Saturday’s Forum
Steven L. Taylor
·
Saturday, November 2, 2024
·
46 comments
OTB relies on its readers to support it. Please consider helping by becoming a monthly contributor through Patreon or making a one-time contribution via PayPal. Thanks for your consideration.
About Steven L. Taylor
Steven L. Taylor is a Professor Emeritus of Political Science and former College of Arts and Sciences Dean. His main areas of expertise include parties, elections, and the institutional design of democracies. His most recent book is the co-authored
A Different Democracy: American Government in a 31-Country Perspective. He earned his Ph.D. from the University of Texas and his BA from the University of California, Irvine. He has been blogging since 2003 (originally at the now defunct Poliblog).
Follow Steven on
Twitter and/or
BlueSky.
So, it seems that “put her in front of nine rifles” has given way to “meet me at the bike racks after class.”
“ Liz Cheney would not fight. I will go with her, wherever she wants, and we will fight.”
https://x.com/acyn/status/1852473139227509129?s=46
Didn’t he say that he was the protector of women? That he’d protect them even if they didn’t want it?
Weird.
Also, this bit… I don’t even know what to say tbh.
https://x.com/projectlincoln/status/1852549296173830212?s=46
@Min:
Dementia = filters and inhibitions fail. It’s remarkable how rapidly it’s progressing now.
@Min:
Blowjob Don’s mental breakdown continues. Those internals must be terrible.
Tim Miller:
@charontwo:
It is. Yet, some ppl insist on saying “that’s Trump being Trump”
Ron Filipkowski:
Somehow, Donny Blowjob’s unhinged hate rally isn’t even a week old yet lol
He picked quite a time to have one of the worst weeks ever seen in a presidential campaign. Yikes.
You or I might think trying to repaint the lines on a major highway in the middle of the morning rush hour is a crazy idea, but that’s why we’ll never be genius enough to be schedulers for PennDOT
I just scrolled through 25 rows of YouTube as served up to me by the algorithm. There was not one thing that interested me. Not one. Much of it was loaded with crap I’ve been ignoring for weeks.
Does Amazon’s Prime Video do better? Nope. Netflix? Nope.
Google (YouTube) and Amazon and Netflix all have enormous amounts of data about me. I have no secrets from them. But with all that data they are completely unable to predict what might interest me. I’ve gone from spending a lot of time on YouTube, to spending about 75% less – thanks to the algorithm. I no longer even bother to glance at Prime or Netflix suggestions, I go there with a specific thing in mind and find it.
I have long suspected that the value of all this data is wildly inflated. Humans are not predictable, and they never will be. We are our DNA, our lived experience, modified by free will and random chance.
Humans are individual, unique decks of cards. Acquire all the data you like, you still won’t understand a human.
@Michael Reynolds: I think it’s more likely that the algorithmic calculations show nothing much to be a good match for you, so it’s trying the best bad matches. Because it will never tell you, “Nope, got nothing for you today”. Doing that is something that an AI could do, but the social media AIs will never do. That’s not aligned with corporate policy.
@Michael Reynolds:
A lot of people report that same problem, but I do not have it. I think it’s a function of who I selectively choose to follow, like, or repost. I am always conscious of not giving the algorithm any bad ideas.
(I am talking Twitter here, I actively avoid interacting with Facebook etc.)
@Michael Reynolds:..you still won’t understand a human.
Every time I see someone holding a leash in one hand and a plastic bag in the other hand ready to pick up the dog shit I have to believe the dog understands the human just fine.
@Jay L Gischer: @charontwo:
I think part of the issue is that I’m a ‘mile wide, inch deep’ guy, as opposed to going deep on one or two things. I always want something new that I don’t expect, or even know that I want. And if I don’t know what I want, the algorithm sure as hell doesn’t know.
Then there’s the fact that the algorithm is leaned on by efforts to push certain things and suppress others. And there’s the fact that the existence of the algorithm in itself is a factor which can subvert the utility of the algorithm.
But my basic point remains that algorithms, like AI, (and incidentally literary agents and entertainment executives) drive through life using only their rear-view mirrors. They can only see what has already happened, and any effort to extrapolate from that to predict the future, will fail. Or at least will perform no better than any average human, and much worse than an artist.
