It's a bold move. Educational material -- particularly frontend -- is an ever-shrinking pie, so Geoff is betting on claiming a very large share of that. I hope it pays off.
I'm sure it will be a great resource for LLMs to read.
But seriously, good luck to them. I used to make Photoshop tutorials before YouTube, and it was a lot of fun. When YouTube came around, all the eyeballs went there, rightfully so -- videos made more sense for Photoshop. But I didn't have it in me to pivot.
And these days you don't even have to know how to do things like mask objects. There's one-click solutions, there's generative fill, light matching, face filters, etc. Hell, you can rotate a 2d layer in 3d now. Soon, live prompting in Photoshop will be the norm.
This article falls into the same trap, I dunno what to call them, artsy-upper-class-society-and-film reviewers often do; writing vague yet meaningful-sounding things about the obvious subtext of a movie as if it was a revelation, yet without any specific observations about it.
Yes, this is a documentary about a person. And how the documentary is presented will color your perspective of the person. This should (hopefully) be apparent to anyone who's watched a documentary about a person before.
But how does it do it? What tools does it use to that effect? I appreciate that this film remains relatively neutral/ambiguous, but what makes this unique from other documentaries that deal with ambiguous subject matter?
I respectfully disagree. It's post-postmodern performance art. It's like a box of Post cereal with a Post-It note on it nailed to a post in from of the lamppost at the corner of Meta's headquarters in the metaverse. It's a relentless onslaught of origami tesseracts made out of pages from a Baudrillard book that an NYU student used to roll a cigarette of imported European tobacco while they post on Xitter in all lowercase. It's pure never-been-stepped-on A24.
I'm glad I saw this film in theaters as the people laughing and shouting was part of this experience but I'll never watch it again, i.e. it was good but definitely not "fun". I agree with those who say that excising the final third would've made it tighter.
> But how does it do it? What tools does it use to that effect? I appreciate that this film remains relatively neutral/ambiguous, but what makes this unique from other documentaries that deal with ambiguous subject matter?
If you’re that interested in getting these answers, you probably should watch the film
Unfortunately, it has such a limited release that most people who want to watch it outside North America won't be able to. No one has been able to explain why it has had such a limited release.
Seems like most documentaries from my point of view in Spain. Seems we won't be able to see this in cinemas until February sometime, and before that it'll probably be available via other channels. In the end, most interested will have seen it before February already, so when times come around, not that many people see it. So the cinemas adjust future viewings since not that many end up seeing them, as it takes so long time to arrive here.
Been like this with most mainstream documentaries like this one.
> this film remains relatively neutral/ambiguous [...] with ambiguous subject matter?
There's a big difference in ambiguousness and ambivalence, especially in documentaries. "Ambiguousness" represents an uncertain logical conclusion, whereas "ambivalence" stands in for (the strength of) conflicting forces.
I'd wager that an ambiguous documentary would be close to being unwatchable, and I can't remember that I've seen any. Since I've yet to see You Can See Everything, unfortunately I can't comment about it in any way.
Without having watched the film, I'm sure they can make it look like Sunny Balwani was the guy in charge, or something like that. I don't even know the whole story of Theranos and never followed it with much interest - startup scams are a dime-a-dozen. HP's acquisition of Autonomy always struck me as such a tale: CFO takes the fall for CEO and CTO for a $11 b acquisition that a few years later was worth a tenth because of accounting fraud and when he gets out the CEO gives him a sinecure; then the week of CEO and CTO acquittal the latter gets killed in a bicycling accident and the former dies when a tornado sinks his yacht.
Sam Bankman-Fried and Elizabeth Holmes are interesting characters, true, but there's so much nuts in tech, it's just fertile ground for stories.
The story is so much more than a startup scam. It's about large numbers of supposedly savvy men being duped by a kid. What kind of kid could and would do such a thing?
It seems that everyone interviewed about this story has their own red flag moment -- the point where they knew that something was way off, and that they should stay far away. Silicon Valley VCs being the prime example. It is incorrect to say that Theranos was a VC company, it was more funded by angels who were incapable of objectively judging what was being proposed.
At Stanford a prof (important to say: a woman) who knew that the idea was technically infeasible. (And interesting to contrast with another very senior prof -- a man -- who left his tenured position to join Theranos). I remember reading a New Yorker profile of EH years before Theranos blew up, and the story described how she had all of these extremely powerful old men on her board: George Schultz, Henry Kissinger. That was my red flag moment. Yes, they are brilliant and powerful, but they are also clueless about biotech and medical tech. So why get them involved? Something seemed very off to me. Lots of people knew that something was very off, and yet ...
I’ve watched the film. There are things that happen during the filming of it, involving the people making the project, that I would indeed classify as big spoilers and knowing them beforehand would lessen the experience.
besides the topic that is being explored is deception so its expected the documentary Will use a rethorical or film technique that has some connection to that theme of deception?
Really depends on the land in question. Some forests have thick underbrush, some are almost completely clear. A lot of undeveloped land isn't even forest - swamp, desert, rocky mountains. Sometimes it's far easier to navigate than a parking lot, IME.
