Search Results for "september 21"

With all the talk about banning TikTok in the United States, Tom dives in to explain who actually owns and controls its parent company, ByteDance, and why there’s so much discourse about security.

Featuring Tom Merritt.

MP3

Please SUBSCRIBE HERE.

A special thanks to all our supporters–without you, none of this would be possible. Become a supporter at the Know A Little More Patreon page for exclusive content and ad-free episodes.

Thanks to Kevin MacLeod of Incompetech.com for the theme music.

Thanks to Garrett Weinzierl for the logo!

Thanks to our mods, Kylde, Jack_Shid, KAPT_Kipper, and scottierowland on the subreddit

Send us email to [email protected]

Episode transcript:

There’s a lot of concern about TikTok these days. It’s banned in India. And many members of the US government would like to ban it there as well or at least force a sale. They assert that it is a threat to national security. Is it? That depends a lot on how you see its parent company ByteDance.

Let’s help you know a little more about ByteDance

Bytedance grew out of real estate search engine called 99fang that was started in 2009 by Zhang Yiming and Jiang Rubo. Those two and some 99fang employees started work in 2012 on an app that would use big data algorithms to classify news stories according to user preference. They founded ByteDance to make that app which later become Toutiao.

It is in fact ByteDance’s core product. You may have run into its video hosting service which is similar to YouTube in that users can upload their own video. Or sometimes, like YouTube, videos owned by somebody else. Like TV shows.

ByteDance has also operated many other news apps mostly in India and Southeast Asia.

Among its many efforts at apps was one called A.me, launched in China in September 2016 and later changed to be called Douyin. Within a year it had more than 100 million users in China.

As Chinese companies often do, ByteDance decided to market Doyin under a different name globally. It launched TikTok in September 2017 in India. In November of that year it acquired startup musical.ly which was also growing fast and in addition to its Shanghai office had a headquarters in Santa Monica. Doyin was only available in China and TikTok only in India but musical.ly was available worldwide. ByteDance decided to merge it into TikTok on August 2, 2018. After the merger, Douyin operated in China and TikTok operated everywhere else.

ByteDance has gone to great lengths to offshore itself as much as possible while still being able to operate in China. ByteDance itself is registered as a Cayman Islands company with a principle business address in China. TikTok Limited is a wholly owned subsidiary also a Cayman Islands company, which itself operates several LLCs which are registered in Singapore, the US, Australia and the United Kingdom.

So is it a Chinese company? Depends who’s asking and what you mean. It’s not a lie to say no, it’s a Cayman Islands company. It’s not a lie when you’re talking about the US version of TikTok, to say no, it’s a Delaware LLC. But in the end a lot of people who work on TikTok, work for ByteDance which has offices in Beijing.

So if you’re asking whether ByteDance or TikTok is a Chinese company, you’re asking the wrong question. The question is how much of what TikTok does happen in China?

For Douyin, all of it. For the other regional variants of TikTok, some. The main algorithm appears to be developed by ByteDance employees in Beijing. So it’s not impossible that someone in the Chinese Government could ask someone at ByteDance to give them some information or make certain tweaks in the algorithm. The way things work these days in China, it wouldn’t even have to be an order. An insinuation that people in power would be pleased if certain things happen would be enough.

So let’s look at each of those scenarios, both data access and algorithm tweaks.

There is no evidence of either happening. But let’s start with data access.

For the US in particular, US user data is stored on Oracle servers in Texas, as part of something called “Project Texas.” The data is handled by an independent company called USDS, which has member of the US Committee on Foreign Investment on its board. USDS acts as a contractor to The Delaware LLC, TikTok LLC in the US. Im practice ByteDance employees in Beijing have reportedly been copied on coding projects that included US user data. Those instances have all been incidental. An engineer needs help on something. Or a moderator needs help on something. So they share a screen or file that they technically shouldn’t. This wouldn’t necessarily deliver useful info into the hands of a ByteDance employee and therefore makes it less likely that this path is how the Chinese government would get data on people.

It would be easy for Oracle or CFIUS to catch any more widespread accessing or searching of US user data and there have been no reports of that. So we’re left with the possibility that maybe some of the US user data might have made its way into the hands of people in China. But it would be much less useful data than anyone in the world could buy from a data broker on the open market.

So an objective evaluation would seem to say that the exposure of US TikTok user data to Chinese authorities is not very high, and in fact more guarded from access by Chinese authorities than data stored with other internet companies.

Next would be the algorithm. Again in the US the algorithm is implemented by USDS and audited by Oracle. That said, it’s a complicated algorithm and it takes a long time to audit. It’s quite possible to miss certain effects of changes. Especially because even the engineers in Beijing creating the algorithm don’t always know exactly how it will act.

One suspicion is that China would order the algorithm to push out recommendations of divisive topics in order to foment unrest in the US. To do that engineers in Beijing would have to make tweaks and hope they are not noticed by engineers in Texas. An achievable goal if not done too often. However it would not be certain to succeed, since it could either be caught and altered by USDS employees, or just not work.

From the other side, the difference between China manipulating the algorithm, and humans just causing the algorithm to do what it does best is indistinguishable. Algorithms at YouTube, Facebook and elsewhere are known to push some people into extreme or divisive content. So the fact that the algorithm did that would not itself be evidence of anything.

However I think an objective evaluation would say that China putting its thumb on the scale of the algorithm is possible, while its effects would not be guaranteed. And the risk here is probably slightly greater than the risk to society from domestic algorithms from US companies.

All of this needs to be considered under the knowledge that if China got caught doing any of the things its government is suspected of doing, it might tank the value of TikTok and therefore ByteDance. That might be a risk China’s government would be willing to take, but what about ByteDance’s investors?

