#June 2021 contest | Explore Tumblr Posts and Blogs

arofili · 1 year

Text

Tolkien Fandom Event Calendar

Recently I’ve received some asks about events/weeks in the Tolkien fandom, so I thought I’d compile a list of those that I know about. This is not exhaustive, and dates are subject to change by the organizers of these events!

Other blogs you can check out are @tolkieneventsblog and @tolkienfandomevents, though I’m not sure how active those are. The @silmarillionwritersguild Discord also has a channel dedicated to signal boosts for all sorts of Tolkien-related & general fandom happenings, which is another excellent way to keep up with fandom goings-on.

Want to run your own event? Here’s some of my tips!

If your event is not on here and you’d like it to be, let me know and I can add it :) Note: I will only add events that have announced dates!

JANUARY Screw Yule My Slashy Valentine @myslashyvalentine — work time Lord of the Rings Secret Santa @lotr-sesa — reveals Thorin’s Spring Forge @thorinsspringforge — signups Second Age Week @secondageweek

FEBRUARY Hidden Paths My Slashy Valentine — reveals Thorin’s Spring Forge — claims Maedhros and Maglor Week @maedhrosmaglorweek

MARCH Back to Middle-earth Month @spring-into-arda Thorin’s Spring Forge — work time Fëanorian Week Fun with Fanon Fest Round 1 @funwithfanon

APRIL Tolkien Reverse Summer Bang @tolkienrsb — signups Silm Remix @tolkienremix — signups & assignments Thorin’s Spring Forge — reveals Aralas Week @aralas-week Barduil Month @bi-widower-dads All of Arda is Autistic @all-of-arda-is-autistic F3: Focus on Friendship & Family, Phase I @spring-into-arda

MAY Tolkien Reverse Summer Bang — claims Silm Remix — reveals Aspec Arda Week @aspecardaweek Angbang Week @angbangweek Gondolin Week @gondolinweek F3: Focus on Friendship & Family, Phase II

JUNE Tolkien Reverse Summer Bang — work time Scribbles and Drabbles @fall-for-tolkien — signups Tolkien Ekphrasis Week @tolkienekphrasisweek F3: Focus on Friendship & Family, Phase III

JULY Tolkien Reverse Summer Bang — work time Scribbles and Drabbles — claims Tolkien Gen Week @tolkiengenweek LotR Ladies Week @lotrladiessource Tolkien Appreciation Week @tolkienweek Tolkien Latin American & Caribbean Week @tolkienlatamandcaribbeanweek

AUGUST Tolkien Reverse Summer Bang — deadlines Scribbles and Drabbles — art reveals Innumerable Stars Exchange @innumerable-stars — nominations & signups Tolkien of Colour Week @tolkienofcolourweek Silvergifting Week @silvergiftingweek Tolkien OC Week @tolkienocweek

SEPTEMBER Tolkien Reverse Summer Bang — reveals Scribbles and Drabbles — work time Innumerable Stars Exchange — signups & assignments Sindar Week @sindarweek Dor Cúarthol Week @dorcuartholweek Finwëan Ladies Week @finweanladiesweek

OCTOBER Innumerable Stars Exchange — reveals Scribbles and Drabbles — work time Half-elven Week @halfelvenweek

NOVEMBER Tolkien Secret Santa @officialtolkiensecretsanta — signups & assignments Scribbles and Drabbles — fic reveals Nolofinwean Week @nolofinweanweek

DECEMBER Tolkien Secret Santa — advent calendar & reveals My Slashy Valentine @myslashyvalentine — signups & assignments Lord of the Rings Secret Santa — claims Khazad Week @khazadweek

MONTHLY EVENTS: These events have prompts/challenges occurring every month. Teitho Contest Tolkien Short Fanworks Silmarillion Writers’ Guild @silmarillionwritersguild

(this list was last updated 5/4/23)

LEGACY EVENTS: These events used to occur, but have not happened within the last year. Arda Needs More Pride @ardaneedsmorepride (bimonthly; last run 2020) Kiliel Week @kilielweek (timing variable; last run 2021) @oneringnet monthly events (last run 2021) Atani Week @ataniweek (January; last run 2021) Legendarium Ladies April @legendariumladiesapril (April; last run 2020) Gates of Summer Exchange @gatesofsummerexchange (May-June, last run 2022) Tolkien South Asian Week, run by @arwenindomiel (June; last run 2022) Arafinwëan Week @arafinweanweek (July; last run 2019) Fëanturi Week (August; last run 2019; no official blog and the creator has deactivated) Imladrim Week @imladrimweek (November; last run 2019) Doriath Week @doriathweek (November; last run 2020) Tolkien Family Week @tolkienfamilyweek (November; last run 2021)

#tolkien #tolkien fandom #the silm #the silmarillion #silm fandom #silmarillion #silm #lotr #lord of the rings #lotr fandom #the hobbit #hobbit fandom #tefain nin #tolkien fandom event calendar #tolkien fandom events #save

351 notes · View notes

jokeroutsubs · 7 months

Text

Käärijä's ESC bestie Bojan's band performs in Finland - "There's just something between us"

The Slovenian band Joker Out drove the audience wild on Wednesday at Tavastia in Helsinki. After the Eurovision Song Contest, the band has become a favourite especially among teenagers.

Video caption: Slovenian Joker Out went wild at Tavastia

Ville Vedenpää

When Käärijä and Bojan Cvjetićanin, the lead singer of the Slovenian band Joker Out, meet again at Tavastia's backstage, there is no end to talking and laughing. They have a special connection, which was formed in the spring during the Eurovision Song Contest.

The duo's friendship or brotherly love, bromance, did not go unnoticed by the public and it became a phenomenon on social media.

The last time the social media crowd went wild was in June, when Bojan visited Finland and went to Käärijä's gig at Tavastia.

Now Bojan's band is performing at the same venue and Käärijä is the guest star.

How does it feel to meet again?

"Terrible!" Bojan exclaims, and the duo laughs.

"I was so looking forward to meeting Bojan again. I remember the first time we met in Madrid. I didn't talk much, so I just watched and smiled," says Käärijä.

"Now he talks. Back then, he was just eating dinner", Bojan laughs.

"There's just something between us. I felt it when we first met. This is not only Bojan but also Joker Out", Käärijä specifies.

The duo continue to talk like Chip 'n' Dale, about how they are like brothers, but maybe more. The joking that started in the spring continues, with perhaps only the duo themselves keeping up with it.

Photo caption: The Joker Out band at Tavastia before their first performance in Finland. Photo by Jukka Lehto

Joker Out got the fans in a frenzy

Joker Out played two shows at Tavastia in Helsinki on Wednesday. The first of the two shows had no age limit and was sold out. Before both shows, queues outside Tavastia were hundreds of metres long.

