Tech News: When is a nanometer not a nanometer?

Apple's semiconductor chip manufacturing company of choice which is t SMC, and to be fair, that's kind of the choice for pretty much every company that's making advanced chips and is a problem in its own right, but that's a matter for a different podcast episode. It is reportedly working on a three nanometer chip for upcoming MacBook computers. Now, the reason I decided to include this was because I wanted to do a little deconstruction on the nomenclature we use

for chips, because it is wildly misleading. So for ages, the semiconductor industry has differentiated chips by using the size of nodes as the naming convention, and by nodes, we're really talking about the length of transistor gates. So the length of the transistor gate in whatever unit was an indicator of the chips sophistication. Generally, think speaking, you know, the more you can cram onto a chip, the more powerful the chip can be. That's not always the case,

but that was kind of the rule of thumb. Now, does this mean that a three nanometer chip has a transistor gate that is three nanometers long. No, it does not, because for more than a decade, companies have shifted away from focusing on reducing components size almost exclusively and looked more at stuff like chip arc a texture and increasing

the density of transistors and that sort of thing. So chips are still getting more powerful and more sophisticated, but the transistor gates aren't shrinking at the same crazy rate that they were before. However, the naming convention that we use where we use that transistor gate size as the name for the next generation of manufacturing processes, has stuck around. So if your transistor gates aren't getting that much smaller, but you're still dependent upon that that naming convention, then

things rapidly stop measuring out. So that means that ten nanometer chip doesn't necessarily have transistor gates that are ten nanometers long. In fact, some of them have transistor gates that are nearly twice as long as that, so it's really just a naming convention. But a lot of folks think that this naming convention is dumb because for one thing, you know, it's not accurate. For another, since we keep going down, you know, we're reducing the size, and now

we're talking about a three nanometer process. We're running out of nanometers. We're about to get down to the atomic scale, y'all.

Because a nanometer is one billionth of a meter. And that also means that consumers have been really confused for a while and often draw the wrong conclusions because you can have a so called ten nanometer chip from Company A and a seven nanometer chip from Company B. And because there's this implication that the smaller number means more powerful chips, you would naturally think the seven nanometer chip

is superior. But that's not necessarily true because we're really talking about things like architecture and power efficiency, and even the size of the components of the ten nanometer chip could be smaller than that on the seven nanometer chip.

Because you're talking about two different companies and you're just talking about them using a naming convention to market a new generation of semi conductor chip, it isn't actually measuring anything, so a ten ter chip and a seven nimeter chip from two different companies could be made in such a way that the ten nimeter isn't always superior to the seven nanometer. That's why I can get confusing. It's this

marketing strategy that creates confusion, and it perpetuates confusion. So yes, I just used a news story to give a quick lesson on why the semi conductor industry is using misleading marketing material and you should do research before you choose a processor and not just do it based off the supposed node size. Former Apple employee Xiao Lang Jong has pled guilty of charges of stealing proprietary information from Apple for the purposes of sharing it with another company, one

in China for that matter. Now, this story actually started back in when Jong was first arrested. Jeong had turned to Apple after taking a trip to China, and then not long after his return, he resigned from Apple, and he also started sending corporate documents to his wife's computer, including documents that, in fact, as far as I can tell, exclusively focusing on documents related to Apple's worst kept secret.

It's you know, autonomous electric vehicle project that everyone knows about, but Apple is never publicly at knowledge that included a twenty five page document with detailed schematics of a circuit board that Apple was designing for the vehicle. Moreover, Jeong had told Apple that he was going to return to China and he was also going to work for a company called ex Paying Motors, which is an electric vehicle

manufacturing company that's also developing an autonomous vehicle. Jong had previously led not guilty to the charges after being arrested, but now he has changed his flea to guilty, and he faces up to ten years in prison and a fine of up to two fifty thousand bucks. And you might remember the story of former Google employee Anthony Lewandowski, whom Google accused of stealing documents from its autonomous car

project that would become way Mo. Lewandowski subsequently worked for Uber, and that led to a nasty court battle between Google and Uber plus, Uber unceremoniously ending its relationship with Lewandowski. Lewandowski was subsequently tried and convicted of stealing documents, but then former President Trump pardoned Lewandowski on the last day of his presidency. Anyway, it looks like autonomous vehicle research is the hottest target for industrial espionage in tech right now,