@Mister Bluster:
Hey, I get up of my own free will at 6:00 AM to walk the dogs, it has nothing to do with a Pug snout in my face. Also, long, intensely-focused dog stares do not cause me to feed them, that is coincidence.
@Jay L Gischer:
But that’s the point: I don’t want more of what I’ve already seen, I want something new and different from all the things I’ve already seen. Which means the algorithms will never be of any use to me.
This is a media trend I really dislike. Headlines on news sites which never actually tell you what the news story is about. Like, “This Issue No One is Talking About Could Really Swing the Election for Dems,” or “The Real Reason Republicans Confident in this Senate Race.” I mean, when did every headline. even at respectable news orgs, all just become clickbait.
I’ve also come up with a new term for unbalanced reporting by the media against Dems in lieu of whole Biden “garbage” nonsense. Let’s call it insanewashing, whereby the presso takes a completely anodyne quote from a Dem and then reinterprets it to it’s worst possible political interpretation and runs like wild with it in the nome of bilance. This of course is the opposite of the newly coined sanewashing that goes on with the press’s rewriting of the meaning and actual word salads, lies and gibberish from tfg into something it obviously wasn’t, but they report on in the most favorable light.
Yesterday, Spotify’s algorithm generated for me a custom playlist, Romantic Mix, that included Bruce Springsteen’s “Philadelphia,” one of the least romantic songs in existence.
@clarkontheweekend: “nome of bilance” – I actually kinda love this phrase for the effect you’re describing.
@ptfe:
“Bilance” = Bipartisan balance.
Music for the weekend. Two short works:
1) Flight of the Bumblebee
2) In the Hall of the Mountain King
@Michael Reynolds: Well, an inclination to variety is a thing that definitely varies between humans, and I’m quite prepared to credit you as being on the high end.
For comparison, as regards my choice in music, I don’t limit myself by style or genre. And an algorithm might try to do that.
AND, there is music I will pass by every time. Unless a person I trust recommends them. Perhaps multiple times.
Maybe you are the same? I think there are features and topics that will never appeal to you. If the algorithm was tuned to help you discover new things you might like, it could probably help you. After all, it knows pretty much everything you’ve seen.
This is a harder task, with a lower hit rate. That’s it’s nature. It isn’t an impossible one, though.
But it IS one that is contrary to corporate interests.
I would like to see some sort of human agency. For instance, a channel that takes a half hour or so to introduce some new music that the human making the channel likes. The legal/business elements to make this work aren’t in place, though. Making them work is beyond my pay grade.
If the AIs were set up to assist, rather than control, they could be quite useful, really.
@Michael Reynolds: What annoys me about all this time and money Netflix, Max, Spotify etc spend on “curating” my feed with machine learning is that it seems to come at the expense of powerful search tools, making it more difficult for me to navigate their library.
@clarkontheweekend: I think this is the root cause of the “But Her Emails!” press that we have, rather than deliberate malignancy. The public trains the media and what type of stories it wants, what we will click on.
@Michael Reynolds: “But that’s the point: I don’t want more of what I’ve already seen, I want something new and different from all the things I’ve already seen.”
Amazon’s the same way – they used to have a feature showing what people who bought the book you’re looking at also looked at or bought. Sometimes the topic was similar but there was a diversity of topics too. It introduced me to a number of authors whose books interested me. That feature tends not to be there anymore and instead Amazon generates what they think I’d like based on my browsing history. 99% of the time they’re wrong.
Speaking of AI curating what I see; I’ve already voted and mailed my ballot. Can I be exempted from seeing political ads? I have carefully considered the claims of all the candidates that their opponents are intent on destroying America.
@Jay L Gischer:
That’s an interesting thought. What if the AI reasoned, Michael was interested in the economics of Turkey for that one video, but he probably doesn’t want 19 more – he has a high desire for novelty and really hates being steered. Sooooo, how about a video on throat singing?
That would be helpful. An algorithm that says, Here’s some shit you’ve shown no previous interest in. That would be excellent.
