"If you manually place your ad breaks, avoid placing breaks at disruptive points
Ah, yes, because the auto-place system is so good at choosing non-disruptive moments like <squints> 44 seconds into the video.
Seriously @TeamYouTube, I can't tell which I hate worse, the midrolls less than a minute into a video or the midrolls immediately before the credits, but both have to go.
What's happening is so many creators like to have a cold open, then play a small title card, and that creates exactly the kind of pause that the auto-placement looks for. But the system isn't considering how users actually use the system.
An ad break less than a few minutes into the video (I would honestly say 5 minutes, personally, minimum) is *extremely* disruptive, and the new system clearly doesn't care if the viewer was just served an ad as a pre-roll.
Like what the actual hell @TeamYouTube?! In what world is this placement algorithm acceptable?
There's no way to efficiently check each and every old video, it's a lot of clicks and a lot of scrolling per-video, so I've mostly just checked my most popular videos and handle others as I happen to have reason to check them. It's exhausting.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Okay, so, I was re-watching some of James Randi's debunkings and I got to his encounter with James Hydrick and, wait, something's off about this... he's not... is he...?
Oh no.
Oh no.
Oh nooooooooo.
Now, maybe I'm focusing on the wrong thing in the life of this convicted child molester, but, yes, James Hydrick used to goose his psychic act by putting on "oriental" makeup.
Between the original production, promotion, reshoots, distribution, Snyder Cut, and god knows what else, this is going to have cost WB almost $600 million.
Parent AT&T is already over-leveraged AF and getting desperate, so definitely nothing good.
The Q faithful have started showing up in the comments, and this one long heartbreaking post is, I think, a good example of synthesis and integration (in addition to many, many, many other things)
The thing I want to call attention to is the "only 9% of child abductions are reported."
At some point a Q pundit made some ludicrous claim about how many kids are abducted from white, middle class, American neighbourhoods every year. It get repeated credulously.
Someone, at some point, pushed back and said "wait, that's a lot, way more than are actually reported." They tried to debunk with facts (reporting numbers) and logic (parents would be looking for their kids). Instead that debunking gets folded into the mythology.
As I've been talking a lot about the hangups in captioning on YouTube lately, it's deeply unfortunate that they've decided to drop this (flawed) feature rather than improving it.
The main reason they're discontinuing is "low use". The issues with the community captions system which contribute to that low use are that it's vulnerable to abuse and it's largely invisible to channel owners.
The alerts system has been broken for ages, now. Either you get a notification with every single subscription (i.e. potentially hundreds of thousands or even millions) or you get none.
I spent a lot of the day thinking more about apocalypse fetishists who like to compare themselves to wolves and other predators, but the funny thing is that they know so little about how predators actually behave.
First of all, most predators are actually pretty bad hunters on paper, failing the vast majority of their hunts.
But there's a very good reason for that: they're well aware the odds aren't in their favour.
If you're a megafauna carnivore (i.e. eat things like deer or elk and that don't just pick on things 1/20th your size) your ability to eat is directly tied to your ability to win fights, and injuries of any kind have an outsized impact on your ability to win fights.
Spoilers: it's not a better idea because it means-tests based on "lost your job due to coronovirus-related reasons" which fails to encompass literally the entire self-employed gig economy who will never formally lose their jobs, because they don't technically have jobs.
It also doesn't encompass people who aren't fired but aren't getting hours or salary and people who weren't working but have had their support network collapse.
Worrying about rich people accidentally getting some extra cash is a waste of time and effort. The number of people who really have no use for an extra $1000 a month is dwarfed by the number of people it would literally save.