ByteDance is 20% owned by its founders and 20% owned by employees. China’s state-run Internet Investment Fund owns 1% of the Beijing subsidiary. So the government owns a small part of ByteDance but that in itself is not unique to ByteDance. It’s true of many Chinese companies that do not face the kind of scrutiny Bytedance has faced. 60% of ByteDance is owned by global investors. That includes Jeff Yass’ Susquehanna International Group, Kohlberg Kravis Roberts, SoftBank Group, Sequoia Capital, General Atlantic, and Hillhouse Capital Group. Those groups are based in Dublin, New York, Tokyo, Menlo Park, California, New York and Singapore.

Suffice to say the only interest behind Bytedance is mostly American. The exception is Hillhouse Capital, which is Asian, run by a Yale graduate who was born in China.

Let’s take a look at the board of directors to see if we can detect any indication of more Chinese control than we see from the investment side.

The chairman is co-founder Rubo Jiang.
Then there’s Arthur Dantchik from Susquehanna. William E. Ford from General Atlantic. Philippe Laffont from a New York investment management company called Coatue. And finally Neil Shen, a Chinese-born investment manager who now lives in Singapore and runs Susquehanna’s Chinese investment arm.

The Chinese influence on the board appears to be the co-founder and an employee of a western company who no longer lives in China. Not nothing,. But not a smoking gun for Chinese influence on the board.

For comparison, 5 of the 8 board members for rival TenCent were born in China and 4 of them still live there. Of the three- non-Chinese board members, one is from Hong Kong, one from South African and the other Russian.

ByteDance appears to have a heavily western-influenced board of directors and slate of investors.

So let’s recap what have we found.

Technologically it is possible but not easy for the Chinese government to somehow make changes to the algorithm of TikTok. It would be hard to detect from background noise and difficult to predict how effective it might or might not be. It’s less likely that the Chinese government would use its rather inconsistent access to US user data in any useful manner when it has much better options.

And the motivations behind the company appear to be mostly western. While it has obfuscated its ownership with several paper companies, the trail leads back to western, and mostly American investors.

One name I haven’t mentioned in this entire explanation is Shou Zi Chew, the CEO of TikTok LLC. That’s the one that runs the other subsidiary LLCs in multiple regions. Chew is a Singaporean. Born in Singapore, schooled in London, and previously employed by Golden Sachs as well as Russian Yuri Milner’s DST Global.

I haven’t mentioned him because he’s somewhat immaterial to the deduction. He makes a nice face for TikTok in the US because he can credibly say he’s not Chinese. He’s not. But his ultimate boss is Liang Rubo, the CEO and Chairman of the board of ByteDance.

You are more likely to have heard of Zhang Yemin, who cofounded ByteDance and is the more colorful of the two. But Zhang resigned as CEO in May 2021 to focus on long-term strategy. Often code in tech companies for being bored with being a manager and wanting to do fun stuff. Like Larry and Sergei from Google did.

And to my way of thinking, that covers all the angles on ByteDance related to TikTok in the US. So what are we left with?

Concerns about TikTok’s treatment of user data and the effect of its algorithms on public opinion are well founded. However they do not seem to be particularly unique from similar concerns at all other tech companies who operate in the same space, such as Meta and Google. There are some areas where the risk might be less because of the strict oversight of US user data in Texas. There are areas where the concern may be more such as the influence on the algorithm by a non-US government.

But neither of those appear to be in any way extreme concerns or reassurances. It may be that strict rules on how to protect user data or a solid system for users to guard it themselves could solve many of these problems.

Preventing TikTok from operating in the US would certainly solve it for TikTok but likely with little effect on overall public safety. Forcing Bytedance to sell TikTok to a US owner would only change who might influence the algorithm or misuse private data. And would likely not be able to replicate the user experience.

In other words, I hope you know a little more about ByteDance.

CREDITS
Know A Little More is researched, written and hosted by me, Tom Merritt. Editing and production provided by Anthony Lemos and Dog and Pony Show Audio. The public key cryptography players were Sarah Lane as Alice, Shannon Morse as Eve and Andrew Heaton as Bob. It’s issued under a Creative Commons Share Attribution 4.0 International License.

What’s really being celebrated on Black Friday and is it the biggest shopping day in the world?

Featuring Tom Merritt.

MP3

Please SUBSCRIBE HERE.

A special thanks to all our supporters–without you, none of this would be possible. Become a supporter at the Know A Little More Patreon page for exclusive content and ad-free episodes.

Thanks to Kevin MacLeod of Incompetech.com for the theme music.

Thanks to Garrett Weinzierl for the logo!

Thanks to our mods, Kylde, Jack_Shid, KAPT_Kipper, and scottierowland on the subreddit

Send us email to [email protected]

Episode transcript:

At the end of the civil war the US government was deeply in debt, both from the cost of fighting and reconstruction. When General Ulysses S. Grant was elected president 1868 it had grown to $2.8 billion. That would be around 103.8billion in 2023.
To help pay for the war the government had begun issuing “greenback” dollars. These were not backed by gold or silver but promised an unspecified future payment. They had the effect fo driving up the price of gold.
So Grant’s administration pursued a policy of selling gold to buy up wartime bonds and by May 1869 the debt had been reduced to $12 million and the price of gold was suppressed.
All that cheap gold gaveJay Gould an idea. He was friends with Abel Corbin, who just happened to be married to Jennie Grant, the president’s sister. If they could prevail not he president to stop selling gold, the price would start going up. Knowing this in advance they could start buying up gold drilling up the price faster. Done right, they could corner the gold market and get unreasonably rich.
Gould enlisted one of his fellow directors at the Erie railroad, James Fisk into the plant.
On September 1, 1869 they started buying up large amounts of gold under other people’s names and driving up the price. Corbin planted the idea with Grant that selling gold would hurt western farmers and the plan should be suspended. But they got greedy. And when Grant’s personal secretary turned down an offer to open a gold account, they did it anyway. When he told the president about it Grant figured out what was happening. And on Friday September 24, 1869, the government resumed selling gold. Gold prices plummeted. And hundreds of people who were riding the gold wave along with Gould and Fisk, lost everything.
Stock prices plummeted 20 percent between that Friday and October 1st. Brokerage firms went bankrupt. Farmers really did get hurt this time with wheat and corn prices dropping by half. The economic turmoil lasted for months. Anti was all traced back to that one Friday. That Black Friday, in September 1869.
And it was that day that would, somewhat ironically, lend its name to what would become the biggest shopping day in the US.
Let’s help you Know a Little More about Black Friday.

