At least if it follows some model of "Mondays and Thursdays at the office". Well, what if I need to visit the lab at office on Tuesday - Might as well spend the rest of the day there too.

If it's some sort of general goal ("spend at least two days a week in office"), I have not so far heard of any org that actually plans (even on team level) what you are supposed to do during those office days. Do you schedule lunch together? Have your standups? If everyone just shows up, gets to their desk and has a bunch of remote meetings the benefits are...what exactly? Goals of office attendance reached?

Well, just watched the new Naked Gun with Liam Neeson two weeks ago. And you know...it's Good. Liam doing the whole "Straight man in wacky world" bit like Leslie Nielsen used to, all the background jokes, heck, even the absurdity of the plot. Of course there are the old favorites like "Have a seat" - "No thanks, I have several at home", Frank's driving skills updated to use EV.

By some definition it's probably woke - I mean, there is no comparable joke to e.g. the one in NG 33 1/3 where Frank Drebin finds out that the Bad Girl is actually a guy (and proceeds to vomit into a tuba).

So, the new Naked Gun shows either that 1) It's possible to make comedy *without* offending half of your audience 2) I'm just too easy to please. But hell yeah I laughed at Buffy references and Clippy popping up in your self-driving car (and that entire scene). And finally 3) Good comedy can also make some interesting points. Not the "smart devices can turn you into a lunatic" bit, but the whole "I want to get rid of my enemies by hacking their self-driving car and driving to an ocean". Might make a viewer or two think about riding autonomous cars. Same as NG 2 1/2 did with their environmental statements.

That change in vacation policy? The real, legitimate notification of it will be in an email from an external bulk-mailing service telling employees to click on the link included. There's nothing in it to distinguish it from a phishing attempt, and employees are supposed to trust it.

That's really up to the (in)competence of organization. At least where I have worked, yes, there are sometimes such mails, but they are very rare. Most often, if an external service is being used (e.g. for employee surveys), you first get a mail from internal address (that is also signed with corporate cert) that says something to the effect of "You'll be getting a mail over the next few days from SurveyPartner. The e-mail originates from @domain.com and it has a link to https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fsurveybox.something%2Fou... something".

In my current workplace, I have gotten *one* legit-but-appearing-fake mail and within two days gotten it again but with a preamble from another source telling me to expect such mail.

Just tell the IT department to get their act together.

I sort of figure that their pitch is that AI figures out "what I mean" and you do not need to be that explicit. So you would tell it "Computer, get me the sales report I edited last week".

And it would know what you mean. Because Recall saves everything you do so it supposedly knows what you mean.

But it would *still* be faster to type that in the search box, especially if you need to iterate "that's not what I meant...".

I guess it would work for people who cannot gather their thoughts when typing, and need to do lots of "umm", "eh", "the thing", "Whatchacallit" and the like filler words when speaking.

Heck, even in Star Trek, the crew *always* kept their fingers on their buttons/PADDs/controls. They only talked to a computer when they needed to have a crew member expose their thoughts and didn't have anyone else to talk to.

Heck, Windows has been pushing the Speech-to-text since Windows 2000 at least, if not earlier, and the use cases are few and far between.

Speech is *slow*. That's why people watch Youtube videos at 2x speed. I'd much rather read whatever talking heads have to say, but I'll settle for 2x-3x speed to find the part of video I'm actually interested in.

Maybe if they can get that Neuralink to work so it really goes at speed of thought, but even those experiments have so far been mostly about moving mouse cursors and the like.

Only places where speech makes sense from UI perspective are when you cannot really type, like when driving and you want to tell the car to navigate to wherever.

Good pic for example at https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fwww.reddit.com%2Fr%2Fthink... (or just google image search for Thinkpad T25 keyboard).

I'd pay extra to get my hands on one. There are couple of DIY projects attempting to create this with frame.work 16" model, but nothing beyond that.

And unfortunately, my trusty old T25's keyboard membrane is just starting to wear out - not sure if it's possible to repair it, some keys only work when PC is heated. No spare parts available, except with JP layout.

I *want* my cluster of del/home/end/pgup/pgdown keys. I *want* my print screen, scroll lock, pause keys. I want my cursor keys neatly ordered in a cross. I want my function keys to be distinct from number keys.

Just make a laptop with 7-row keyboard and take my money.

Wil Wheaton (CleverNickName of Slashdot) did a blog post years ago what's it like to be a voice actor in practice, see here (context is the strike a decade ago): https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fwilwheaton.net%2F2015%2F09...

Both Google Family Link and Apple's equivalent allow you to drop any apps you want, including social media apps. So if you want a barebones phone, it's quite easy to achieve - and as a bonus, you can incrementally allow more stuff as kids grow.

I dont know if this is fixeable outside of massive automation

Why would that be a problem? Japan is one of the leaders in robotics. If they can build thousands of semi-autonomous drones on both sides of Ukraine war weekly, then shouldn't be a problem to create caretaker bots.

I basically keep listing of stuff I want to watch and what service it is on. For the obvious ones (Stranger Things S5, Andor's new season, Wednesday etc), it's of course obvious, but if I suddenly decide to check something out (recently: Star Trek Strange New Worlds), I just use justwatch.com to find what platform it is on.

Once there's enough stuff on one service and too little on the one I'm currently subscribed to, I just switch. Kid has also grown up enough that she doesn't demand Disney+.