so I guess it's fashionable. A senior fellow for the Irish Council for Civil Liberties named Johnny Ryan has spearheaded a class action lawsuit in the United States targeting the computer technology company Oracle. Now, in case you're not familiar with Oracle, it is primarily a A B two B kind of company, a business to business like Its clients are other businesses, and it works in software and database

management and cloud services as well as hardware. Ryan's lawsuit alleges that Oracle has illegally been collecting the information of

around five billion people. Essentially that Oracle is assembling dossiers on folks, and those dossiers can contain information that includes stuff like names, physical addresses, email addresses, political views, purchase history, geolocation data, meaning that Oracle has been tracking people or at least has access to tracking information so they know where people have been, as well as records of online activity. So essentially all the personal data stuff that we talk

about in other you know, news articles and such. So Ryan disclaiming Oracle is collecting all of that and organizing it into what he calls dossier's Ryan has brought this lawsuit to California, probably because that is the U s state that has the most strict privacy laws, and this is a massive endeavor. It's too early to say how it's all going to turn out, but some folks at least suspect that this is a push to encourage the United States to adopt stronger privacy laws more in line

with what we see over in the European Union. Joshua Benton at Nieman Lab dot org has a great article. It's titled are you legally Liable for the contents of every web page You linked? To? Australia finally gets sensible? All right, some backstory on this. So back in the first decade of this millennium, I just hate saying. The early two thousand's, an Australian lawyer named George Defterros was

arrested and charged with conspiring to commit murder. Defterro was known as a lawyer who represented people accused of belonging to organized crime gangs. Anyway, an Australian newspaper published an article about defter Ros, alleging that he was in fact part of this conspiracy and such, and Google ended up linking to that article and its search results because Google indexes the web, and when people do searches for things, you get the links. Right. Well, flash forward many many years.

The lawyers representing Defterros, who by the way, had all charges dropped against him, so he did not he did

not stand for those those charges they were dropped. His lawyers were seeking to have this article removed from the from the Internet, and they went to the newspaper and demanded that the newspaper removed the article, and the newspaper said no. So when that proved fruitless, they then went after Google, and their argument was that Google, by publishing a link to this article, was kind of endorsing the article, that Google itself was acting as a publisher, and that

it was almost as if the offending piece had come from Google because it was linking to it. So that's kind of wild, right, like that a link can somehow imply that you're responsible for the material that the link goes to. While initially a court in Australia ruled that Google was in fact responsible, but then it got appealed. It went to Australia's High Court and the High Court has reversed that ruling and essentially said this is ludicrous.

If we follow this logic, anyone who links to anything that is later claimed, not even proven, just claimed to be defamatory, shares responsibility and therefore could be sued for liabel. That seems pretty extreme, doesn't it that a link alone could make you respond constable for libel. So what if you were to come across a link to a story and you shared it on your social media plat platforms like on Facebook or on Twitter. Maybe you saw the story,

you just thought it was interesting you wanted to share it. Well, if this earlier court ruling had been upheld, it would have said a precedent that suggests you could be found guilty of libel yourself just by sharing the link, and that you could potentially face charges for it even though you didn't write the supposed defamatory material. By the way, a big part of this story is that while the lawyers were claiming that the article was found to be defamatory,

it never actually went to court. It was settled out of court. So because of that, the claims were spurious, and yet they still went through UH and got this initial decision by the court that was then overturned by the High Court. So it's good that the High Court saw this for what it was, or at least five of the seven judges saw it for what it was. Two of them dissented and argued that Google was in

fact responsible. Not sure what they were thinking. Okay, we've got a few more news stories to go, but before we get to that, let's take a quick break. We're back US automaker Ford announced it is laying off three thousand employees and includes around two thousand salaried positions and one thousand contractors, and the company says this is all part of its strategy to pivot from focusing primarily on internal combustion engine vehicles and to change to put more

emphasis on electric vehicle production. Ford CEO Jump Farley denies that the cuts are a cost saving move, but rather they indicate how Ford is serious about fundamentally each changing course by committing to the future of electric vehicles. My heart goes out to all the folks who got their walking papers. It is an increasingly tough job market, particularly when other auto manufacturers like Tesla have also been laying off employees or making other kinds of cost saving cuts.