ETA: Make that happen, computer people.
@Min:
He was very clearly saying that they would go to the Middle East, or France or somewhere and fight the locals, side by side, to see who is really a man. Apparently 9 locals.
Ok, he was semi-clearly saying… her fondness for wars she would never have to fight in was mentioned. Which was a little rich from Cadet Bonespurs.
Still insane, but not a threat.
I’d be happy with them going pheasant hunting with her father.
Oh, he’s escalating.
@Michael Reynolds:
My experience is the Youtube algorithm is structured around type-casting, not individual tendencies. I keep tabs on the Ukraine war, and when I view the Russian mil-blog channels it serves up a pile of RW nutjob Trumpist stuff. When I look at Lincoln Project or anything like that, it serves up the opposite.
@Michael Reynolds: Spotify did a much, much better job of playing random stuff after my playlists ended than Apple Music. So, it definitely varies.
YouTube is known to be one of the worst. If you feel like vomiting, get a fresh account, and watch two videos of children’s gymnastics, and then look at the results.
DO NOT DO THIS UNLESS YOU WANT TO FEEL SICK. And if you look at the comments, it’s all men linking to the “best” timestamps.
I haven’t heard about this in a while though, so maybe they fixed it. I will not check.
A good recommendations algorithm will have a certain percentage of random crap, notice that you don’t click on certain topics, and rotate the random crap.
@Jay L Gischer:
Short term very slightly, long term no. And it depends on whether you have to act to accept the recommendation (other people bought this, or video recommendations on YouTube’s home page) or whether you have to act to stop it (Spotify playing stuff after your playlist ends, YouTube automatically playing another video.)
The cost of the former is the space that would show you more of what you are more likely to like (but, there’s more than one recommendation, so that’s minor) versus reminding you that you should turn this off and go do something productive with your life.
The quality of recommendations has actually gone backwards in general.
@Not the IT Dept.: Walmart’s website had something similar, and when there wasn’t a really good match it would pick whatever the hot items were that week. So, as Martin Luther King day was approaching, right after a Planet of the Apes movie came out on DVD, they created many amazingly racist pairings.
I was at Amazon at the time, and the internal reaction was “oh, shit, we’re probably doing that too” and a quick scramble to mark the Planet of the Apes DVD as ineligible for cross-sell on black history books and videos.
@Not the IT Dept.:
Kindle still does this, and I see this feature on Amazon from my desktop occasionally. I haven’t found the feature valuable to any degree.
Just after trashing Kamala Harris in Michigan yesterday, Trump bragged repeatedly about his “beautiful white skin” and how he was too busy to get a tan on the beautiful beaches he owns, which are the most beautiful in the world.
According to Tim Alberta of The Atlantic, he wanted to call Biden “Retard Joe Biden” but was begged not to by his aides.
@MarkedMan: I just wish these sites allowed me to sort by descending IMDB score or something. I’m open to watching just about any genre so long as it’s not junk.
Re algorithmic recommendations, I’ve spoked about Youtube. The streaming services mostly try to push whatever is new or popular. Rarely what I may be interested in, unless it happens to be new or popular. And sometimes not even then.
For instance, I have The Diplomat on my Netflix list. I found out season 2 was near by browsing my list. And I found out it dropped on a Youtube channel reviewing new shows for the fall… Netflix near daily emails said nothing about it.
But I can see their point. Why would I be interested in a show just because I put it on my list and watched the whole first season?
Re: algorithms
I’m a pre-school teacher with middleish school sons and am experiencing a quasi mid-life crisis that has be doing things like saying “Maybe I was right to buy that Limp Bizkit album.”
The recs are WILD!
I feel two different types of admiration, based on how I see myself relative to the object of admiration.
If I think I can exist on the same number line as they are on, that’s one type. IOW, if I posit John LeCarré as a ten on the fiction writing scale, I don’t think I can get to ten, but if I work at it, I can get to eight. We exist in the same universe.