I still hope they'd at some point get the equivalent of Spotify or Apple music or Tidal. Apart from very few exclusives, everyone is on all services, and musicians get their due depending on the number of plays. Why couldn't the streaming services just provide access to everything and pass the fees to producers depending on views? ...like, you know, Netflix used to!

I have sort of the same deal. I hate "exercise" without any real purpose than busy-work, but I still started swimming around 5 years go. I typically swim about 1 km three times a week. I chose swimming because at least you don't sweat while doing that (or rather, it goes directly to the pool) and you get to do different styles to alleviate some of the boredom.

Result: Weight has been exactly the same as before I started, around 100 kg. Bloodwork shows that cholesterol levels and other indicators are much better than 5 years ago, but absolutely zero effect on weight. Also tried changing my diet - several times. However, all "lighter" diets just results in hunger. I don't mind feeling hungry during the day, but can't go to sleep if stomach is growling for food - so also have tried the 16/8 variant where I only grab water for most of the day and maybe one cookie just to keep blood sugar up, and then eat just before bedtime - no dice. My supposedly ideal weight would be -20kg, and I've never been able to go below 95 for the past 30 years or so.

I *guess* some of that earlier fat has turned into muscle, but frankly, my dad belly still looks the same as ever before.

If a single book is 30 plus hours to read it'll never be a successful series.

To be fair, at least around 1/5 or even 1/3 of each book is
  - Recaps of stuff that happened as the storytelling pans towards PoV character.
  - Descriptions of scenery.
  - Slow as hell travel scenes

On TV, you can skip the recaps, scenery is implicit, and the travel montages can be made much faster.

I mean, story summaries are pretty much:

    - Book 1, works as an independent story, the three country boys and two gals are yanked onto the adventure, during which they learn a bunch of lore and at the end they have a bossfight
    - Book 2&3, the boss was just a sub-boss of the real Big Bad, need to grab a bunch of gear. Gals go to level up in wizard school, boys start hunting for that extra gear. Add a couple of prophecies into the mix.
    - Book 4-7: The main character does a bunch of level ups with Fremen (oh wait, Aiel) and defeats a few more bosses. A new group (Seachan) show up in force, while they have so far been only hinted at.
  - Book 8-11: Folks bounce around and nothing much happens, but getting a bunch of setup for the final battle.
  - Books 12-14: Payoff to the previous four books set-up. Plotlines finally getting tied up with massive battles everywhere.

(Yeah, personally at around book 9 I stopped for several years, but then when news got around that the series had finally finished, I got around to reading them. It was ok-ish in the end, but I'll probably not even try reading them again).

    Anyway, there's so much filler in the books that condensing those to around Book 1 = Season 1 and then 1 season for every 2-3 books is not *that* difficult if you just cull a bit of the plot kudzu.

What's the new test that AI is supposed to pass to be considered generally intelligent? Given that humans are defined as generally intelligent, it has to be a test that a below-average human would pass.

That's a rather philosophical question. Check out Chinese Room argument (https://ancillary-proxy.atarimworker.io?url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FChinese_room). LLMs are essentially filling that role. So the big question is indeed do the LLMs actually "understand" Chinese.

The big thing as it comes to AGI is probably emergence. At some point we just note that yeah, it is now impossible to even try to understand what exactly went down when the AI agent did $THING, even if you have full visibility to all the internal states. Yeah, that's an "if you build it, they will come"-type argument, but seems to be the way things are going.

I'm happily driving a 30 year old truck and a newer passenger car that does not have any outside connections or built in cellular modem. I'm afraid it'll be harder to find something like that as time goes on.

Actually, it might not be so bad. The telematics units are shipped with the cellular network technology of the date. In Europe at least, 2G/3G networks are being turned off (in my country, 2G is going away in 2029, in Sweden in 2027). Well, lots of cars have only 2G/3G cellular radios in them, with no way to upgrade.

=> You can buy an used car that only has cellular module for technology that is not deployed anymore. So no need to be stuck with 90's cars.

The problem stems from supposed "realism". It used to be that acting in movies and TV had roots in theater - where actors are used to projecting their stage voice. Also, I sort of have thought that they are using less ADR these days.

Anyway, you can see this at work even in later productions - watch Picard, for example. Patrick Stewart still brings his Voice - you do not need subtitles to understand what Picard is saying. Compare to the younger actors in the same production - yes, they talk more "realistically", but the voices sound like they are coming from your random home video recorded on a phone or something.

The other issue is technical. It used to be that all the voice acting was placed via the front-center channel. That's why I have set up my amplifier so that in my 5.1 setup the center speaker has +3dB more than the others - it brings out the actor's lines out much more clearly - in OLDER movies ("Older" in this case meaning anything before about 2005 or so).

With the newfangled Dolby Atmos, however, you get "spatial audio" instead of discrete channels, so you can apparently no longer get that nice and easy way to isolate the speaking parts. Yes, the characters voices are coming from the "correct" location in relation to the screen, but they are mixed in with all the background sound effects. I can no longer explicitly bring out the lines by fiddling with my mixer settings.

And I have a 5.1 setup. If all of this is shunted via a soundbar, or TV speakers, or even (shudder) an iPad or a phone, it gets much worse.

So, these days I'm watching spoken English movies with English subtitles, despite understanding the language.

Give me a technical means to say to my mixer "Increase ADR tracks by +6dB", and I do not need subtitles anymore.