[1] very clearly just cherrypicked some particularly popular posts from before the Ukrainian strike, and some particularly unpopular and/or very new posts from after.
Show me some graphs! Some actual data! Not just the 6 exact posts necessary to make the point.
This argument consistently comes up in techie circles when it comes to regulating or outlawing, well, just about anything, and it really gets my goat.
Making something illegal is the prerequisite to enforcement. These are separate parts of the government. You can't build enforcement infrastructure around something that isn't yet illegal! You don't even know what that would look like until you know what the laws you're enforcing are. Enforcement is the goal of these processes, the laws are the foundation from which enforcement stems.
It is true that simply putting a law down doesn't mean the problem is solved. But solving these problems doesn't happen all at once, so there's no reason to criticize the first step for not being a leap to the finish line.
It's the same critiques that come up with taxing the rich. "Oh people will just evade it". Cool then nail them then. Or if they all want to move to tiny tax shelters then tax them when they move their money there.
They want you to believe it's not possible to restrict them so that you don't try.
Good comparison! The best part is that it's not even true! Tax evasion is far from common, as no one wants to go to jail. It's tax… optimization? that's the problem.
In other words, the rich don't generally break the law to avoid paying taxes. Rather, they find _legal_ loopholes that work for them, and even create them through lobbying.
Which suggests that legal measures _would_ be likely to work!
To the common man, there is no real difference between tax evasion and tax "optimization". Especially since the people who most use those "optimizations" often have excessive impact on the writing of the tax code in the first place!
Well sure, making things illegal is better than letting things be illegal, but the current government gets away with a lot of illegal shit so the skepticism is warranted.
I think what people want is just tearing all of the cameras down. Law enforcement had their chance to prove flock cameras were good. They proved to be poor stewards for the technology. Now get them the fuck off our streets.
Third party audits should be part of the enforcement. And even then, those third parties should be vetted and have full daylight as to ownership, conflicts of interest, etc.
Or it's just a flood of comments trying to take the moral high ground without actually helping in any way the cause. I'll call it pure aura farming or something, because I totally fail to see how constant bickering about any fixes is actually doing any fixing. Yes, I'm irritated.
In this case what you've proposed is the exact same setup FISA, and other "national security" collection works under. Does this system work? who knows the people policing themselves won't say
So in the second push add a periodic audit requirement. In the third push mandate unmodifiable firmware and certain physical anti-tamper features. At some point circumventing the law using the existing infrastructure becomes similarly difficult to installing all new infrastructure.
The problem is, these "consecutive pushes" for reform rarely happen in practice and get silently shelved. That's how it went in Germany with a lot of things, anywhere from the Hartz IV reforms two decades ago all the way through the half-finished cannabis legalisation - and we have four years of legislative periods, while the US House is reelected every two years which leaves just one year (if not less) of productive work time.
Having laws is generally a good thing when they are enforced. A law that is not enforced, selectively enforced or not enforceable is actually worse than no law since it gives power and agency to the enforcer without providing much to the people.
Edit: this being said, the New Hampshire laws above seem to be mandating that the data is deleted. Not that it is not accessed. It seems like pretty good start.
But there also could be a law stating "do not set up mass surveillance camera network in public areas".
My prediction is that the pluralistic axes will come back to China, but they will be superficial and strictly controlled. The recovering axes will be aesthetic and consumptive such as fandoms, hobby subcultures, fitness, foreign credentials... They will give young people somewhere to hold face when the main ladder rejects them.
Notably, these axes will be different from the pre-1949 ones: lineage, religion, and reputation came from institutions that were alternative sources of authority. The 2021 crackdowns on fan clubs and gaming showed the state will regulate any subculture that starts organizing people outside the approved channels. So I expect a split: people with exit options (e.g., overseas degrees, foreign income, portable skills) get a real pluralistic status life, while everyone else stays on the single entangled axis, where competition likely intensifies as the pool narrows to those who cannot leave.
As this progresses, we basically just watch to see if any new axis produces its own gatekeepers not credentialed by the state. I'm guessing every one instead gets an official association within a few years of reaching noticable scale.
You've misread/misunderstood the book review. It's saying the exact opposite, and the same thing as you; that in the book, the aliens are more like westerners, and the humans Chinese.
The thesis as given - "The real aliens are the Chinese people" (note the 'real') - is saying that, to an American reading the book, the portrayal of the humans feels more alien than the extraterrestrial aliens, while the extraterrestrial aliens are closer to familiar western culture.
With all these new Jev-like models popping up, has anyone actually started building anything with them yet? It's odd how quickly they've multiplied despite being relatively niche in their use cases, as far as I can tell. I suppose they're simple and cheap enough to make that it's a sort of 'why not' thing for a lot of these companies.
I'm trialling using it to scan user signups for signs of gambling spam, phishing etc. My current process uses an LLM, but seems like a good fit for these classifiers.
Good news, they've been working on adding C# support for a year or two now. I think it's still experimental, but from what I understand they already had a robust extension system that supported C/C++/Java/etc., so fitting C# in wasn't much of a stretch.
(it is only extension support -- not scripting -- but the extension system is such that it can be used for any game logic AFAIK)
reply