In addition to Helsinki, the Slovenian band will also perform in Turku and Tampere.

Joker Out represented Slovenia in this year's Eurovision Song Contest. The band came 21st in the final with the song Carpe Diem.

Ida Oljemark from Sipoo has been listening to Joker Out's music since the ESC. She is interested in the band's music, even though she doesn't understand everything. She has also learned some Slovenian through the songs.

"We've spent weeks painting lighters and key fobs and everything. We've joked that we'll remain on the floor here. We're not going to handle this."

Photo caption: Ida Oljemark from Sipoo has been fascinated to see how the friendship between Käärijä and Joker Out's Bojan was born over the language barrier. Photo by Jukka Lehto

Fans had come from all over the world to see Joker Out in Finland, from as far away as Mexico.

Alina Badzian had come to the gig from Ukraine, as the band cannot perform there.

"They became a safe haven for me through the Eurovision Song Contest. Each of their songs has a special meaning for me."

Photo caption: Ukrainian Alina Badzian had written a cardboard note for the Joker Out gig, in which she wants to steal guitarist Jan's guitar pick since he has stolen her heart. Photo by Jukka Lehto

No more Eurovision

Indie rock band Joker Out was formed in 2016. The band consists of Bojan Cvjetićanin (vocals), Kris Guštin (guitar), Jan Peteh (guitar), Jure Maček (drums) and Nace Jordan (bass).

The band has released two albums Umazane misli (2021), and Demoni (2022).

Joker Out describes their music as Shagadelic Rock n' Roll. The name refers to the Austin Powers movies and the protagonist's comment about feeling good.

"Our style is very authentic. We play energetically on stage and we hope it affects the audience," says singer Bojan Cvjetićanin.

The Eurovision Song Contest took the band's popularity to a whole new level. Now they perform all over Europe and fans sing along in Slovenian.

The band members receive many gifts from their fans and even proposals of marriage.

"If the music speaks to the heart, we can easily get married," laughs the singer.

The Eurovision Song Contest was a unique and educational experience for Joker Out. However, they will not be entering the competition a second time.

"We learned a lot and gained even more. You can only experience it once and then leave room for others."

Photo caption: "It's amazing how people sing in Slovenian at gigs. It seems even more difficult than Finnish," said Joker Out at the Tavastia gigs. Photo by Jukka Lehto

Special relationship with Finland

Joker Out will start recording a new album next year and then head off to summer festivals. They hope to play on festivals in Finland as well.

Finland has become a special place for the band through Käärijä. The band dreams of a trip to Lapland, where they could go ice fishing and see the northern lights.

Photo caption: Joker Out says that they have no plans to make music with Käärijä, at least not at this stage. Photo by Jukka Lehto

The band feels that Finns are similar to Slovenians when it comes to partying, for example.

However, there is one difference.

"The Finns have their own sense of humour. You never know whether they're joking at your expense or whether they're serious. They do it in such a raw way that it is scary, even," the band members laugh.

Translation cr: @niini5

#joker out #bojan cvjetićanin #bojan cvjeticanin #jan peteh #nace jordan #jure maček #jure macek #kris guštin #kris gustin #käärijä #kaarija #jokeroutsubs

79 notes · View notes

slimysalamander1114 · 3 months

Text

Redraw of Whiteout from a pride art contest ages ago. (Ages being June 2021)

Well, I kinda went overboard lol

Total time 2h37m

#digital art #art #wof #wings of fire #whiteout wof #whiteout #ibispaintx #pan #redraw #improvement #fanart

38 notes · View notes

biggbosstruefan · 2 years

Text

Bigg Boss 16 Contestants Name List 2022 with Photos

Bigg Boss 15 Contestants Name List 2021, Salman Khan show Bigg Boss 15 Expected Contestants, Common Man List Out, Celebrity Names, Big Boss 15 Contestants List on Auditionformdates.in | Bigg Boss Season 15 Contestants List Coming Soon

Here below you’ll get all the Roumered contestants of Bigg Boss Season 16. So quickly check this exclusive list of Bigg Boss 15 Contestant Names & Photos.

Hello Friends we bring latest News for all BBians. Are you looking for Bigg Boss 16 Contestants? Are you a huge fan of Bigg Boss TV show? Are you egarly waiting to know who are the contestants of Bigg Boss 2022? If yes, then you are at the right place as here I will tell you about all the contestants who will join Bigg Boss this year.

Bigg Boss 16 Contestants Name

Bigg Boss 16 Contestants list has revealed. Check the Celebrity and Commoners contestants’ details. Bigg Boss is one of the most popular and controversial shows of India. Bigg Boss 15, finished in February 2021. Thereafter, its auditions for the 16th season started in May/June 2022. From those auditions, 10 or 12 People were selected as Bigg Boss 16 Contestants. Also, Celebrity contestants have been finalized by the Cast. Cast had approached some celebrity contestants for Bigg Boss 16. Out of which 14 celebs signed the show ‘BB16’. You can see their all details here.

#bb16

915 notes · View notes

mythicalviper-fr · 10 months

Text

my festival submissions from June 2021 - June 2023! two years, 32 designs (not including recolors), 14 wins, 7 losses. I'm blown away by how much my art has improved, and I'm very grateful for the artists that have mentored me along the way ^^ especially august, who competed with me throughout and pushed me to be a better artist.

unfortunately, Brightshine 2023 is probably my last hurrah. I will still resubmit my fest rejects, but I was diagnosed with tendonitis in both arms last week (work-related), leaving my right arm in a cast. I don't know if or when I might recover, but considering the severity of the diagnosis I was given I don't want to risk injuring myself further :/

I am currently writing up something on festival skin contests, mostly with tips that I hope will help and encourage other artists to enter.

until then, happy Brightshine everyone! here's to another (the best and the snek-est) fest cycle!

#flight rising skins #flight rising #flightrising #fr skins and accents #flight rising art #fr #drink water and take breaks everyone ^^"#brightshine jubilee

91 notes · View notes

beardedmrbean · 5 days

Text

A former Northern California teacher pleaded no contest Monday to having sex with an eighth-grade student on his graduation day in a classroom, prosecutors said.

Michelle Christine Solis, 46, pleaded no contest to having unlawful sex with a minor and distributing harmful material to the 14-year-old boy, according to the Butte County District Attorney’s Office. The investigation began after parents learned their children had inappropriate pictures of Solis, a former employee of the Gridley Unified School District.

Solis, who had taught for more than 20 years, was teaching the 14-year-old boy in June 2021 when she “friended” him on Instagram, prosecutors said. They began talking online in the weeks leading up to his Sycamore Middle School graduation.

