Muse has been downloaded 9 million times in the U.S. and Canada since its launch on September 8, market intelligence company AppMagic, told Rest of World.

I don’t even want to know where they got this figure. Downloads are not the same as users. I would imagine a third is tech influencers and journalists who played with it for a hot minute, produced their piece on it, and uninstalled. The figure is immediately less impressive, isn’t it.

The question of “Western” values over local ones in developing countries is valid. While I would assume the models will be able to learn, I also think the users there will quickly realise that this stuff doesn’t work and stop using it. A similar thing will happen in areas of sketchy internet coverage. People will just not use this crap. And that the vegetable stall on the street market is going to implement an inventory system and publish an agent-friendly website so they can be part of this magnificent technology sounds very unlikely to me. Payment systems, okay. If you can do that on a phone connection. But this much complication? In the PRC in a big city maybe, Burkina Faso maybe not so much.

The agent apocalypse won’t be as bad, I don’t think. There will be plenty of stories of how the Meta agent rather than booking a 10:30am ordered a military drone strike on the nail salon by mistake or something.

The languages are a thing. Even in developed nations like Switzerland or Belgium there are challenges that the “American Intelligence” people have probably not thought about. Brussels is good example where dual infrastructure for Flemish and French speakers exist and really 2.1 language groups live separately and together at the same time. The .1 is all the EU and NATO foreigners in the city plus the odd Belgian German speaker. Booking an appointment at a doctor that doesn’t speak your language through your clever agent isn’t helpful.

I read about some research (and it may have been funded by the socials, I don’t remember) that in a Coke/Pepsi blind test, users preferred algorithmic feeds. Most people do not think too much about what is happening behind the scenes and are happy to receive content they like. The unwashed masses are less concerned with things like privacy or moderation of content. I think there is a strong correlation between people who care about such things and their preference for chronological feeds. And we’re the minority.

An algorithmic feed will always be a better vessel to sell ads. Or in a more conspiratorial sense, to serve you “approved content.” So in the end, it boils down to money. Money prefers algorithms.

It’s precedent-ish. The first court in Munich did the hard work from scratch. The Aachen court copied and pasted their logic into their own ruling. There is a strong chance that for the time being other courts will come up with similar rulings under comparable case conditions. But Google may be waiting for a case where they are not quite as likely to lose if they appeal. Funny shit can still happen, especially if the highest federal court can be convinced to rule on a case like this. So it’s too early to tell if this precedent has more long lasting staying power.

I started a rewatch of DS9 a couple of months ago and I’m a third done with S7. I saw Oceans 11 on DS9 earlier today. Didn’t even skip all the bad Ferengi episodes in the rewatch.

I suppose it was hard for the makers of Disco to square the circle of telling Star Trek stories in a post breaking bad streaming era. In TNG S1 they use the phaser as a visual effect maybe a dozen times, most of those in Conspiracy, because it cost too much. Now they can do whatever for a dime so they overlooked that we liked the character stuff they wrote to fill the hour so they wouldn’t blow most of the budget on SFX. I hope that in a future fourth wave Trek they fill the hollowness we feel.

Yeah, I have been looking at the veritable content drought we’ll be entering since there is nothing in production any more. And in streaming TV stuff takes ages to get off the ground or even renewed. I read on the grapevine that letting all existing contracts expire is a way to kick Kurtzman out and after that there is a chance for a fresh start under new leadership. But that will take time until after Academy S2 goes online. So considering all this stuff going on behind the scenes, I’m okay with Muppets and Casablanca in space. Whatever! Better than nothing.

A prequel! What a novel idea.

This should be an org chart at the Star Trek office today when they look at show ideas: is it another prequel? Yes - don’t pursue this idea. No - Allamarain! Move along home.

Granted, they got lucky (initially) with SNW. But that needed a backdoor pilot through Disco to get launched.

If you get to that position and pet peeves are the first thing you’re thinking about you will have proven the Peter Principle correct. Again.

How would you know the baby was your deceased significant other? What have you sniffed this morning?

I stopped watching Marvel movies because I found the plots virtually nonexistent. And don’t get mad at me, dear reader, if you love them. That’s okay, I’m not taking that away from you. I’m just saying how I felt.

So I don’t know if I would sign on to your theory. I did enjoy a lot of the characters but it didn’t carry an ironman or Thor movie for me.

I would like a sequel of Hail Mary though.

Time keeping developed as a local phenomenon. We are stuck with this dumb 24-hour clock. People don’t want to have to get up at 22:00 in Vanuatu to go to work when it should be 7:00 AM because it’s dawn outside. Time keeping stays local. Unless you live in the West of the PRC.

GMT has been rebranded UTC to remove the reference to Greenwich and whose acronym doesn’t stand for anything so not to annoy the French. NATO uses Zulu time, which is UTC, regardless of where they are on the planet.

We’d improve the situation if we all just agreed to get rid of daylight savings everywhere. If it wasn’t for that, the UTC +/- whatever would work much better. Alas, politics.