The other type is the kind of admiration I have for Eddie Van Halen or Simone Biles. Let’s make Eddie and Simone each tens in their respective fields. At no point in my life could I have, by dint of all the effort in the world, existed on the same number line as EVH or Biles. I’m made out of beef and whiskey, Simone Biles is made out of spring steel and helium. There is zero chance that I ever could have even been a one to their tens.
The first group I feel comfortable criticizing, because at least I can tell when they’re doing their best work, or phoning it in. The other group I basically have nothing to say about because they’re doing magic alien stuff.
The problem with that first thing, from my narrow perspective, is that everybody thinks they can be a writer. And they’re right. If you can write your name and home address you’re already on my number line. But the poor dumb bastards who have it worst are politicians. Everyone is on their number line, and a hell of a lot of them are well upstream from the pols. It amazes me anyone wants any job in politics. I have to suspect they either have serious psychological issues, or they’re just not very smart. It doesn’t even pay well, if you’re honest.
The final 2024 Ann Selzer/DMR Iowa poll
“Iowa Poll: Kamala Harris leapfrogs Donald Trump to take lead near Election Day.”
https://eu.desmoinesregister.com/story/news/politics/iowa-poll/2024/11/02/iowa-poll-kamala-harris-leads-donald-trump-2024-presidential-race/75354033007/
@Min: Hehe. How are our doomers doing today?
@DK:
Probably not happy 😀
I mean there was a certain “holy s–t” aspect to the 2014, 2016, and 2020 final Selzer polls, but this takes the cake.
@Min:
@DK:
Ann Seltzer is considered the gold standard of Political Pollsters.
Ann Selzer Is The Best Pollster In Politics
If I were Rick Scott, Tim Sheehy, Ted Cruz, and Bernie Moreno, I’d be shitting my pants and yelling at my staff.
Another poll came out yesterday, showing Ohio within the margin of error.
https://miamioh.edu/news/2024/11/miami-university-students-conduct-survey-of-ohio-voters.html
@EddieInCA:
Seen on Twitter:
“Male pollster: Tie game, we’ve adjusted for Trump
Male pollster: Tie game, we’ve adjusted for Trump
Male pollster: Tie game, we’ve adjusted for Trump
Male pollster: Tie game, we’ve adjusted for Trump
Ann Selzer: Have you boys heard of Dobbs?”
Welcome to Roevember.
@EddieInCA:
I wouldn’t want to be anywhere near El Felon when he heard the news.
Btw, we can probably expect Trump to sue NBC sometime soon. You know, since apparently Harris will be there today for the SNL episode.
https://x.com/ap/status/1852864756610113771?s=46
@DK: Pyrrhus of Epirus would be very well acquainted with the sort of victory Dobbs has proved to be for the GOP.
@DK:
Seen on Twitter
Ann Saltzer History polls vs Results
2022 Senate: R+12 (R+12)
2020 President: R+7 (R+8)
2020 Senate: R+4 (R+7)
2018 Governor: D+2 (R+3)
2016 President: R+7 (R+9)
2014 Senate: R+7 (R+8)
2012 President: D+5 (D+6)
I’m phonebanking this weekend for Harris/Walz – from Los Angeles – to voters in Michigan, PA, and WI… and TEXAS! Yes. Texas.
@Min:
I would. Schadenfreude, I’ve a feeling, is best when served live.
And it’s not hubris, so you’d be ok with the gods.
My weekend just keeps getting better and better….
Neo-Nazi Nick Fuentes turns on Trump after garbage comments: ‘Liberals are right’
Trump is now toxic to actual Nazis.
I just opened up a bottle of Weller.
Remember RoeVember. I just want it to be over, so we can start dealing with the next Trumpster fire.
I am exhausted. I’m certain most of you don’t know what this means, but we preg tested cattle today, and I got a 95% breedback rate.
I am so ready for a 3 month nap. 😛
@Jax: I don’t know what it means to you, but those are some very enthusiastic cows.
@EddieInCA: You know, I saw that Harris had done a rally in Houston with Beyonce, and I wondered if this was one of those endgame bluffs. Just kind of trick the opposition into spending money somewhere they didn’t need to?
Maybe not.
@Jay L Gischer:
I do not know, but my guess is “try again, use a different bull.”