Prosecutors said investigators found evidence that Solis sent four explicit pictures of herself and then had sex with him inside a locked classroom on the day of the student’s eighth-grade graduation.

Gridley police began examining Solis’ conduct after rumors of their relationship began spreading two years after the illegal contact, prosecutors said.

Investigators reached out to the boy and found Solis’ explicit pictures and their communications on his phone. She was arrested in November and entered her plea Monday in Butte Superior Court during a preliminary hearing.

Solis must register as a sex offender as part of her sentence, prosecutors said. Her sentencing is scheduled for June 6; she faces four years in prison.

Gridley is a city of 7,200 residents in Butte County between Chico and Yuba City.

#nunyas news

14 notes · View notes

golbrocklovely · 2 months

Text

haven't done one of these in months, so i thought it was the perfect time to bring this back

here’s some of colby’s tweets from 2021.

if it’s bold and italicized, it’s someone’s tweet to him.

if it’s in (), that’s just me commenting lol

added bonus: if they have a * next to them, that means it’s been deleted

~~~~~~~~~~~

April 1 - just wanna make things right

April 4 - happy easter i’m hungover goodnight ?

@/mannymua733: 30 flirty and thriving

happy birthday!!! @/mannymua733: thank you my love!

fan: all manny wants from colby is this (video of colby grinding)

@mannymua733: that’s all i’m asking for colby if this is your b day wish i gotchu

hey if any of you guys have a semi truck could you please run me over with it thanks

@/deefizzy: Can i join the more the merrier !

April 7 - watch the time go by

there’s an angry squirrel in our backyard that tries to kill me every night

April 9 - i can admit that i have big commitment issues when it comes to relationships

April 11 - to the two girls i just met at chipotle … you made my day thank you for lifting my spirits i’m so grateful

April 13 - sometimes life just feels like a big dream to me it’s so hard to explain.

April 15 - who’s your comfort person?

April 19 - it’s hard to find things that excite me as much as they use to and i find that kinda sad. i miss being stoked about every little thing. wanna get that back

learning to let go

no matter what changes in the future .. just know i’ll remember the past

April 20 - i know you don’t believe it but i’d do anything for ya

April 27 - @/ohkailno: when im vaccinated im coming for u @/ColbyBrock

😈

May 2 - i got a PCR test yesterday that literally felt like they were trying to tickle the back of my Cranium

May 11 - fan: So you just gonna leave us hanging like that @/ColbyBrock (photo of him with the caption "it's time")

😏🖤

May 13 - @/katstuartmusic: should i make colby bald

May 17 - i hope you’re happy today i love you

May 23 - someone broke into our rental car in San Fransisco and they stole everything. but i’m most upset about my journal i’ve been writing in daily for the past 6 months… all my emotions, memories, details of life just gone. my 2021 story i can’t get back .. i’m heartbroken

@/lukewaale: duuude what :( that's messed up man, i'm sorry it’s all good man 💔 thanks for sayin something. gotta look at the bright side .. everything’s replaceable

May 28 - you can’t miss what you forget

(it's so clear to me now that this man was clearly going thru so much. this is when things really changed for him and how he interacted with us. it's sad to look back on this now)

June 19 - feels good to be back.

June 28 - fan: I FORGOT @/ColbyBrock IS AN UNUS ANNUS FAN SIR EHAT WAS UR FAV VIDEO

the one where they have a staring contest ! hahahah

#colby brock #cb tweets #text #personal

8 notes · View notes

pronoun-fucker · 2 years

Text

“The Flash” star Ezra Miller says they are seeking treatment for “complex mental health issues.”

In a statement provided to Variety by a representative of the actor, Miller broke their silence about the troubling behavior that they have exhibited in recent years, which has led to a series of legal issues and assault and abuse allegations. Miller also apologized for their actions.

“Having recently gone through a time of intense crisis, I now understand that I am suffering complex mental health issues and have begun ongoing treatment,” Miller says. “I want to apologize to everyone that I have alarmed and upset with my past behavior. I am committed to doing the necessary work to get back to a healthy, safe and productive stage in my life.”

The embattled actor, who is non-binary and uses they and them pronouns, played the DC Comics character Barry Allen in 2017’s “Justice League” and its 2021 counterpart, “Zack Snyder’s Justice League.” But their role in the comic book franchise has been overshadowed by their off-screen behavior, which has alarmed friends and associates of the actor, as well as Warner Bros., the studio behind “The Flash.”

Most recently, Miller was charged with felony burglary in Vermont, after state police investigated an incident involving several bottles of alcohol being taken from a residence while the homeowners were not present. The actor made headlines in 2020 after a recording surfaced in which they appeared to be choking a woman outside of a bar in Iceland. No charges were filed. The actor has been arrested twice in Hawaii this year, once for disorderly conduct and harassment. In the Hawaii incidents, Miller pled no contest to a single count of misdemeanor disorderly conduct and paid a $500 fine and $30 in court costs. The harassment charge was dismissed.

The series of scandals has become a snowballing public relations issue, not only for Miller, but also for Warner Bros. Discovery, which has employed the actor across two of its cornerstone franchises — DC Comics films and the “Harry Potter” spinoff series “Fantastic Beasts.” A source close to the studio says that Warner Bros. supports Miller’s decision to seek professional help.

“The Flash” is set to open in theaters on June 23, 2023 and reportedly serves as a key entry in the studio’s DC Comics films, featuring appearances by Ben Affleck and Michael Keaton as Batman.

Link | Archived Link

#I mean #WB literally just got him to apologise so that they can still release The Flash #so I don’t believe this apology is sincere in the slightest #it’s also a few years too late tbh #radical feminism #radfem #gender critical #misogyny #trans activism #transgender #trans #nonbinary #Ezra Miller #the flash

164 notes · View notes

choices-binglebonkus · 1 year

Text

Every 2022 Choices Release – Rated on a scale of 1-10

Several people have said that 2022 has been the best year for Choices since 2018. And…as sad as that is, it’s pretty true. Like 2020, 2022’s releases had high highs and low lows. As I’ve said once before, the good books were awesome and the bad ones were really, really, REALLY bad. This post is going to be covering every 2022 book release and rating them on a scale of 1-10 because I haven’t done that on this blog and felt like spicing things up. It’s the new year, bitches!

Quick disclaimer: to qualify for this list, a book must have been initially released to wide audiences (non-VIPs) in 2022, because we don’t do that VIP shit here. So, for example, Wake the Dead was released on December 29, 2021, so it doesn’t qualify for this list even though it ended in 2022. With that out of the way, let’s dive right in.

Ms. Match: January 28, 2022 – April 29, 2022

Ms. Match’s story itself was not terrible, but the general skeleton of the plot is super overdone. I am sick to death of the plots centered around rising to the top of the competition and blowing everyone else away, and that’s pretty much exactly what Ms. Match does. The idea behind it is cute, though: we join a matchmaking company and compete in matchmaking contests and challenges to determine who the next CEO will be. My biggest complaint is that the love interest is…not great. The Rival Matchmaker is far from the worst love interest, but my god, they’re so cocky, full of themselves, and overall annoying as hell, not to mention totally uninteresting. Quite frankly? The cutest couple in the book, as unexpected as it was for me, was Veronica and our character’s dad. 4/10, not a terrible book, but it was a chore to play through at times.