Thanksgiving day was established by the US first constitutional president, George Washington in 1789. On the recommendation of Congress, President Washington proclaimed Thursday, November 26, 1789 as a Day of Public Thanksgiving. Thanksgiving days were proclaimed by subsequent presidents on a regular basis but the dates varied. It wasn’t until 1863 in the midst of the Civil War, that President Abraham Lincoln proclaimed that Thanksgiving should be commemorated on the Last Thursday of November each year.
The regularity made it a nice signpost not he calendar. Retailers began promoting holiday shopping starting the day after Thanksgiving.
That lasted until another economic depression, the great one. In 1939, the economy was showing signs of recovery. But Thanksgiving that year would fall on the very last day of the month. That meant the shortest possible holiday shopping season, meaning the smallest boost to the economy So President Franklin D. Roosevelt issued a proclamation that Thanksgiving would take place on the second to last Thursday of November, adding a week to the shopping calendar.
16 states refused to move the date and for two years, a third of the country celebrated Thanksgiving a week after the rest of the country.
So in October 1941, Congress passed a resolution declaring the fourth Thursday in November to be Thanksgiving. This kept it as the last Thursday most years, unless November happened to have 5 Thursdays. That kept the holiday shopping from getting too small without pushing it so far up the calendar.
Once that pattern was set, the Friday after Thanksgiving started to take on a character of its own. Workers began to call in sick on Friday in order to have a four day weekend. In 1951, the journal Factory Management and Maintenance began to refer to this phenomenon as Black Friday, referencing the panic of 1869. Friday also became a huge shopping day of course, and police in Philadelphia and Rochester began referring it to Black Friday as well because of the crowd management.
But the reference did not become common. The New York Times first called the Friday after Thanksgiving “Black Friday” in tis November 29, 1975 issue referring to the traffic in Philadelphia. But even as late as 1985, the Philadelphia Inquirer reported that merchants Cincinnati and Los Angeles were unaware of the term.
Meanwhile merchants were trying to avoid the usages connotations of a panic and disasters. As early as November 28, 1981, the Philadelphia Inquirer picked up an explanation put out by merchants that it was being called Black Friday because it was when retailers got “in the black” – aka profitable.
But by the late 1980s the term had gained wide acceptance. Retailers across the US began advertising Black Friday sales. More companies began to just give in and give workers Friday off since they were going to call in sick anyway.
By the mid 2000s the day had inspired “Cyber Monday” when workers came into their offices with computers and high bandwidth and shopped for deals online. Giving Tuesday was a counter-celebration to encourage people to spend money on charities instead of products.
The lockdowns because of COVID caused a lot of people to shift to online shopping on all days and by 2021, the Black Friday sales were no longer limited to Friday.
The prevalence of US-based retailers have caused the promotion of Black Friday sales outside of the US, even though those countries do not have the November Thanksgiving holiday. Some countries even promote Black Week or Black Month sales.
The success of Black Friday sales may have inspired Alibaba to co-opt a dating holiday in China called Singles day – on November 11th – to be a big sales day which now has passed Black Friday as the largest shopping day in the world.
Steely Dan wrote a song called Black Friday, released in 1975 just as the New York Times was picking up on the phrase in its post-Thanksgiving context. Steely Dan was writing about the 1869 panic but its words could apply to both

When Black Friday comes
I’ll collect everything I’m owed
And before my friends find out
I’ll be on the road

I hope you appreciate the probably unintentional double meaning. And hope you know a little more about Black Friday.

CREDITS
Know A Little More is researched, written and hosted by me, Tom Merritt. Editing and production provided by Anthony Lemos and Dog and Pony Show Audio. The public key cryptography players were Sarah Lane as Alice, Shannon Morse as Eve and Andrew Heaton as Bob. It’s issued under a Creative Commons Share Attribution 4.0 International License.

Douglas Engelbart demonstrated how two people connected over a network could work together editing a document at the same time in 1968. It took much longer to see a real product.

Featuring Tom Merritt.

MP3

Please SUBSCRIBE HERE.

A special thanks to all our supporters–without you, none of this would be possible. Become a supporter at the Know A Little More Patreon page for exclusive content and ad-free episodes.

Thanks to Kevin MacLeod of Incompetech.com for the theme music.

Thanks to Garrett Weinzierl for the logo!

