ey b0ss

I live in a constant state of fear and misery

Do ya miss me anymore?

Look closer at the beauty mark, I flipped the emoji

"☹️".reverse() == "☹️"

Correction: you don’t work around dangerous things assuming you’ll make a mistake long

When I was young, people (read: other kids) would accuse me of being pretentious for using vocab words. I learned to dull down my speech to please them, and lost most of my vocabulary in the process. Now i talk nurmal lik ery1 else

Take the passive-aggressive nerd approach:

1. Start a niche online movement that only cares about one aspect of computing and convinces people all their problems are caused by your pet peeve

2. let the company dig its grave

3. create a FOSS alternative

4. sell a premium version for businesses (it includes phone support and management-friendly marketing matetials)

5. congrats, you are now the de facto standard software in your field

In the year 2025
if man is still alive
if woman can survive…

I think it’s just the utopia image with colors inverted, which is a visual gag that I’m loving

right? I know the unix philosophy disagrees, bur goddamn is it so convenient and versatile.

plus overpowered CLI tools like ffmpeg are the definitive case where CLI is definitely preferable to GUI by almost every metric. Why go hunting through a labyrinth of dropdowns and dialog boxes when I could just search the man page and type a few characters.

stupid question, but does your server have a video card or only the CPU’s (or SOC’s) built-in GPU?

If you do, you might want to make sure you’ve configured ffmpeg to use the hardware acceleration

Also, could be that unmanic is trying to reencode the files, which would eat up wayyyyyyyy more resources than necessary. It looks like the least overhead/performance hungry option will be to just run ffmpeg directly, so you know exactly what’s being done.

thanks for the red circle, otherwise i’d be illiterate and unable to see the only thing in the foreground

Oh god, you probably got the gonocacococcus

That’s great news! I hope my contributions didn’t require major overhauls on your part, and I’d love to see the end result (scrubbed of any private info of course) if you care to share.

No worries about the FOSS stuff, I was just fishing for excuses to absolve myself of guilt for not finishing it 😬

Here’s the journal article itself

This whole thing stinks of advertising. They use a sample size of 24 people. The results are underwhelming: “Each disease cohort concordantly demonstrated increases in the NAD+/NADH ratio but did not reach significance individually” (“when we massaged the numbers, the result we wanted popped out”). Looking at the graphs, it looks like the upper half (highest baselines) of the patients worsened over the 12 week study.

My favorite lines?

“The objective of the Phase 2 REPAIR-MS and REPAIR-PD clinical trials was to investigate the effects of CNM-Au8” and “CNM-Au8® is a suspension of faceted, clean-surfaced gold nanocrystals” and course:

“Clene Nanomedicine, Inc., funded this study.”

None of the above. This study is as fishy as they come

So we’ve extracted so much gold from the earth there’s not enough left for our bodies?

I know this probably isn’t the case, but the image of extracting precious metals from human tissue solely to squeeze every last drop out of nature is hilarious to me. Plausible, but hilarious nonetheless. Mine owners leaving toxic lakes in the middle of Parkinson’s patients when the earth runs out of gold.

@jawa21@lemmy.sdf.org

here’s the script

I wasn’t able to implement catbox nor youtube uploads for a few reasons:

1. neither are FOSS services, and I sneer at those in my moral superiority (disregard the above usage of github)
2. I’m lazy
3. This script isn’t for me and I will likely never use it
4. I’ve already spent over 2 hours on this
5. catbox.moe is a small site, with little-to-no API documentation. I didn’t want to hammer them with 100 API POST requests with blank files while I test the script and figure out their API through trial and error
6. youtube requires an API key via oauth… I really don’t have time to work that fustercluck out

I left in some helpful comments anyway.

All in all a fun exercise that brought me and lemmy closer together, and that’s what really matters. Turns out the real API docs were the friends we made along the way

sounds like a fun challenge. My thought process is the control flow. To that end:

0. script runs

handled by cron

1. Script selects files to upload, community to upload to and titles for them.

These are the hardest parts, because it will be the only variable parts. It basically requires user input. I’m thinking perhaps parsing a text file where you input the file name, the community, and the title.

2. Script authenticates with lemmy API

I found this really handy “unofficial” API documentation, which includes examples in many different languages, through the official docs.

Unfortunately, it looks like it’s going to require a hardcoded password to get the appropriate authorization token.

3. script posts to lemmy

We can use the api documentation above to figure out this part too

For other sites like youtube, you can expand on this script and have parts for youtube as well, but you’re basically doubling its size and complexity. Personally, I would go for the KISS method instead and just have a separate script & info file.

After work today, I’ll see if I can slam out a quick proof of concept in Python

Right next to the Yarb Sale