Surrender book 1: February 11, 2022 – June 10, 2022

There are a lot of bad books in Choices, but Surrender is definitely one of the worst. We play as a vulnerable divorce-seeking woman whose spouse has cheated on her, which got her into BDSM. As soon as I saw that that was going to be the plot, it was all over. It was not a great look for the BDSM community, which is already stigmatized enough as is. But it gets worse. Our love interest is Reagan Thorne, our billionaire lawyer who’s also into BDSM and agrees to dom for the MC and teach her about BDSM even though she’s a client, which already introduces an imbalance of power. Barring the shitty plot, terrible love interest, and horrible misrepresentations of BDSM, the dialogue is all sorts of fucked in the book. Reagan utters some of the cringiest lines in the entire app and calls us such embarrassingly bad nicknames like…ugh, little lamb. 1/10, horrible book, horrible story, horrible love interest, horrible everything. Some of the outfits were cute. That’s it.

Crimes of Passion book 1: April 6, 2022 – July 20, 2022

Crimes of Passion is one of the best books we’ve seen in a while. Sort of like Most Wanted, it has us teaming up with a love interest (in this case, Trystan Thorne) to investigate and stop a string of murders tied to a serial killer. Unlike Most Wanted, however, the murders in this book are much, MUCH grislier and more visceral, and are tied to a cult as we come to find out. The plot was very gripping and every clue we uncovered was exhilarating. I cared about pretty much all of our friends/coworkers: Mafalda, Ruby, Luke, and Trystan. I wasn’t totally sold on Trystan as a love interest, though, and I’m still not. 9/10, really good story, crazy fucking shit, love it. One point off for Trystan because I’m not really into the whole cocky/flirty and stubborn partner pairing, and I think we should’ve been able to romance Ruby and Luke.

Untameable book 1: April 15, 2022 – August 5, 2022

We return from college to help out on the family ranch which has been barely kept afloat by our controlling misogynist of a brother. We spend the whole book sneaking around with Kit Jackson, our bland-ass cowboy/cowgirl love interest because we don’t want to upset our brother. The ranch is on the brink of failure and we also have to deal with that too, but we’ve already had this same plot a whole bunch in other books, so it’s not super noteworthy. That’s it, that’s the book. 1.5/10, no interesting characters, plot is uninspired, dialogue is SUPER cringey, and goddamn it, stop reusing cowboy shit, PB! You’re never going to make it happen!

The Princess Swap: May 13, 2022 - September 2, 2022

The premise of this story is that our character studies abroad for a semester in the kingdom of Monterre, where she accidentally stumbles upon another girl, the princess, who looks just like her. Spoiler alert: they’re twin sisters and don’t know it until much later in the book. The two switch places The Prince and the Pauper style: the princess takes on the role of college student and sorority girl while our character takes on the role of princess. Both grapple with challenges unique to each other’s lives and navigate them as best they can. Yes, the story was cheesy as hell, but you know what? It was a really pleasant read. I liked Devin all right, but only as a friend for the princess, and I didn’t care for Clarke at all. Otherwise, The Princess Swap was a lighthearted, feel-good book. 6.5/10, by no means is this book amazing, but I’m a sucker for princess stories, and like I said, the book was pretty light and enjoyable.

The Cursed Heart book 1: June 8, 2022 – September 14, 2022

Let me preface this by saying that I did not like Kieran at all. They were ugly, creepy, and possessive, and yes, I’m aware they eventually change for the better, but they really left a bad taste in my mouth. The plot was centered around us being claimed by Kieran and eventually trying to find their missing heart pieces to cure them of their beastly form and general hollowness. That being said, I love the rest of the book. The outfits, backgrounds, character designs, and music are all gorgeous. No joke, The Cursed Heart, in my opinion, is the most aesthetically beautiful book in the app. Not to mention, we meet Leaf, who is absolutely precious. 5.5/10, the rating would be MUCH higher, but Kieran really takes away from the book since the plot is centered entirely around them.

The Nanny Affair book 3: June 24, 2022 – November 4, 2022

The only good things about this book were Carter, the twins, and the fact that this was the last book in this wretched series. We deal with Sam’s former wife coming back from the dead and contend with forced, trashy interpersonal conflict with her while we plan our wedding. Sam and the MC’s relationship continues to be toxic as hell, unpleasant to watch, and impossible to root for. Even worse, Jenny and Aditya start their own affair because Sam and the MC have enabled them, and unsurprisingly, we’re forced to continue enabling them. I can’t even do the horrid quality of this book justice so I’m not even going to try. 0/10, yes I know it’s a 1-10 rating list but fuck it, this book is just that bad. As I said, the only redeeming qualities of the book are Carter and the twins, but not even they can get this book any points. Speaking from experience, falling into a hole is more fulfilling than engaging with this content. This book and series as a whole are so transcendently bad that if they were an actual real-life book, it would be a Worstseller, and I’m so happy this two-year-long nightmare is finally over.

Immortal Desires book 1: July 6, 2022 – October 12, 2022

Immortal Desires is kind of a weird one for me. On one hand, the beginning to early middle is so slow and such a huge chore to play through, but on the other hand, the rest of the book plus Gabe and Cas are so awesome. The premise of a vampire serial killer and cult is honestly kinda metal as fuck, but Gabe and Cas are where the book REALLY shines. They’re each other’s perfect foils and I love interacting with them and seeing them interact with each other. Plus, my parrot loves M!Gabe. And, full disclosure, Gabe and Cas were the only love interests I actively wanted to pursue since Tom Sato from It Lives Beneath. 8.5/10, the story’s pacing was kind of fucked but Gabe and Cas are priceless.

Slow Burn: August 19, 2022 – November 25, 2022

Not much to say about this one. Generic and repetitive contest/competition plot, bland MC, uncompelling Big Bad. I didn’t care about Julian/Julia, and Everett/Yvette had absolutely ZERO chill and always got on my nerves. 3/10, super boring and uninspired, Chef Flynt Stop Losing Your Fucking Temper Challenge 2022.

Murder at Homecoming: September 14, 2022 – December 21, 2022

This one really got my attention when it was announced: our character investigates a murder at homecoming, but I bet you couldn’t guess that from the title. Our character is sort of a busybody and they’re absolutely not perfect in any way, but I truly do think they’re the best MC we’ve gotten in a very, very, VERY long time. Also, they’re named after Valentine from The Two Gentlemen of Verona, and I am a shameless sucker for Shakespeare. I also like all the love interests, and I love the non-binary rep we got. The love story between Gabbie Navarro and Joanna Morgan was incredibly tragic and really made me stop and think for a while. But unfortunately, as is the case with most modern Choices books, the ending was very…lackluster, to say the least. 8/10, I love a good murder mystery and the high school setting actually made it considerably more interesting. The ending was a letdown, though.