Thanks to our mods, Kylde, Jack_Shid, KAPT_Kipper, and scottierowland on the subreddit

Send us email to [email protected]

Episode transcript:

Douglas Adams is a hero to computer geeks everywhere. His Hitchhiker’s Guide to the Galaxy right there with Monty Python in the Hall of Geek memes. Legend has it Adams was one of the first people in the UK to get a Macintosh computer. And on that Mac he used a program called FullWrite. We could probably do a whole episode on the rise and fall of FullWrite. It had one of the first true What You See is What You get interfaces, handled long documents particularly well and included an outliner. A true dream list of features for writers in 1988. Scientist and writer Douglas Hofstader was also a big fan. But it was also buggy and eventually it was crushed by Microsoft Word.
So why bring it up? Its lead engineer was a guy named Steve Newman. And in 1989, a young programmer named Steve Schillace joined as a programmer on FullWrite. Those two got to know each other, and later, with the help of a self-described party person, Claudia Carpenter, they would revolutionize how everyone works every day.
Let’s help you know a little more about Collaborative Editing.
That’s Douglas Engelbart in 1968, demonstrating how two people connected over a network could work together editing a document at the same time. It was one of the many impressive technological advancements Engelbart showed off- the mouse, hyperlinks, video conferencing and more. if you missed our episode on that, you definitely want to go back and take a listen.
One of the themes from that episode, that people who did listen already know, is that almost nothing came out of that demo directly. It was oddly too far ahead of its time. But Collaborative editing takes the cake on that point. It would be 35 years or so between Engelbart’s demo of collaborative editing in 1968 and the rise of mainstream use of that technology.
One of the *earliest* examples that had any significant use was Instant Update for the MacOS. And it didn’t arrive until 1991. Multiple users of Instant Update could edit a single document at the same time over a LAN – a local area network– so they had to be in the same location. And it needed a workgroup server to work. It didn’t catch on.
It wasn’t until the internet became widespread that real-time collaborative editing became common. One of the first successful efforts was from a student at Gdańsk University of Technology in Poland. Tom Dobrowolski created Multi-Editoro in 2003, later renamed MoonEdit. It used code from Ken Silverman’s BUILD video game Engine. Up to 14 users could edit at the same time, text from each person was highlighted a different color and it featured infinite undo and a time-slider so you could replay edits. It was available for Linux, Windows or FreeBSD, and it needed somebody to set up a server for it to work.
Then there was SubEthaEdit, named after the communication network in Douglas Adam’s Hitchhiker’s Guide. Oddly it was originally named Hydra but had to change the name to SubEthaEdit for legal reasons. It was put together in early 2003 in an attempt to win an Apple Design award, which it did, at Apple’s WWDC 2003. It was mainly a word processor but you could turn on a collaborative mode that worked without configuration on a LAN or with a little set up, over the internet. It’s still developed and you can get it on GitHib.
But neither Moonedit or SubethaEdit used the Web. And we were fast moving into a world where people didn’t want to download an app if they didn’t have to. That was thanks to Ajax, a name for a collection of web development techniques that could send and receive data in the background on a web page. Without Ajax a web page usually had to reload to significantly update itself. It wasn’t even new in 2003. It had been around since 1999. It wasn’t a language, just a way of using HTML, CSS, and javascript. But in 2003 it exploded in popularity as part of the Web 2.0 movement. Everybody was scrambling to make more dynamic web pages like Google’s Gmail.
Among the folks playing with Ajax were Steve Newman and Steve Schillace. Since last we saw them working on the doomed FullWrite, they had worked together on Claris Home Page, a what you see is what you get HTML editor, and started a company called BitCraft that developed a JavaScript application engine. They sold Bitcraft to Macromedia in 2000 and by 2002 were consulting with Intuit to build a customer manager for QuickBooks.
In the usability lab there they met Claudia Carpenter.
That’s how Carpenter described her teen years to the Clickup podcast. But by 1980 she was studying computer science at Cal State Chico and regularly topping her class. Out of college she worked as software engineer for HP for 6 years before moving over to Intuit where she eventually found herself working on QuickBooks and one day, in a conference room with Newman and Schillace.
After Carpenter had her second child, she wanted to try something new and get out of the office. So she, Newman and Schillace left Intuit to try some stuff. They worked in Newman’s attic room above his garage. A classic Silicon Valley story.
They didn’t know exactly what they were going to do but Schillace later told The Verge they knew they wanted to work with Ajax. In fact they started by just trying to find a way to keep document locking from getting in the way of two people working on the same document and eventually that led to a basic word processor. They referred to it as Docster, but later changed the name to Writely. They bet that users would want speed and the convenience of always having the latest version of a document available, so they wrote a lean 10 pages of javascript that didn’t include more advanced functions like rich formatting, margins or pagination. It was a version of collaborative editing that would have looked familiar to Douglas Engelbart.
They didn’t even include a save button which was one of the earliest complaints from a generation of computer users trained to save regularly or face the peril of a lost document. Carpenter’s brother was an MBA professor and made one of his classes do their assignments in Writely. Students complained about the limited features and lack of save button, so much, so, that Carpenter said they tried to get the dean to overrule her brother.
That experience led them to add the only other employee their small company would ever have, Jennifer Mazzon, another Intuit colleague, who Carpenter convinced to work for them as a second job. Mazzon eventually left Intuit to work on Writely full time.
Writely officially launched in August 2005. How did it go? Again Carpenter talking to the Clickup podcast:
It was popular. And they knew they would need help somehow to keep it working.
The group had always thought that Google was a likely home for their product but they spent the better part of a year being told no by Google then saying no to Google themselves and so on.
Sam Schillace took the meetings with Google and they eventually did lead to Google acquiring Writely on March 9, 2006.
Writely became Google Docs.
And as you know it didn’t stand alone.
That same year, Google acquired 2Web Technologies which was working on a collaborative spreadsheet, and the next year made two acquisitions to create Google Slides.
The popularity of Writely continued as Google Docs. And eventually Microsoft, the titan of word processing, bowed to the pressure and introduced Office Web apps in 2010.
And now collaborative editing is the norm. People use Google Docs and Office for the Web, every day. It’s normal. But it took a while to get there.
And the heroes of our story?
Steve Newman worked for Google as an engineer until 2010, then left for online storage company Box in 2011 and after his stint there, founded a data service company called Scalyr which he left after acquisition in February 2023. He now says on his LinkedIn that he’s “Blogging on AI and Climate Change while I think about what to do next.”
Schillace pursued a lot of projects. Among them he went to Box with Newman in 2010, then came back to Google in 2016 to work on Google Maps. But in September 2021, he left Google again, this time to become a deputy CTO at Microsoft.
And Claudia Carpenter left Google in 2010 to work on her own projects as Restartle & Girl Power Software, then worked with Steve Newman again at Scalyr until February 2021.
On her LinkedIn she says she’s retired but also that she is “absolutely terrible at retiring.”
And that wraps up the last and latest of the innovations of the Mother of all demos to come into the mainstream. This episode was written in Docster, wait no Writely, no, Google Docs.
In other words, I hope you Know a Little More about collaborative editing.
CREDITS
Know A Little More is available without ads to direct supporters at patreon.com/knowalittlemore. It was researched, written and hosted by me, Tom Merritt. Editing and production provided by Anthony Lemos and Dog and Pony Show Audio. It’s issued under a Creative Commons Share Attribution 4.0 International License.

