Please contact me at @otter@lemmy.ca for any urgent issues.

This is an alt account used for testing and scheduling posts. I have retired the previous accounts I used for this purpose ( otters_raft@lemmy.ca and otters_raft@piefed.ca )

profile image: Photo by Kedar Gadge

  • 367 Posts
  • 62 Comments
Joined 9 months ago
cake
Cake day: February 15th, 2026

help-circle






















  • Hello!

    I’m sorry for the confusion. I’m not actually with coopcycle.org, I’m one of the admins on lemmy.ca . I made this post to ask if it would be ok to invite them and see if someone here would like to take this on. I’m in Canada and French is not my first language, so it didn’t seem like my place to talk to them :) I’m happy to do it, but I figured I should at least speak to your team first since I would be referring them to one of your instances.

    I agree with you that Piefed makes more sense for their use case. I was thinking that they could set up a community here instead of their Facebook page, which doesn’t seem to get that much engagement anyway.

    Language wise, I am more fluent in English. I’m sorry for the poor French, I didn’t get a lot of practice after learning French in high school.





  • It would be a playlist of the videos that people post in this community, I’m not making any videos. Right now I manually open videos from this community and save it to YouTube to watch later, and each one takes a while to load.

    As for AI use in the setup, I asked LLMs some questions to figure out how to configure the Google account / YouTube permissions, because the official guides seemed out of date and the third party guides contradicted each other. It feels like they don’t maintain proper guides anymore

    The actual code is fairly minimal, and easy to look over. Right now the setup is:

    • Use the Lemmy API to check for the latest posts, and remove duplicates / posts that are too new
    • Use the YouTube API with my account and a pre-configured playlist, and replace all the videos with the new set
    • repeat on a reasonable schedule
    • docker container so that it’s easy for me to manage




  • Niri does look nice, that’s the one I want to try someday.

    Back when I first started using Linux, I had some issues after trying to customize things (ex. Being locked out of the plasma desktop), and it scared me away from any big changes on my main device. I might take another whack at it now that it’s been a few years



  • From my limited understanding, this is about upgrading the backend systems that e transfers (and other payment systems) use.

    Etransfers feel instantaneous in the current system, but in the background the money doesn’t actually make it in to your account right away. That’s also why some payments take time to clear (whether a few hours or days), and why certain things can’t be done after hours / on bank holidays.

    Interac is the entity that will be running this new system: https://www.payments.ca/payments-canada-selects-interac-corp-exchange-solution-provider-canadas-new-real-time-payments

    More from this article:

    Canada is the only country in the G7 without an instant payment rail. RTR is meant to support higher-priced transactions of up to $100,000 and settle them instantly. Existing payment rails are limited by speed and transfer size. When making payments, Canadians typically see money leave their accounts right away, but transactions are actually settled within hours or days—which is why it can take up to three business days for a credit card payment to post. While Interac’s e-Transfer system is faster than those traditional payment rails, it typically limits consumer transfers to $3,000, and businesses to $25,000.

    The Bank of Canada’s senior deputy governor, Carolyn Rogers, said in a speech last year that RTR and open banking are two important changes coming to Canada’s financial sector that will likely improve competition. The establishment of RTR could bring more than $3 billion in efficiency gains to the Canadian economy over its first five years, according to a 2023 study by think tank C.D. Howe Institute.



  • I don’t think this would work consistently, since creative work and inspiration take time and the detection errors might depend on the length of the work. I also think that making some groups of people jump through hoops before they can clear their name, because a generative algorithm said so, is only perpetuating the same problem.

    Better would be to judge individual cases based on edit history + timestamps, and judging a tool based on multiple experiments like this. If it fails a few cases, then it should not be used.

    Back when such “detectors” first popped up, our university sent out a guide to profs saying that they are unreliable and not to be trusted. We need similar education for more institutions since this isn’t common knowledge yet.