Laws of Attraction book 2: September 30, 2022 – present

I was originally very worried about this book, because the cliffhanger at the end of book one made it seem like it was going to be an us vs. Martin book. In a way, it is, but Pixelberry really threw us a curveball when we, Aislinn, and Gabe left McGraw-Byrne and formed our own firm. We spend the book uncovering a conservatorship conspiracy which runs all the way up to the mayoral candidate of New York City. I’ve always loved the MC, and they only get better in this book. My only complaint is Joaquin. As someone who has repeatedly hooked up with him in diamond scenes, he is creepy as hell to people who aren’t interested in him. Otherwise, I’ve thoroughly enjoyed the book so far, and though it still has one chapter to go, my rating isn’t going to change unless they REALLY fuck it up somehow. 9/10, likable cast of characters, interesting plot and mystery, badass MC, Sorcha fucking Flannery.

The Phantom Agent: October 19, 2022 – present

Not much to say about this one either. The Phantom Agent is basically an amalgamation of Most Wanted and Crimes of Passion. Agent Gray is an incredibly dull love interest and I can’t stand their dynamic with the MC because it’s that super boring cocky/flirty and stubborn partner pairing again. I also can’t stand the MC either because they’re so goddamn annoying. I find myself very bored and unimpressed as I progress through the book and at this point, nothing can save it for me. 3/10, everything is boring. That is all.

Surrender book 2: December 28, 2022 – present

I hate that we have to have TWO Surrender books on this list, but this one unfortunately barely meets the date cutoff. Solely based on the two chapters I’ve read so far, this book is shaping up to be just as awful as the first. 1.5/10, all the same reasons as the first book, but that extra half a point is because I think this book has the potential to improve on the original material. Because let’s be honest here, I don’t really think they could make the book worse.

86 notes · View notes

steampaul · 4 months

Text

i've started a little series of weekly posts documenting every MoC i made (digitally) over the last few years:

Week 8: 2021

Bonsai, Contests, Old themes and Pride

brickset had another contest where you could enter mods of 10281 bonsai tree. i made two that i thought would look interesting as more alien trees and one using bionicle kanohi (plus jaller)

there was also a category for that contest where you'd make a custom bonsai. i didn't have to many ideas other than "make small" and "abandoned". i don't really like the way those trees look.

the great cloud gardens was a game that i played a lot during that time, so i decided to make a bonsai-like MoC based on the city levels of the game. i still love this build and i hope that i'll be able to build it irl some time in the future.

these are also more directly based on cloud gardens. they depict the levels where you get the seeds for the Wisteria, the Monstera, the opuntia, the bryophyta and the pothos. (i kinda had to guess the names based on the pixelated names on the cards and the closest names i could find for plants that are similar looking to the ones ingame)

i never posted these anywhere because i wanted to finish making all of the plants before posting them. i never finished.

there was another contest on lego ideas where you could build a new GWP with the theme "space" i originally built it as an Ice Planet style-vehicle, but then i looked more into the lore of Ice Planet and found out that they specialized in Rocket science and something about satellites, i think?

this made me redesign this MoC as a futuron vehicle because that subtheme had a focus on transportation and more civilian activities.

i had the idea to make more modern Rock Raiders MoCs based on interesting modern Construction vehicles, i think i kinda overdid it with the amount of colours present on these MoCs, but i'll probably revisit this idea some time in the future.

another ideas contest. this time for a summer themed GWP. i based this on some north sea beach memories i made as a kid. i really like the sunny atmosphere i created with this MoC.

i also played narita boy that year and wanted to make a MoC based on that amazing game, but this never went anywhere. this was also the first (and probably last) time i've used PartDesigner to make a custom part (the sword) and some cutom prints.

LEGO released the Everyone is Awesome set in June 2021, so ideas had an activity to post some Pride Month inspired Creations. i originally made the Stand For the Flag(s) build that i made several variations of and now turned into a .gif, but i didn't post it because i really disliked the colours i used and couldn't think of better ones. instead i posted the Building Bridges MoC. i stimm like it and i think it would make a nice desk decorations similar to the EiA set.

anyone remember Atlantis? i had the idea to build some place to hold all eight Treasure Keys that where released in both waves of that theme. i think the theming is a bit off with this MoC, but i still kinda like the colours i used in it.

first week last week next week

#lego #lego moc #lego bonsai #cloud gardens #narita boy #atlantis #rock raiders #ice planet 2002 #futuron #pride flags

7 notes · View notes

cadybear420 · 2 months

Text

Cadybear's Reviews Brief Thoughts- America's Most Eligible

Welcome to the twelfth official Cadybear's Reviews Brief Thoughts post! Today I'll be talking about America's Most Eligible, which I have ranked on the "Gold Tier" at 8 stars out of a possible 10. My last and only playthrough of this series was back in May-June 2021. I had started another playthrough in July 2023, but it's been put on hiatus.

This one was pretty fun.

I think it did a pretty good job of getting you into the crazy atmosphere of reality TV shows. Whether you love it or hate it, you gotta admit it’s quite filled with some pretty wild drama. I also loved being able to establish our own confession cam persona for the MC– MCs that give the players agency in their character are always a win for me.

My only real problem is that besides maybe Jen, the LIs were mostly forgettable and flat in personality. Let me clap Carter’s cheeks instead please!

That and, a lot of the male MC outfits were painfully mid. Like, they seem to really enjoy dressing male MC in basic casual and/or toilet swirl color scheme outfits, even though this is a reality TV show where ideally you’d want to grab people’s attention and make yourself look stylish. I don’t like the outfits that f!MC gets as they’re not my style, but at least I can buy that they’d be impactful on this kind of show. But for m!MC… why do they want him to do a photoshoot in a big winter coat that looks like the same coat worn by PETA’s version of Ghetsis from Pokemon? Why are they saying this is his last chance to “show off his hot bod” and then giving him an outfit with a big coat that looks like an off-brand Doctor Who cosplay?

I know it’s pretty par for the course for premium outfit setup scenes to be obviously written with f!MC’s outfits in mind, but it’s far more frequent in this series, and far more glaring in this kind of story too because this is a romance-based reality TV show. Ideally, the hardcore dedicated contestants would want to dress in ways that grab people’s attention and make themselves seem more desirable. And I can assure you, no one’s underwear is gonna be flying off at giant peepeepoopoo-colored winter coats.