By 1998 the US had passed its Digital Millennium Copyright Act. And partly because the US generates so much copyrightable material, and partly just because it’s the US and is a little pushy on the world stage, the DMCA became the de facto way of handling copyright protections on the internet around the world.

But what is it? Why did we need the DMCA or the WIPO copyright treaty at all?

Let’s help you Know a Little more about the DMCA.

Featuring Tom Merritt.

MP3

Please SUBSCRIBE HERE.

A special thanks to all our supporters–without you, none of this would be possible.

Thanks to Kevin MacLeod of Incompetech.com for the theme music.

Thanks to Garrett Weinzierl for the logo!

Thanks to our mods, Kylde, Jack_Shid, KAPT_Kipper, and scottierowland on the subreddit

Send us email to [email protected]

Episode transcript:

It’s April 26, 1970. Joe Cocker is playing live at the Fillmore. The Jackson 5’s ABC is dominating the charts In Novo Mesto, Slovenia, little Melanija Knavs is born. And after three years of planning, the World Intellectual Property Organization has begun operations. The purpose of the specialized agency is to provide a place for countries to work together on their various intellectual property laws and rules. Copyright is of course the most well known type of intellectual property these days but it also includes trademarks and patents and such. WIPO is meant to be a clearing house. A place to try to harmonize. I’ll respect your patents if you respect mine etc. In fact its first big achievement is the Patent Cooperation Treaty which to oversimplify, made filing a patent in one country equivalent to filing in all. Now different countries still had latitude to approve or deny patents according to their own laws, but it made things a lot simpler.
WIPO made lots of other treaties and systems to make it easier to handle trademarks and service marks. It created mediation and arbitration to help resolve disputes between countries over these kinds of matters.
And in September 1995 it took up the digital agenda. Copyright came to the fore. And somehow. Some way, WIPO agreed on new rules faster than it almost ever agreed on anything. By December 1996 there was a diplomatic conference to approve the WIPO Copyright Treaty and the WIPO Performances and Phonograms Treaty.
Those two treaties brought countries together to agree on how to handle digital copyright protection. Each country then had to pass its own law to implement the treaty.
By 1998 the US had passed its Digital Millennium Copyright Act. And partly because the US generates so much copyrightable material, and partly just because it’s the US and is a little pushy on the world stage, the DMCA became the de facto way of handling copyright protections on the internet around the world.
But what is it? Why did we need the DMCA or the WIPO copyright treaty at all?
Let’s help you Know a Little more about the DMCA.

