There’s a weird bug in Ubuntu 24.04 where some windows cannot be resized if you’ve got your system set to automatically log in when you boot up the computer.
Odd.
If you log out and log back in, everything works fine again. But I don’t want to have to do that. So I delved into shortcuts and turns out in Ubuntu’s GUI, Gnome, if you hold down the Super-key (Windows-key on a Windows-themed keyboard) and drag the edge of the window with a mousewheel click (or middle-click if you’re still rocking a 3-button mouse) you can resize the window in either direction.
I have a GoPro Hero4 Session. This little camera was amazing when it came out. I still think it’s still pretty amazing today. (Aside from one annoyance: I can’t get Quik to connect to it from my Android phone because it insists that there’s a PIN set on it.) It does what I want it to do. Takes half decent action videos in 1080p at 50Hz and wide-angled quite average stills. In good light the photos are pretty usable as happy snaps. It’s waterproof without needing a case. And it comes with all manner of mounts. It’s currently being used on the front of my bike to capture any excitement that might happen as I ride around the place.
But where it falls over completely is file naming convention. Videos start off with a perfectly fine “GOPR6014.mp4” format. The 6,014th photo or video shot on the camera. Perfect. But then, when that video gets to the filesize limit, it creates a second file with the name GP016014.mp4. And the next file it creates is GP026014.mp4.
This means that when you sort the files by name, you get a list of all of the second files for the videos, then third files for the videos, then fourth, etc., until you get to a list of all the base files.
It’s horrible and makes a mess of any storyboarding file editing software you can imagine. (Unless there are ones out there that will accept this horrid GoPro standard that I’ve not yet found).
import os
import re
# Get all files in the current directory and skip directories
for filename in os.listdir('.'):
if not os.path.isfile(filename):
continue
name, ext = os.path.splitext(filename)
# Rename GOPR files: GOPR6014.mp4 → 6014-00.mp4
if name.startswith("GOPR"):
new_name = name[4:] + "-00" + ext
os.rename(filename, new_name)
print(f"Renamed {filename} → {new_name}")
# Rename GP files: GP016015.mp4 → 6015-01.mp4
elif name.startswith("GP"):
match = re.match(r"GP(\d{2})(\d+)", name)
if match:
suffix = match.group(1)
main_number = match.group(2)
new_name = f"{main_number}-{suffix}{ext}"
os.rename(filename, new_name)
print(f"Renamed {filename} → {new_name}")
This takes all the files in the directory that you launch the script from and renames them into a more sensible structure.
GOPR6014.mp4 becomes 6014-00.mp4. GP016014.mp4 becomes 6014-01.mp4. GP026014.mp4 becomes 6014-02.mp4. and so on.
This now sorts properly, and makes life much easier.
Enjoy.
Test this before you use it on something important.
It would have been better if I’d spent some money on new disks. But I kinda got the shits up when i opened the start menu and there was yet another notification bubble on the gamified thing where you collect Microsoft points for somethingarather by doing something. Then there was a thing about unicorns. And then Copilot was back, telling me that with AI I wouldn’t have endless search results in my start menu.
I don’t want search results in my start menu, unless it’s an application I have installed on my computer.
So anyway. A bit of a hiccough with my ownCloud and I’m syncing that down from the cloud again (sorry, old-old work hosting it still, but your upload is going to be about 700GB in the next couple of days (you have my number if you want me to stop)).
Email is rolling in. Signal and Whatsapp are linked. Browser stuff is all working fine. 1Password was disturbingly easy with the QR code scan. I’ve mounted the NAS drives. Music is playing.
Todo:
tidal-dl
yt-dlp
Torrent client until I set up a “proper” setup
So far so good.
This is the first time that I’ve used Ubuntu and apt on a properly powerful machine, and I’m kinda blown away by how fast it all goes. I’m used to being able to read what packages are, but no chance with a desktop CPU, 32GB of RAM and an M.2 disk.