Speaking of m!MC writing coded for f!MC, this is kind of an elephant in the room, but I don’t think this series as a whole is quite as “female coded” as people have made it out to be. Granted it has the most gender errors I’ve ever seen in a GOC Choices story, like some of the outfit descriptions or that “he brushes a stray strand of hair from your face” line. But a lot of the other stuff like the shoulder carrying stuff or MC’s sometimes gushy personality or the walking down the aisle stuff, more or less boils down to lazy coding in general rather than lazy coding specifically for a male MC. I mean, that tends to be the case for the majority of the “male MC is female coded” complaints I see, but since this book in particular has gotten a lot of those sorts of complaints, I figured it’s worth bringing up.

So yeah. Overall this one was pretty fun. Would love to play this again and see the different kinds of outcomes I can have.

#choices stories you play #choices #choices stories we play fandom #choices game #choices stories we play #america's most eligible #choices america's most eligible #ame #choices ame #cadybear's brief thoughts

5 notes · View notes

black-arcana · 1 month

Text

LZZY HALE: 'If It Weren't For SKID ROW In My Bones, I Would Not Be The Rocker I Am Today'

HALESTORM's Lzzy Hale says that "there will be blood left on the stage" after she performs with SKID ROW later this spring.

The 40-year-old singer, who has fronted her band HALESTORM for more than two and a half decades, will handle the vocal duties for SKID ROW for four shows only after the group's fourth frontman since Sebastian Bach's departure — "Swedish Idol" contestant Erik Grönwall — quit the band to focus on his health.

Earlier today (Friday, March 29),Lzzy shared an admat for her four appearances with SKID ROW and she wrote in an accompanying message: "This is totally wild! I can't believe my VHS audition tape finally made it in the mail! (I kid, I kid)

"I started listening SKID ROW when I was a teenager growing up on the east coast in central Pennsylvania. They were one of the only bands at that time that truly bridged the gap for my love of 80s big choruses and riffs, but had the same poignant dirt and grime that was seeping into the cracks of my mind during the 90s. By the early 00s HALESTORM was well coming into its own, inspired by these powerful songs that helped me unlock a door within myself as a young musician. I can say without a second thought that if it weren't for SKID ROW in my bones, I would not be the rocker I am today.

"And this summer, my world turns full circle. Because not only am I able to share the stage as acting front person for this integral band, but even more importantly, I get to call these boys my chosen family. No matter how much we both tour, or how much time has gone by…we light up when we see each other. They have seen me through my rollercoaster ride of ups and downs, all while cheering me on. Time, space, history, age…are all relative, irrelevant and elastic cuz we are all just astronauts cut from the same cloth, worshipping the same magic that is this music. I'm so grateful for their support of me in my career, and as a fellow human.

"Love you guys! There will be blood left on the stage!"

In the comments section below Lzzy's post, SKID ROW bassist Rachel Bolan wrote, "Sister from another mister", while SKID ROW guitarist Scotti Hill added, "This is going to be so much fun!"

SKID ROW shows with Lzzy Hale on vocals:

May 17 - Walker’s Bluff Casino Resort – Carterville, IL May 18 - Riverside Casino & Golf Resort – Riverside, IA May 31 - Nugget Casino Resort – Sparks, NV June 01 - Hard Rock Live Sacramento – Wheatland, CA

HALESTORM covered SKID ROW's "Slave To The Grind" for the 2011 EP "ReAniMate: The CoVeRs eP". The band has also performed the track live, as can be seen in the video below.

Grönwall, who was diagnosed with acute lymphoblastic leukemia in March 2021, is immunocompromised, which made touring difficult.

"I'm getting stronger and healthier every day but after consulting my doctor I need to allow myself more time to recover, which I can't do as the lead singer of SKID ROW," he said in a statement. "That's why I have reached the tough decision to move on."

The SKID ROW members said in a statement that they are "proud of what they have created and accomplished with Erik over the past two years" and "wish nothing but the best to him and his health. To celebrate the last two years, the band will be releasing a live album that perfectly captures this moment of time in the band's 35-plus-year history, to be announced soon."

In September 2021, just four months before he joined SKID ROW, Grönwall released his new cover version of "18 And Life" via all streaming platforms.

In late March 2022, SKID ROW released its first single with Grönwall, "The Gang's All Here". The song is the title track of the band's latest album, which arrived in October 2022 via earMUSIC.

SKID ROW played its first show with Grönwall on March 26, 2022 at Zappos Theater at Planet Hollywood Resort & Casino in Las Vegas, Nevada as the support act on the rescheduled dates for SCORPIONS' "Sin City Nights" residency.

youtube

#lzzy hale #halestorm #Youtube

4 notes · View notes

the-monkey-ruler · 7 days

Text

Aotu World (2021) 凹凸世界

Date: June 3, 2020 Platform: iOS / Android Developer: Seven Chuangshe Publisher: Seven Chuangshe Genre: Strategy Role Play Type: Appropriation

Summary:

The Creator God created countless planets and people, and established seven envoys to carry out God's will. However, resource differences between planets make the world unfair. Only by participating in the "Bump Contest" and winning the final victory can people change their destiny. And just today, the "Bump Competition" suddenly became abnormal...

As a trainee Angel has forgotten his name and past, the player is resurrected from the Yuanli Pool by the Seven Gods and sent to the Bump Competition to investigate the initiator behind the abnormality of Warcraft. After coming into contact with more and more contestants, Angel gradually developed a desire to understand his past, but this desire was repeatedly blocked by referee Ball 160628, which seemed to contain a very dangerous secret.

Players will face many dangers and unknowns side by side with the contestants

Source: https://baike.baidu.com/item/%E5%87%B9%E5%87%B8%E4%B8%96%E7%95%8C/23390325?fromModule=lemma_search-box

Link: https://www.taptap.cn/app/55881

#Aotu World #凹凸世界 #jttw media #video game #game #Seven Chuangshe #sun wukong cameo #sun wukong #appropriation

6 notes · View notes

nostalgebraist · 2 years

Text

on "ai forecasting: one year in"

AI is improving so fast, even expert forecasters are surprised!

. . . wait, is that true?

Who are these experts? And what exactly was it that surprised them?

If you have been following along with the LessWrong-adjacent conversation about AI, you have probably heard some form of the bolded claim at the top. You might have heard it via

Ayeja Cotra:

As a result, I didn’t closely track specific capabilities advances over the last two years; I’d have probably deferred to superforecasters and the like about the timescales for particular near-term achievements. But progress on some not-cherry-picked benchmarks was notably faster than what forecasters predicted, so that should be some update toward shorter timelines for me.

or Dan Hendrycks et al:

Capability advancements have surprised many in the broader ML community: as they have made discussion of AGI more possible, they can also contribute to making discussion of existential safety more possible.

or Scott Alexander:

Jacob Steinhardt describes the results of his AI forecasting contest last year. Short version: AI is progressing faster than forecasters expected, safety is going slower. Uh oh.

All of these people cite the same blog post as their source.

In the last example, Scott is . . . well, just linking to a blog post, and it's clear that his "short version" is a summary of the blog post, not necessarily of what's-actually-true.

But in the other two examples, the claim is being treated as a "stylized fact," a generalization about reality on the basis of an empirical result. It's not about Jacob Steinhardt and his contest, but about "forecasters" and "capabilities," as entire categories.