Since the Internet became more than just something university IT experts used, worries about copyright violations on the internet have existed.
Digital content is infinitely copyable and the internet makes it infinitely transferable. That’s a nightmare for businesses built on physical limitations to copying, like music, movies and others.
To extend these older business models onto the internet, companies use digital rights management or DRM. This is a name for varying ways of trying to lock up content so that only a user who is authorized to view it can. It’s an attempt to make it not be infinitely copyable. DRM is tricky though because you have to balance access for the person who does have the right– like a paying customer– with denying access to anyone who doesn’t. Those are cross-purposes. If you leave a door open for authorized viewers, eventually unauthorized viewers will figure a way into it.
So the industry quickly turned to the law, and we get the Digital Millennium Copyright Act. Or DMCA. While this is only a law in the US, it affects anyone who publishes content in the US, such as on YouTube and has provided a model for laws like it around the world.
The problem it solves is that no matter what digital locks you put on a file, someone can figure out a way to break them. So the law fixes this by making it illegal to break them.
That’s one of the main misunderstandings about the DMCA. It doesn’t just make unauthorized access illegal. That was already illegal under copyright law. It makes circumventing access protections illegal, punishable by fines and imprisonment.
Copyright holders can seek up to $2,500 per violation, or statutory damages up to $25,000. Repeat offenders can face more. If you are accused of willfully violating the DMCA for personal or commercial financial gain, you can be tried as a criminal offender. A first-time criminal DMCA violator can face a fine of up to $500,000, up to five years in jail, or both. Repeat offenders can be fined up to a million dollars and up to ten years in prison.
Screen capturing is a circumvention of the DMCA in some cases. Keep that in mind.
The DMCA was passed as an amendment to the US Copyright Act in 1998. It implemented those two 1996 treaties of the World Intellectual Property Organization.
It makes it illegal to produce or disseminate (even if you give them away free) any device or service INTENDED to circumvent measures that control access to copyrighted works. Courts decide whether a device or service is intended to do this. Because you know, computers can do this, but it’s not their sole intention. And that’s why screen-capturing software is not just illegal.
The other aspect of the DMCA is it makes it illegal to circumvent access control EVEN IF copyright is not infringed. Yep. If you have a fair use for something, like making a backup of a DVD, it is illegal under the DMCA to circumvent copyright protection in order to make fair use of that backup. The DMCA includes some limited exemptions such as for security research and government research but they are few.
Now if you’re saying hold on I thought they changed that and made DVD copying legal. We’ll get to that later but yes and no.
There are a couple more aspects of this to keep in mind. One, is that the United States Copyright Office ( part of the Library of Congress) was given the power to create (and get rid of) further exemptions to the DMCA. So it can restore fair uses on a case by case basis. More on that later.
And then there’s a safe harbor for platforms. Online Service Providers, which includes platforms like YouTube and Facebook– are exempt from liability for their users copyright infringement as long as they follow certain procedures. Platforms keep their safe harbor by promptly blocking access to infringing material once they are notified of an infringement claim. This called the “notice and takedown” process. It also provides for a counter notification from a user who claims the material is not infringing.
There’s also an exemption for a repair person who makes limited copies solely for the purpose of repairing a machine. In other words, imaging a drive to restore it on a replacement drive doesn’t violate the DMCA. There are also some provisions for distance education, ephemeral copies made in the process of broadcasting and more.
DMCA’s Title V is my favorite. Title V provides protection for Boat Hull designs because Boat Hull designs are not covered by copyright as they cannot be separated from their useful function and therefore are better protected by patents than copyright. This section of the DMCA was added in 1998 after the Supreme Court ruled — in Bonito Boats, Inc. v. Thunder Craft Boats, Inc. –that Boat Hulls did not have copyright protection. So, immediately boat manufacturers lobbied congress to add the exemption to the DMCA. As of 2019 there have been 538 applications for registrations for Boat Hull designs under the DMCA, compared to more than 70,000 patents granted.
OK back to the notice and takedown system.
The notice and takedown system is governed by Section 512 of the DMCA.
In order to get the safe harbor protection, a service provider has to have an agent on file who takes notifications. The provider can’t have reasonably known about the infringing activity or directly benefit financially from infringing activity. In other words your main business can’t be infringement.
Ok. Now you’re a safe harbor protected platform. How does it work if somebody thinks their copyright has been infringed on your platform?
Well it works differently for every system but here are the parts required by Section 512.
The notifier must send a formal takedown request notification under penalty of perjury. They can’t knowingly lie about it.
Once a notice is received, the provider must “expeditiously take down or block access to the material.” right away. No grace period. It must also promptly notify the user that the content has been removed or disabled.
The user can then file a counter-notification, also under penalty of perjury, that its content was identified as infringing through a mistake or misidentification.
That sends it back to the notifier. If they do not file a court order against the user, the provider must restore the content within 10-14 days.
So yes. Send a takedown notice the content goes down immediately. Send a counter notice it takes 10-14 days to get it back up.
So you could abuse the system by just sending notices for anything you wanted to disappear from the internet for a couple of weeks right?
