• 0 Posts
  • 16 Comments
Joined 1 month ago
cake
Cake day: January 15th, 2026

help-circle

  • Well, you’d lose support for devices which can’t handle software DRM playback, old YouTube clients installed on things like TVs which no longer get updated, if you want to support things that only get Widevine L3 support (most devices) you’re not really going to move the needle since Widevine L3 had been broken since like forever, etc.

    The main thing YouTube would gain in practice from such a move would be to get DMCA as a legal tool to crack down on people ripping YouTube videos, but that’d require some very significant resources invested into driving Legal processes against average consumers ripping videos, and the return on investment for that is almost certainly abysmal.

    EDIT: I thought of two more reasons:

    • Given the enormous scale of YouTube’s transcoding pipeline, just adding in a DRM step into the mix will cost non-significant amounts of money
    • All old content will remain non-DRM, because a re-transcode of the full YouTube catalog would cost an impossible amount of money


  • While I am broadly speaking in favour of free transit, one thing that should be noted in any conversation about it is that it often does not significantly increase ridership by users who were previously not using transit - the increase mainly comes from increased utilization by prior users, and sometimes you can get negative effects such as the newly free transit trips displacing walking.

    Hence, other investments into the transit system may be prudent - for example, expanding the system, increasing headways and so on.

    We should try to be brave enough to do all of the above, but sometimes you’re forced to choose one or the other.







  • So far, there is serious cognitive step needed that LLM just can’t do to get productive. They can output code but they don’t understand what’s going on. They don’t grasp architecture. Large projects don’t fit on their token window.

    There’s a remarkably effective solution for this, that helps both humans and models alike - write documentation.

    It’s actually kind of funny how the LLM wave has sparked a renaissance of high-quality documentation. Who would have thought?