This is a pretty striking conclusion to draw. "Big if true," as they say. So a lot is resting on the shoulders of that one blog post and contest. Do they justify the stylized fact?

----

In August 2021, Jacob Steinhardt organized a forecasting contest on the platform Hypermind.

In July 2022, he summarized the results up to that point, in the blog post everyone's citing.

Here's how Steinhardt begins his summary:

Last August, my research group created a forecasting contest to predict AI progress on four benchmarks. Forecasts were asked to predict state-of-the-art performance (SOTA) on each benchmark for June 30th 2022, 2023, 2024, and 2025. It’s now past June 30th, so we can evaluate the performance of the forecasters so far.

That is:

Forecasters were asked to predict 4 numbers, each one at various times in the future.

The earliest of those times has come and gone, so we have something to compare their predictions to.

The predictions we can evaluate were about what would happen a little under a year in the future.

Each of these predictions is about the best published result on some ML benchmark, as of the date in question.

How did the forecasters do on those one-year-ahead questions?

On two of the four questions (MATH and MMLU), the actual value was at the extreme high end of the forecasters' probability distribution.

On one of the questions (Something Something v2), the actual value was on the high end of the distribution, but not to the same extreme extent.

On the other question (adversarial CIFAR-10), the actual value was on the low end of the distribution.

(By "the forecasters' probability distribution" here I mean Hypermind's aggregated crowd forecast, though you should also read Eli Lifland's personal notes on his own predictions. Lifland was surprised in the same direction as the crowd on MATH and MMLU.)

----

The first thing I want to point out here is that this is not a large sample! There are lots of important benchmarks in ML, not just these 4.

This is most relevant to the stronger version of the claim -- that "capabilities" are moving fast, but "safety" is moving slow. Here, a single benchmark is being as a proxy for the entirety of "AI safety progress." Did safety really move slowly, or did people just not care that much about adversarial CIFAR-10 over the last year?

(Do people care about adversarial CIFAR-10? I mean, obviously people care about adversarial robustness, there are thousands of papers on it, but is it really a good proxy for AI safety as a whole? When you ask yourself what the most promising AI safety researchers are doing these days, is the answer really "trying to get better numbers on adversarial CIFAR-10"?)

This contest is, at best, really weak/noisy evidence for the stronger version of the claim. Definitely not "stylized fact"-caliber evidence IMO.

With that aside, let's think about what these forecasts actually mean.

These are forecasts about state-of-the-art (SOTA) results.

A SOTA result is not something that gradually creeps upward in a smooth, regular way. It's a maximum over every result that's been published. So it stays constant most of the time, and then occasionally jumps upward instantaneously when a new high score is published.

How often are new SOTA results published? This varies, of course, but the typical frequency is on the order of once a year, give or take.

For example, here's 5 and a half years of Something Something v2 SOTAs:

The value only changed 7 times in those 5+ years. And the spacing was uneven: judged by SOTA results, absolutely no "progress" occurred over the entire two-year interval from late 2018 to late 2020.

This means that a well-calibrated forecast with a single-year time horizon should always under-predict progress, if any progress actually occurred!

When you answer one of these questions on a one-year horizon, you're not actually saying "here is how much progress is likely to happen." You're effectively making a binary guess about whether anyone will publish any progress -- which doesn't happen every year -- and then combining that with an estimate about the new high score, conditional on its existence.

If there is progress (and even moreso if there's significant progress), it will look like relatively fast progress according to your distribution, because your distribution had to spend some of its mass on the possibility of no progress.

Indeed, any serious forecast distribution on these questions ought to have some amount of point mass at the current value, since it's possible that no one will report a new SOTA. The distribution would have at least two modes, one at the current value and one above it.

But Hypermind's interface constrains you to unimodal distributions. So the forecasters -- if they were doing the task right -- had to approximate the bimodal truth by tweaking a unimodal distribution so it puts significant mass near the current value. And since these distributions are nice and smooth, that inevitably drags everything down and makes any actual progress look "surprisingly high."

(Sidenote: if I understand Hypermind's scoring function correctly, it actually encourages you to report a distribution more concentrated around the mode of the true distribution than the true distribution itself. So if I were in this contest and just trying to maximize winnings, I'd probably just predict "no progress" with the highest confidence they'll allow me to use. I don't know if anyone did that, though.)

Still, though . . . even if you can only use unimodal distributions, hopefully at least you can flatten them out so they capture both the "no progress" side of the coin and the "how much progress, if any?" side -- right?

Well, no. Apparently Hypermind has a maximum on the std. dev. of your distribution, and in Eli Lifland's case this was too low to let him express his true distribution! He writes:

I didn't run up to the maximum standard deviation [on MATH], but I probably would have given more weight to larger values if I had been able to forecast a mixture of components like on Metaculus. [...]

I think [the std. dev. limit] maybe (40% for my forecast) would have flipped the MMLU forecast to be inside the 90th credible interval, at least for mine and perhaps for the crowd.

In my notes on the MMLU forecast I wrote "Why is the max SD so low???"

And indeed, his notes reveal that this is a pretty severe issue, affecting many of the questions:

The max SD it will let me input is 10… want a bit higher here and obviously would want even higher for later dates. The interface is fairly frustrating compared to Metaculus in general tbh.

Why is the max SD 7.5… I want it larger. Still think most likely is in teens but want a really long tail. Have to split the difference

Ok the max SD still being 2.5 is incredibly frustrating… still think 49.8 should be the modal outcome but want my mean to be able be higher. I guess for 2023 I’ll settle on the 49.8 modal prediction and for 2024 start going higher.

Wow, the max SD is insanely low yet again… my actual mean is higher, probably in the 68-70 range

Really wish the SD could be higher (and ditto for below).

Remember that we're already doing stats on a tiny sample -- at most 4 data points, and if we insist on forming "capabilities" and "safety" subgroups then we only have 3 data points and 1 respectively.

And remember that, because progress is "spiky" and only spikes ~1 time a year, we know one of two things is going to happen:

either there will be ~0 progress and it will look qualitatively like people "overpredicted" progress, or

there will be > 0 progress, and it will look qualitatively like people "underpredicted" progress

And now consider that -- although the above is true even in the best case, where everyone reports their real distribution -- we are not in the best case. The forecasters in this contest are literally saying stuff like "my actual mean is higher [but the interface won't let me say that]"!