Well, those perjury conditions are meant to keep the system from being abused but in practice they’re hard to prove. Just being mistaken is not the same as perjury so you have to prove that a company KNEW the content was not infringing when it sent the notice, not just that it was mistaken. And end users are much more likely not to want to risk a perjury lawsuit than the large companies who send bulk notices, so most takedown notices are successful. But willful and malicious abuses are rare. Mistakes however, are rampant. Lots of companies have been accused of sending inaccurate bulk takedown notices, sometimes ending up affecting their own employees. But that’s not the same as perjury.
Also there is a chilling effect of the DMCA. A content-hosting platform can avoid falling afoul of the DMCA by just not hosting some material altogether. It’s not required to host it. So some companies, like YouTube have employed “informal” takedown notices that are not meant to be the legally required notices. These are usually constructed as terms of service violations. This lets them take down content without risking the perjury charge. Companies have the right to operate outside the DMCA in this way because the law can’t force them to host content they don’t want to. A copyright holder is only subject to perjury restrictions if they are following a “formal” takedown procedure. YouTube does have a method of proceeding from informal takedowns to formal ones.
For years YouTube used a bot system called ContentID to look for possibly infringing content. If the bot thought it saw a match to a database of content provided by big copyright holders, it would pull the content off the site and notify the user it had been pulled. This was not part of the DMCA.
If the user disputed the Content ID claim, YouTube would then contact the alleged rights holder. The rightsholder could release the claim and the content would go back up or could uphold the claim and the user would be notified that the rights holder still claimed the content was infringing and it would stay down. This was partly DMCA as this could also serve as the rightsholder’s formal Takedown notice. But since the bot had identified the content as infringing the risk of perjury for the rights holder was almost nothing.
If the user did not have an account in good standing or had already appealed three other claims that was it. The DMCA never entered into it for the user. YouTube just declined to host the content because they didn’t want to.
However if the user was in good standing and had not reached the appeal limit a DMCA counterclaim would then be issued to the rightsholder — with the risk of perjury for the user still there– and the normal DMCA takedown procedure would take place. The rightsholder would then have to decide whether to pursue it in court or not.
As I mentioned earlier the US Copyright Office can make exemptions to the DMCA. It regularly reviews exemptions and can add, extend or remove them.
The Copyright Office has issued exemptions to the DMCA over the years. Here’s a look at a few of them.
The first two in 2000 were for website filtering — you know like safe sites for kids kind of stuff– and preservation of damaged or obsolete software and databases.
In 2003 an exemption was given to screen readers for e-books and one for video games distributed in obsolete formats.
A brief exemption was given in 2006 for sound recordings protected by software with security flaws, specifically the Sony Rootkit. And one for unlocking wireless phones.
In 2010 an exemption for breaking DVD’s Content Scrambling System was issued for educational, documentary, noncommercial or preservation uses. Security testing of video games.
In 2012 an exemption for excerpting short portions of movies for criticism or comment was given.
In 2018 one for 3D printers if the sole purpose is to use alternate feedstock. As well as ones to expand exemptions for preservation and security research.
In October 2021, an exemption was given for repairing any consumer device that relies on software as well as medical devices and land sea and air vehicles even if they aren’t consumer-focused.
What if you’re outside the US? Why should you care? On the one hand, you’re right, US law doesn’t apply outside the US. However copyright owners from outside the US can still issue takedown notices on US sites. But the bigger thing to remember is that the DMCA is the US implementation of the WIPO Copyright Treaty and the WIPO Performances and Phonograms Treaty. The WIPO Copyright treaty was signed by 110 countries and most members of the World Intellectual Property Organization or WIPO have agreed to accept DMCA takedown notices. Think of it like this. A country adopted the WIPO treaties, the US created a system to enforce it and the country just borrows that system. It’s not that US law is enforceable in their country it’s that the US enforcement system of the WIPO treaty is a nice prepackaged way to do things. Copyright-enforcement as a service!
Some countries however are known as DMCA ignored countries. These are countries who either have not agreed to WIPO’s provisions, systematically ignore those provisions or prioritize their own copyright laws over those of the US and so websites do not honor DMCA requests.
These include Russia, Bulgaria. Luxembourg, the Netherlands, Hong Kong, Singapore, Malaysia, Switzerland, and Moldova. They are often promoted as places to host websites if you’re concerned about copyright infringement, though each carries its own set of concerns either with local laws or political speech. China doesn’t necessarily honor the DMCA, but has enough other restrictions that it’s not generally not included on these lists.
Nobody loves the DMCA but it has proved to be surprisingly stable. It’s next big test will be machine generated works like ChatGPT and the multiple text-to-image generators.
So far the discussions have been about where copyright applies but that is going to drift into the DMCA and put its uneasy equilibrium to the test.
For example, in April 2023 an unknown composer created a song and used some machine generation to make it sound like Drake and the Weeknd. The song lyrics and beats were original but the artist had used a producer tag that was not. Universal Music Group used that producer tag as the basis for copyright takedowns. But versions without the tag would force the issue.
That’s the first not the last of what will be a long discussion about where machine-generated works fall in copyright. How that discussion plays out will likely determine whether the DMCA stays standing, gets modified or rewritten altogether.
So that is the Digital Millennium Copyright Act, aka the DMCA.
It makes it illegal to circumvent copyright protection unless there is an exemption written in the act itself, or added by the US Copyright Office.
It also provides a way to try to get infringing material removed and a way for a user to combat having that material removed.
I hope this helps you understand why some content is allowed up and some is not and why you don’t see some content at all.
In other words, I hope you know a little more about the DMCA.

