Ruben isn’t super quick to put out updates but he makes up for it in quality. He was slower than some other devs to get Boost for Lemmy out the door but the first release was damn near perfect, stable, fast and only very minor bugs. Personally I prefer quality over constant updates.

These developers owe us nothing and it takes an incredible amount of time and lots of money to develop an app of this quality so no matter which app you choose consider paying and/or donating.

Nobody is coming after you, don’t worry about it. If they were really determined? Maybe, possibly, depending on many factors but you’re a very small fish in a very big pond.

If you have dark mode enabled in your android display settings then you can use the dynamic color option in boost to inherit amoled black from there.

Look I love FOSS, but this mentality that using anything except for FOSS is dumb. An incredible amount of time, money, and effort goes into building an app like Boost and the developer has every right to keep it closed source and charge for it and you have every right not to use it.

Many people are more than willing to pay for great software and others are happy to give up some privacy to get it for free. That’s their choice.

Yea they’re internal. That’s normal for a fully loaded 2u storage server. Some even have 2-4 extra disk slots in the rear to cram in a few more.

I concur and it just gets worse the more hardware you have in them. 256G of memory and 24 disks? Might as well go have lunch while it boots.

That’s awesome. I seeded some this year, hoping to have the same in a year or two.

They’re taking a whack-a-mole approach for sure but it’s either that or shut the whole instance down. I imagine their hope is that either the bad guys give up/lose interest or that it buys them some time.

Either way, it shows they are taking action which ultimately should help limit their liability.

If you think it might help I’ve got a bit of a hack I’ve used in the past to cache a sql database in a compressed ramdisk using zram and bcache. Imagine stuffing a 50G DB into 20G of memory.

It won’t fix the inefficient SQL queries but it would make it so frequently accessed tables get cached in a ram disk cutting query time significantly.

This might be enough to reduce the impact of these attacks until queries can be optimized.

This assumes your database isn’t running on something like RDS though.

This made me laugh. Configuration management systems like ansible, chef, salt, and puppet only exist because people wanted to manage a large numbers of systems and keep them consistent and replaceable, i.e treat them like cattle instead of pets. They were born out of the pets vs cattle analogy.

I realize containerization has taken that a step further but it’s funny to hear someone talk about these tools like they’re something archaic.