It’s a Tuvix level decision I’m glad I didn’t have to make. I think reuniting a child with their biological parent is a general, understandable guideline. And if you look at this clusterfuck in sequence, the first crime in this case - the illegal invasion of Bajor notwithstanding - is Rugal’s abduction by Dukat’s aide and his subsequent depositing in an orphanage. If it hadn’t been for that, Rugal would not have had Bajoran foster parents. Who were understandably not very nuanced about the Cardassians. (There must have been a way to raise him not hating his “kind” unconditionally as he does. So they carry a small dose of guilt for this mess in my opinion.) Which led to Rugal biting Garak.

The biological dad makes a convincing case that he was fucked over. Rugal is fucked no matter what. Sisko went back to where this train was sent on the wrong track and corrected that. I can kind of see the logic of that decision. He had to weigh different factors and decided to give less weight to Rugal’s wishes over the crimes that preceded him ever being in a position to express them this way.

Tough call but he (somehow, there doesn’t appear to be a legal system with a metal sphere gavel and the social worker from Mrs Doubtfire presiding just yet) had to make one. And there probably isn’t a right decision.

As a starter, Duolingo isn’t bad. Sure, it’s mostly slop at this point so keep that in mind. And only stick with it until you realise the mistakes.

1. Is this from Xwitter? And if so, why? Nothing on that platform is worthy of our time and attention.

2. Even if it isn’t Xwitter. This could be just a joke. Admittedly, not a good one. But a screenshot may not be enough to banish this father of the year.

I just wanted to point that out because a lot of praise heaped upon them comes from the more zealoty side.

I appreciate their thinking to the point that it means there is an office suite of software available that doesn’t currently contain any “American Intelligence” features.

LibreOffice said that there is currently “no integration” that meets all of its requirements, such as ensuring that the users’ data stays on their device and that there is no reliance on a single AI provider, among other considerations.

As soon as there is a version of it that can be run entirely locally, they are not averse to adding it though. They’re not on team never-AI.

Japanese kids learn the alphabet in primary school to do Romanisation of their own language and then for English class. There are no problems with the former because they can see and read those everywhere. English is a particularly difficult alphabet language to read so that’s hard. But on a certain level there is relief that there aren’t an additional 2000 Kanji characters to learn to be able to read it. It’s a radically different language so it’s hard either way.

Mandarin learners must pick up Pinyin, at least in the mainland, fairly early. I’m guessing primary school as well.

Both Pinyin and Romaji use the alphabet to make sounds that don’t exist in English. Like rya in Japanese or the distinction between q, ch, and zh in Mandarin. Both mainly use qwerty keyboards for typing their own language.

Gendered languages flummox even learners from the country next door. The biggest obstacle to learning English in Asia isn’t the different writing system or even the insane spelling conventions of English or French. It’s mastering indefinite and definite articles, when to use either or none, that makes them pull their hair out. It doesn’t matter which European language you pick, it makes no sense, it’s vibe based more than anything, and it isn’t the same between French and German or English and Spanish. So knowing one language already is only of limited use.

I’m going to theorise that a writing system with spaces between words helps learners to grasp what is an equivalent unit in their own language. To a point because you get words like antiestablishmentarianism in English or Verwaltungsfachangestellte (an administrative assistant in the public service sector) in German and all bets are off.

I have a feeling that the people who could judge that best probably aren’t present in an English language forum.

I think if we look at the spectrum of popular international animation with South Park America on the low end and Miyazaki output at the top, the care put in to make the spoken word match lip movements often correlates to where you are on that spectrum. And there is a lot of stuff trending towards the low end. They probably don’t do 20 retakes on a line read during voice recording. No matter what language they are doing.

Shit like that can’t be planned. It either catches on or it doesn’t.

Why do you want to make it cooler? Why polish this turd?

Startpage is owned by an American advertising company.

Eff me, really? Can you back this up? I only see a company in The Netherlands.

Google in particular getting worse to get you to click through more results pages or to run more queries so they can sell more ads on each of those is well attested. I used Bing for a while and didn’t think it was better or worse than newly bad Google. So now I use an equally shitty non-US search engine (start page) and just give up sooner.

In their defence, I think the exponentially increasing content of the internet, and the not insignificant portion that just sits there dead or unmaintained, plus the bits and bytes that just disappear as subscriptions or support run out are difficult problems. It’s getting ever harder to write an algorithm to sort through that as well. That plus greed equals shit.

Pre-Google we had very shitty search results. Than Google became the Kleenex, the word everybody used. Then Google got worse. While my faith in “American Intelligence” is somewhat limited, I think we will get back to better results soon. What Google had to do (build big server farms to look at all of the internet to index it) may become trivially easy for a sufficiently good LLM eventually. Although we’re not there yet. Given that it’s not just Microsoft or Google fighting in that market for dominance, I live in hope that competition from OpenAI etc. lights a fire under their arses.