The CEO of the cryptocurrency exchange company Binance, says that LinkedIn is absolutely swarming with people falsely claiming to be Binance employees. And I'm not joking about swarming. He says. There are about fifty real profiles belonging to Finance employees on LinkedIn, but in total it's closer to seven thousand claimed Binance employees, which is a big old yauza. So why would people be lying about working for Binance, Well,

it's probably part of crypto scams. The scammers are likely listing Finance on LinkedIn to give themselves a sense of legitimacy when they're talking to their marks, their targets. They're tricking people into pouring money into various schemes. They're usually types of Ponzi schemes. If you don't know what a Ponzi scheme is, it's a subset of pyramid schemes. So a scammer convinces a group of investors to pour money into you know, whatever it is, in this case, a

cryptocurrency scheme. Then the scammera convinces a second round of investors to do the same, and then pays a percentage out to the first round of investors to keep them happy while pocketing the rest of the money, and then they keep going and so on and so forth, and effective scammers can often convince investors to reinvest into the scheme, so they take the money that they're supposedly getting paid out as the scheme is paying off, and they put it back into it, which just gives more money to

the scammers. And ultimately these schemes all collapse in on themselves. They cannot sustain themselves forever. And so the Binance CEO is warning followers not to assume someone really is a finance employee just because it might say so on a LinkedIn account, particularly if that supposed employee is trying to coerce people into pouring money into a crypto investment scheme. This is pretty tricky because LinkedIn doesn't verify work or

education history. They do claim to respond to reports of false accounts, and they say that they look for false accounts, but yeah, this is a If it's seven thousand fake ones out there, that's a that's a pretty big problem. And you know, folks, fib on resumes all the time. I get it, But this goes well beyond that. Japanese company Fujitsu has partnered with reik In, a research institute, with the intent of developing and selling a quantum computer

boasting sixty four cubits starting next year. Now, to brush up on quantum computers, the fundamental unit of classical computers is the bit, and a bit can either be a zero or a one. The fundamental unit of information in a quantum computer is a cubit, which, thanks to quantum effects, can essentially be a zero and a one at the

same time under specific circumstances. And I'm being very high level with this, but when paired with the right algorithms, that kind of computer, a quantum computer can potentially solve a subset of computer problems far faster than a classical computer can. Uh. It's essentially solving for all potential solutions at the same time and then presenting the one that is most likely to be the best. It deals with probabilities,

not certainties. It gets very wibbly wobbly, but uh. It's also important to remember quantum computers are no better than classical computers for other types of applications, other types of computer problems. You would not be using a quantum computer as a gaming rig for example, but they do potentially have the ability to change really important things that we

depend upon, like encryption in the near future. NASA has narrowed down potential future lunar landing sites to thirteen regions, all of which are not too far from the Moon's south pole. I like to think they were on the lunar equivalent to Zillo at the time. Scientists believe that the region is perfect for future Moon missions because the deep craters in the area could potentially hold hydrogen and

water ice. That kind of stuff would be useful if you wanted to make your own rocket fuel, for example, or if you wanted to perhaps process water ice to create not just water, but maybe oxygen. This falls in line with the goals of the Artemis campaign, which has some really ambitious targets, including creating a base of operations

suitable for long term stays on the Moon. NASA has been planning this out for years, and in fact, the launch of Artemis one, which will be an unscrewed Orion vessel on top of the Space Launch System, which is a super heavy lift launch vehicle, is scheduled to launch on Monday of next week. If everything goes to plan. The actual return mission to the Moon in which humans will head back up there. That one is designated as Artemis three. That's not expected to launch until twenty twenty

five at the earliest. Sony has announced via Instagram of all Things that's new generation of VR hardware for the PlayStation console is likely to launch in early This generation of hardware is going to work with the PlayStation five.

It's reportedly softer with better ergonomic design than the earlier generation of Sony's VR peripherals, and Sony says the headset will display graphics at a four thousand by two thousand forty resolution per eye that breaks down to two thousand by twenty and it will have a refresh rate of nineties slash hurts. It's also gonna have a c through mode, so if you get too close to the wall, it'll show you so you don't bump your nose in there. They have not have any information on how much it

will cost. My guess is it will be a few hundred dollars um, So here's hoping that we find out soon. We know that it's coming in early, and that's about it. Well, that is the news for Tuesday, August two. Hope you are all well, make sure you reach out to me with any suggestions you have for future episodes of tech Stuff. You can do that on the I Heart Radio app or on Twitter at tech stuff H s W and I'll talk to you again really soon. Text Stuff is

an I Heart Radio production. For more podcasts from my Heart Radio, visit the i Heart Radio app, Apple Podcasts, or wherever you listen to your favorite shows.