saying that to me, it isn't hard to believe this was the best option here.

With $MY_JOB in java, that was my assumption