How are you planning on querying the data? I would plan your columns and indexes based on that answer. I’m going to assume arbitrary time ranges.

Just indexing a standard datetime column should be sufficient but that’s going to depend on a few things. Like you’re not accessing date ranges 10,000 times per second (website data source). Or you’re not inserting 100,000 records per second. Even then you might be able to get away with it if you want to throw more hardware at it.

The answer to most of database design questions at larger scales depends on the scale. If you need to query this date once for a report you do it in a spreadsheet. If you are building the next Facebook api then you shard the crap out of it into a time series database and add layers of caching.

My suggestion is to build it simple first and test. Don’t make assumptions about scale before you have data to back it up…

American Chestnut. Have a few seedlings we planted in the front yard. Super excited to be part of the process of restoring them

I have these for some raspberry pis around the house and they work great: https://www.adafruit.com/product/3785

Going home and relaxing. It’s been a physically demanding week and my body could use a break

My workflow for editing css mostly revolves around playing around with the live css directly in the browser until it looks the way I want it to then making those changes in the file and reloading to make sure it’s correct. The immediate feedback of seeing the changes happen makes it rather pleasant

The second icon looks like the Kirac vault pass rewards one. You have to open the chest in the vault for that to go away. Not sure about the first one though.

A lot of this is personal preference but I will suggest the following strategy. Mount all of your drives into subfolders of /mnt or /media (/mnt is usually used for more permanent storage but either is fine). Then symlink various folders on the system to this mount point. Like maybe you want your home folder downloads on one of these drives so /home/spawnsalot/Downloads is symlinkef to /mnt/drive1/Downloads.

This lets you pick and choose various places across your system that are actually on the additional drives but also the ability to see everything on the drives in one place.

Game installation location completely depends on the game itself. Some might install to /usr/bin, others to /opt, etc. You might have to dig around a little after install, move the folder, then symlink it like nothing ever happened.

I have not tried either. I’ll look into them though, thank you!

It’s a work provided device. It was either that or Windows, so it was really the only option. I might try out some tiling options for it, not sure how upset IT will be about that though…

It’s also incredibly useful as a failsafe in a helper method where you need the argument to be a string but someone might pass in something that is sort of a string. Lets you be a little more flexible in how your method gets called

Couldn’t you use painter’s tape? Unless your glue joint is thin, it won’t matter if some of the oil seeps past the tape. Pre-finishing is way easier than finishing after final assembly.

I spent years using i3 as my main machine and I loved everything about it. Fast forward to now where I have to use a Mac. Most of the time I’m in a terminal with tmux so it’s fine but any time I have to deal with a gui element that is under something else I get more and more upset.

I’ve had buttons stop working. The mechanism inside that registers the click is a mechanical switch and they eventually die

If you can somehow manage to install software on your work laptop then I highly recommend Synergy. It is by far a better solution than a KVM and you could keep the keyboard and mouse plugged into your PC so no effect on gaming.

Tung oil: https://youtu.be/l9fqCJ5kJiA?si=342VtEBPuI_JmVkB

Generally you plug a cable in from the UPS to the server and install software that monitors the UPS. You would set up that software to power down the server.

Either dia or inkscape

If it were me I would give it a shot and use a thermal camera to see the temperature differences. Worse case you have to patch the hole. Might also be a good idea to contact an HVAC company to see what they think.

Promtail is for collecting metrics from log files, Loki is for log aggregation. Really depends on what your needs are. Do you want to be able to view the actual log entries or do you only want to see metrics about the logs?

Why would you back up data you don’t care about? I agree, it does seem pointless.