----

Above, I showed a screenshot of SOTA progress over time on one of the benchmarks.

That example actually understated how severe the ~1-result-per-year problem is. I picked the benchmark with the longest history, to make a point about how often SOTAs arrive, on average over a multi-year interval. But that was also the benchmark where progress was the most incremental, the least "spiky" -- and it's one the forecasters did relatively well on.

MATH and MMLU -- the two where the forecasters really lowballed it -- look different. Here's MMLU:

Except this graph is sort of a lie, because MMLU was only introduced in 2021. The earlier data points come from going back and evaluating earlier model on the benchmarks, in some cases fine-tuning them on some of its data.

But fine, let's imagine counterfactually that MMLU existed this whole time. In 3 years, there have essentially been 3 big jumps:

UnifiedQA and GPT-3 in May 2020

Gopher in Jan 2022

Chinchilla in Apr 2022

What would this look like, if you ran this contest at various points in the (counterfactual) past?

It all depends on precisely when you do it!

It took about a year from GPT-2 to the next milestone, so if you start forecasting at GPT-2, there's either "zero progress" or substantial progress -- depending on whether the May 2020 milestone "sneaks in at the last second" or not.

And it took ~2 years from that to the following milestone. If you started in the first half of that interval, there would be "zero progress," and forecasters would qualitatively overpredict. If you started in the second half, there would be "substantial progress" and the forecasters would qualitatively underpredict (since, again, they have to give some mass to the zero-progress case).

You'd get diametrically opposed qualitative results depending on when you run the contest, not because progress was slow in one interval and fast in another, but wholly because of these "edge effects."

OK, that's MMLU. What about MATH? I've saved the big punchline for last:

Yep, there are only two data points. We're trying to (effectively) estimate the slope of a line from literally two points.

More importantly, though: when exactly did that jump happen? Steinhardt writes:

Interestingly, the 50.3% result on MATH was released on the exact day that the forecasts resolved. I'm told this was purely coincidental, but it's certainly interesting that a 1-day difference in resolution date had such a big impact on the result.

These "edge effects" are not hypothetical: the MATH result "snuck in at the last second" in literally the most extreme possible way.

If the Minerva paper had come out one day later, the contest would have resolved with "no progress on MATH," and we would have observed qualitative overprediction.

What would have happened in that world? I imagine Steinhardt would (quite reasonably) have said something like:

"Yes, technically the contest resolved with no progress, and I'll use that for deciding payouts and stuff. But for drawing conclusions about the world, it's 'more true' to count Minerva as being inside the window. It was only one day off, and it was a huge gain, after all."

But then this would have forced people to confront the topic of edge effects, and there would have been a whole discussion on it, and I wouldn't have to belabor the point in a post of my own.

Did you notice the asymmetry? In the world where Minerva came one day too late, I don't think anyone would feel comfortable just writing it off and saying "yep, no progress on MATH this year, end of story." People would have decided that Minerva at least "partly counted" toward their estimate of progress.

But in our world, no one is doing the reverse. No one is saying that Minerva "only partly counts." Steinhardt notes the edge effect, but doesn't say anything about it casting doubt on the implications. He gives it as much weight as the other results, and it ends up being a major driver of his qualitative conclusion.

----

Yes, yes, you are saying. You're right, of course, about all these fiddly statistical points.

But (you continue) isn't the qualitative conclusion just . . . like, true?

We did in fact get surprising breakthrough results on MATH and MMLU in the last year. What exactly are you saying -- that these results didn't happen? That "forecasters" somehow did see them in advance? Which forecasters?

You are right, reader. If the claim is about these specific benchmarks, MMLU and MATH, then it is true that they over-performed expectations over the last year.

Where things go wrong is the leap from that to the stylized fact, about "capabilities moving faster than expected" as a purported real phenomenon.

You've seen the graphs. "These benchmarks over-performed expectations this year" is like saying "the stock market did unusually well this past week." Some years, the SOTAs overperform; some years they underperform (because they don't move at all); which kind of year you're in depends sensitively on where you set the edges. At this time scale, trying to extract a trend is futile.

What's more, you also have to be careful about double-counting. If you're following this area enough to have seen this claim, you probably also heard independently about the Minerva result, and about how it came out of nowhere and surprised everyone.

From this knowledge alone, you could have inferred that expert forecasters wouldn't have guessed it in advance, either. (If not, then where were the voices of those expert forecasters back when the result was announced? Who said "yeah, called it"?)

You would have known this information already, even if this contest had never happened. Now, reading Steinhardt's post, what exactly is it that you learn? What new information is here that you hadn't already priced into your thinking?

I think what people are "learning" from this post is something about the systematic nature of the phenomenon.

You can see Minerva and think "huh, that could be a fluke, I don't know enough to know for sure." But when you hear Steinhardt say "progress on ML benchmarks happened significantly faster than forecasters expected," with supporting numbers and charts, it feels like you're learning that things like Minerva are not flukes: that when you look at the aggregate, this is the trend you see.

But in fact, this result is just Minerva and Chinchilla -- which you'd already seen -- repackaged in a new form, so it's hard to tell you're double-counting.

Viewed as statistical evidence, this stuff is too noisy to be any good, as I detailed above. Viewed as a collection of a few anecdotal stories, well, these stories are noteworthy ones -- but you've already heard them.

I feel like I've seen several cases of this recently, this process of "laundering" existing observations into seemingly novel results pointing in the same direction. I get the same vibe from that one Metaculus AI forecast everyone keeps talking about.

The "forecasters" are not magic -- in many cases they are literally the same people who later go on to interpret the forecasts, or at least socially adjacent to them! They are using publicly available information, and making judgments that are reasonable but routine, based on familiar arguments you've heard before. If you already follow this area, then what really separates you from the "forecasters"? A bit of calibration training, maybe, at most?

And so we have forecasters boggling at each others' forecasts, and updating on each other's updates, and doing statistics to each others' stated opinions and re-updating on the "statistical versions" of the same arguments they've already incorporated into their inside views, and so on, creating this unstable feedback system that can easily spiral one way or another even in the absence of any new information coming in.

#ai tag #statpicking

75 notes · View notes