Their reviews aren’t frequent. This is more of a “we ask around the org for what everyone liked this year.” It’s a consistently good list year after year.

Being picky about it seems like a good idea. I do the same with Japanese; the only thing I don’t fail myself on is pitch accent (Japanese has a high/low pitch thing). Everything else has to be right.

I do catch myself sitting on a card longer than I should at times, though if a session goes long for me it usually means I’m shadowing a lot of sentences out loud. I don’t think I’ve seen specific guidance on how long per card is too long, although if I go longer than around 10 seconds on a card it does start to feel like I’ve lost flow.

I finished Final Fantasy Tactics: The Ivalice Chronicles. I’m a nerd for this game in a particular so I have many nits to pick, but I was happy with it overall. That said, I don’t recommend the hard difficulty; I didn’t get much out of it.

The voice acting, though. Wow. Chapter 1 hit hard this time, especially Milleuda’s and Ramza’s performances. Joe Pitts is excellent in this role, and you can really feel now how much of a turning point the Plateau was for the character. I appreciated the additional dialogue too, although it got oversold in the marketing. Pretty sure most of the script size increase Matsuno was talking up ended up in menus.

Not anticipating playing anything in the genre. I’m sucked into Warhammer 40K: Rogue Trader currently, and I’ve got Trails Beyond the Horizon next month anyway.

Edit: scratch that, after seeing a recommendation I ended up picking up Ni no Kuni: Wrath of the White Witch to play in Japanese. I’ll be playing a little here and there.

Slightly different? Comparing a dead simple plot of employment vs. the performance of the S&P to a DID Poisson regression event study is the coughing baby vs. hydrogen bomb meme.

This Stanford study is just one in a very active field of economic research, so it’s reasonable to be skeptical, but I really hope you don’t think people make decisions based on the kind of thing in that Tiktok video.

Rewatching the original Law and Order for the umpteenth time, season 3. Episodes about corporate responsibility and exploitation of immigrants. The more things change…

If it wasn’t for the payphones it’d feel timeless.

This study controls for the post-COVID hiring spike in three ways: the researchers generated results without including the tech sector, they separated out remote work, and compared trends from 2018-2022 to those after. The hypothesis holds in all cases. The primary regression analysis also included a standard set of controls for hiring trends (such as interest rate fluctuation).

There’s enough here to find a negative correlation between generative AI and entry-level employment.

An hour is quite a bit of time for 100 cards. What’s your recall percentage like? Do your cards have context clues, like phrases, sentences, or even voice clips?

I’ve felt a lot better about my progress once I cut my Anki time down myself, so I certainly agree with the idea of trying to budget more time elsewhere.

Dude, I just heard of Chill With You (from Denfaminicogamer actually) and I’m itching to try it out. Probably tomorrow. I had never seen the concept before, are there more decent ones other than Spirit City?

And yeah, I certainly also feel it when I’ve got other processes tugging on me during the week. Takes a toll.

In my experience this is the biggest knock against it, and it can be fatal for multiplayer games. I had to wait several days for a patch to get pushed to continue my Baldur’s Gate 3 campaign with a friend because she’d picked it up on Steam. We eventually had her keep Steam offline.

Considering the condition games can be released in lately, it can really suck to wait in general, too.

Hell, why not? Something to be said for mild self-accountability when you’re telling people you’re doing something.

I added a monolingual dictionary to my web browser lookup tool (Yomitan). There have been a few undeniable signs recently that my language learning has progressed the farthest it’s ever gone (I started in 2000!), but this is the most glaring, obvious one. It’s kind of surreal, honestly.

Not sure how you got to that conclusion. The survey prompt was “Please indicate whether or not you ever use the following websites or apps.” 29% responded that they do not use the platform, less than 1% did not respond. That’s 70%.

I’d love to see an un-bungled True Detective again.

Dancing around queer romance is a problem in Japanese games as well, and it’s so very disappointing. It’s plain as day to everyone’s what’s going on but they won’t go there.

Tangentially, sanitizing romance in general is also a problem with Japanese games (and games developed abroad that borrow their concepts). Some studios are outright afraid of offending men by having a woman character express a love interest in anyone, so it’s just absent entirely from swathes of the cast, if not everyone.

With gaming having become such a big footprint in entertainment, I worry that sort of thing is going to feed back into TV.

It’s also far from the first time Steam’s content review process has stirred up controversy–even before Collective Shout–which is ultimately the reason why this is getting so much run in games media right now. At some point Steam has to get their shit together, start hiring people, and revamp their scattershot content review system before they get on the wrong side of an incident by either letting something through that stirs up a shitstorm and Congress gets involved, or pissing off the wrong publisher and having the ESA come down on them.

That said, I don’t think this particular game is the horse to back for this effort, so to speak.

I mean, this is the Internet, nothing stopping you from that lol

Here it’s a katakana character purely used for the sound, in this case “ne,” the first half of neko, for cat.

That’s why I said local models. They aren’t automatically taking outputs and training updates on them.

And yeah, we’ve all already had our likenesses folded in somewhere. That’s the bigger problem here.

Yes, if it was on a locally-hosted generative model, I wouldn’t be bothered if someone did this in my likeness. That wouldn’t be meaningfully different than using Photoshop to fake it ten years ago.

Passing it around to their friends and gods know whom else is still just as reprehensible though.

Self-confidence in my studies was pretty high this week. I’m on pace to finish this Tobira chapter in a total of 18 days or so (including the workbooks), which is an amazing speed improvement from before. That puts me on pace to finish the book in spring, which feels really great for my N2 plan. There’s still the whole listening comprehension thing I have to figure out, but right now that’s a Future Me problem~

Also pretty excited because I finshed my Anki reviews in one pomodoro today. I’m thinking within a month I’ll start to see some real results from shifting my time from Anki to more reading.