IT security has been in the news lately with the theft of money from superannuation funds. It’s bad, but when you consider that they only made off with $500,000 from the $4.2 trillion that Australians have in super funds it’s pretty inconsequential. It’s not going to ruin anyone’s day. Not even devops, because they’re saying it was a credential stuffing attack.
To do a bit of maths: $500,000 from $4.2 trillion is 1.2×10-7%. Or 0.00000012%. When it’s easier to work in scientific notation you know it’s insignificant.
But it could have been bad.
Preventing credential stuffing is pretty easy – don’t reuse your passwords. This becomes nearly trivial when you use a password manager. You don’t need to remember the passwords that are generated and they can be long and random. The one password that you need to remember to log in to your password manager can be long and simple. (You can look at the maths behind it here: XKCD.com.)
Then there’s that second factor authentication. When you try to log in, you get an SMS or email with a code sent to the number or address on file that you need to enter before you get let in. Or you need to dig out a number from an app that changes at regular intervals in sync with a service that you’re logging in to.
There’s a third factor that is tricky to implement – some sort of biometric factor. Like eyes, fingerprint, DNA (Severance style).
<pretend I found a jif of the fingerprick scenes from Severance>
I don’t think my super fund has 2FA (or MFA) set up. So I can’t rely on that as a second factor. But I do have an email service that allows for subaddressing email addresses. That link goes to a very dull publication about what subaddressing is.
In short, it’s that you can add details to the left side of an email address – the bit before the @. So if my main email address is ko@example.com, an email sent to ko+sometext@example.com will be delivered to ko@example.com, but with the +sometext as part of the to field. This is really handy for filtering emails.
It’s also handy to find out whether your email has been part of a breach or has been sold. If you suddenly get emails to ko+company1@example.com from Company 2, you know that your email address has either been sold or stolen. Handy for the curious but not that useful.
But where it might come in handy is to add another factor of authentication to limit credential stuffing. If you’ve happened to reuse a password somewhere, but if the username for the places are different, then you’re steps ahead in protecting yourself because there’s no match.
In many ways they never left. But often they were ignored, left to collect cyber-dust and possibly go horribly out of date, needing many updates as they are revived. But they’re coming back. The Great Centralisation that started in the early 2000s with the advent of Facebook seems to be coming to an end. At least in my circles. Old blogs are being revived. New blogs are popping up. People are talking about RSS feeds again. Platforms. Favourite way of doing formatting.
It’s really magnificent.
But then – how do you keep track of everyone’s blogs? How do you know what to pay attention to, if there isn’t an algorithm that manages it all for you?
There are ways. And here are mine.
Some blogs let you subscribe to a newsletter that sends when the blogger posts a new posting. I do this with a few of my blogs. Not this one, though. It’s a nice way of doing it if you don’t hate emails. I don’t actually get many good personal emails any more – it’s mostly notifications – so getting an email full of delightful content from people I like is really refreshing.
RSS. Really Simple Syndication. A standard that’s been around for 26 years (yikes!) and not changed in the last 16. It’s a standardised XML file of the content on a website. There are many ways you can subscribe to an RSS feed. Your browser can (or should be able to) create what looks like a bookmark folder that updates with all the latest postings from a website. You can get a special feed-reader app on your phone. Or a feel-collecting page on a website. Or in an email client in a similar way to the bookmark folder in a browser.
Or my favourite way: RSS-to-email. You can run scripts in Git Hub that will gather all the updates from a collection of RSS feeds, format them neatly into an email and send it to you however often you like. I’ve got my email set up to be delivered around 6am, so it’s there to skim as I’m waking up. It’s a bit of a fiddle to configure initially but having updates delivered daily, without me needing to open Yet Another App. I’m checking emails anyway.
I’m trying to collect a nice blogroll here, too. It’s a work in progress, obviously.
I don’t read a lot. I think it’s because I read slowly. I have occasionally posted reading-log (and one reading-slog) entries here. But I’ve moved that over to Futzle’s Bookwyrm instance.