CREDITS
Know A Little More is researched, written and hosted by me, Tom Merritt. Editing and production provided by Anthony Lemos in conjunction with Will Sattelberg and Dog and Pony Show Audio. It’s issued under a Creative Commons Share Attribution 4.0 International License.

Taiwan is at once one of the most vexing political situations on the globe and one of the most important to the world of technology.

But few people understand how it got to be either. And understanding that is essential to understanding what might happen next and how that matters a LOT for the technology industry.

Let’s help you Know a Little More about Taiwan.

Featuring Tom Merritt.

MP3

Please SUBSCRIBE HERE.

A special thanks to all our supporters–without you, none of this would be possible.

Thanks to Kevin MacLeod of Incompetech.com for the theme music.

Thanks to Garrett Weinzierl for the logo!

Thanks to our mods, Kylde, Jack_Shid, KAPT_Kipper, and scottierowland on the subreddit

Send us email to [email protected]

Episode transcript:

Taiwan is at once one of the most vexing political situations on the globe and one of the most important to the world of technology.
But few people understand how it got to be either. And understanding that is essential to understanding what might happen next and how that matters a LOT for the technology industry.
Let’s help you Know a Little More about Taiwan.
This is not going to be a Dan Carlin-style dive into the history of Taiwan. If you know that history consider this a refresher. But for those of you who know little about the island, consider this an excellent starting point to understanding it. And for all of you, I don’t think understanding it is likely to get less important in the coming years. Because it’s one of the places on Earth where it’s conceivable to see a war involving China and the US. AND it’s one of the most important places in the world for building technology. Chips are in everything these days and the chips are made mostly by companies from Taiwan.
Let’s start with the where.
Taiwan is 168 islands including the Penghu islands but it is mostly one main island with the three main cities, Taipei, Tainan and Taichung. It’s about the size of Vermont. Or Albania.
It is located partway between the Philippines and South Korea, but very close to mainland China. It is 160 kilometers off the coast of southeastern China. About the distance from Dublin to Belfast. You could not fit Ireland between Taiwan and the mainland.
OK so why it’s important is both the tech industry, which we’ll get to later AND the dispute over it.
Let’s start with the dispute over what Taiwan thinks it is and what China thinks it is. Because Taiwan thinks it is China. This is one of the most common confusions I hear from people.
Taiwan’s government officially calls the country, the Republic of China. Well that’s odd, you might say. Isn’t there already a Republic of China? Yes. The People’s Republic of China. That’s the one most people think of as China. The one with its capital in Beijing. Both the People’s Republic of China and the Republic of China, which is on Taiwan, consider themselves the legitimate successor to the Republic founded in China on January 1, 1912 after the overthrow of the Qing dynasty. Where they differ is that Taiwan considers itself the true continuation of that republic and the People’s Republic of China says that republic ended in 1949 and was replaced by the People’s Republic.
So what Taiwan is, depends on who you ask? The People’s Republic of China – to oversimplify– says Taiwan is a breakaway province that is part of the People’s Republic. Eventually it needs to stop denying that fact, cooperate with the central government and unify with the mainland. Hence China’s strong objections to calling Taiwan a country, or having full diplomatic relations with it. The US wouldn’t want anyone having diplomatic relations with Texas, or Hawaii. The Uk doesn’t let Scotland go have separate diplomatic relations with other countries.
Meanwhile, the government of Taiwan still considers itself the rightful ruler of all of China. Hence its insistence on officially calling itself Republic of China.
And so you get weird situations like letting Taiwanese athletes compete separately at the Olympics, but only if they call themselves Chinese Taipei and use the Olympic flag.
There are other similar arrangements. For example England, Scotland, and Wales, all part of the UK, compete in World Cup competition as separate teams. The difference being that they are not all calling themselves the true UK.
I bring it up to illustrate the point the People’s Republic of China is making. If Taiwan is just a province of China, then it’s not odd to let it compete separately in things. So call it under a provincial name, and throw Chinese at the front just so people are clear. Taiwan goes along with this so its athletes can compete separately and they consider themselves China as well, so why not call them Chinese Taipei. It’s WAY more complicated than that but you get the gist and it kind of helps illustrate how seriously these countries take the “on paper” meanings of this dispute.
One thing the two countries agree on is that the Republic of China started in 1912.
Sun Yat-sen was the the founder and first provisional president. He is honored by both the People’s Republic of China and the Republic of China on Taiwan for ending the rule of China’s imperial dynasties. But it didn’t result in stability right away. China’s political history in the 1920s and 1930s is full of disputes between the Nationalist Party – aka the Kuomintang and the Communist Party. Sometimes those disputes became battles. The two parties teamed up in World War II to fight their common enemies, but never fully unified. The People’s Republic of China consider the Republican era to have ended on October 1, 1949 with the proclamation of the People’s Republic of China. The Republic of China in Taiwan, disagrees.
So that’s where the Republic of China and the People’s Republic of China come from.
Let’s talk a little about Taiwan and how it became part of China in the first place.
The smaller Penghu islands off the coast of the main China were often under mainland sway, the larger island though was somewhat independent. In 1622 the Europeans arrived, first the Dutch, then the Spanish. The Dutch called it Ilha Formosa or “the beautiful island” which was shortened to Formosa, which became the European name for the island. The Chinese didn’t like the idea of the Europeans getting too involved there. So it was finally annexed by China’s Qing Dynasty in 1683. There were attempted invasions over the next couple of centuries by Japan and the French, but Taiwan remained in Chinese control for a good two centuries. Then at the end of the war between China and Japan in 1895, Taiwan was ceded to the Japanese Empire, where it remained until the end of World War II.
So it was loosely affiliated with China until the late 1600s, then solidly part of China for a couple centuries, then occupied by the Japanese for 50 years. So it made sense in 1945 that it would go back to China.
In July 1945, The US, UK and China agreed to the Potsdam Declaration. Among its many provisions was that the islands of Taiwan would be restored to China.
On August 15, 1945 Japan’s Emperor accepted the terms of the Potsdam declaration and on October 25, 1945, Japan surrendered. After the surrender of Japan, Japan’s governor-general of Taiwan signed papers handing over administration of the island to General Chen Yi – a Nationalist– of the Republic of China.
One technicality, nowhere did Japan confirm in writing they were giving up their claim to Taiwan. There was no cause for concern on that in reality but it was a detail that needed to be taken care of. An i to be dotted. A t to be crossed. There were a lot of those. For example Japan technically remained at war, even after the surrender. Not something the matter in practice but you kind of wanted everyone to be clear on the point right? There were some official treaties in the works to nail down all those technical details.
Problem was while the paperwork was getting drawn up, China was having a civil war.
The Communists led by Mao Zedong an and the nationalists led by Chiang Kai-shek no longer united by an external foe, started battling for control of the country. And while the allies had been doing all the paper work with the Nationalist, the communists started winning the civil war. Mao felt confident enough to proclaim the People’s Republic of China as a replacement for the Republic on October 1, 1949 and by December 7, Chiang and the nationalists evacuated their army to Taiwan and set up a capital in Taipei. About 2 million Chinese people and soldiers made the move to Taiwan.
Meanwhile there were all those little things that hadn’t been taken care of regarding the end of the war with Japan, like compensation and rebuilding and oh actually declaring the war over.
So the Treaty of San Francisco was created to wrap up all those details and was signed by Japan on September 8, 1951. Among its many provisions, Japan formally renounced its claim to Taiwan.
Great. Except, China didn’t sign it. Because by that time there were two governments claiming to represent China, Mao’s on the mainland and Chiang’s in Taiwan. Chiang had held strong in Taiwan and with US support continued to claim to be the rightful government of China. There was some recent experience with supporting exile governments. An exile government of France had held out in England and recently been restored. So there was some feeling the same might happen in China.
Meanwhile the USSR wanted to support its communist comrades, and argued that Mao had won and so should be recognised as the legitimate government.
And meanwhile Japan just wanted China, any China, to sign something declaring the war over.
To solve that, on April 28, 1952, Japan and the Nationalist Republic of China government on Taiwan, signed the Treaty of Taipei, formally ending the war between Japan and the Republic of China in Taiwan. Not the mainland. But it was enough to satisfy Japan.
And there’s another little wrinkle to this. Back during the war in 1942– when everybody was on better terms, Soon Tse-vung, who’s sister was married to Sun Yat-sen- the man who founded the first Republic of China in 1912, signed a document along with Soviet diplomat Maxim Litvinov and US President Roosevelt and UK Prime Minister Churchill, which later became the basis of the United Nations Declaration. That document gave those four countries a special position in the formation of the UN. And so, the US, UK, Soviet Union and US were guaranteed to be on the UN’s permanent security council.
When the UN was founded in 1945, China got its seat. The civil war was just heating up and Mao hadn’t proclaimed the People’s Republic, so Chiang got the seat.
There was some talk about dual representation maybe but that ended in 1949 when the People’s Republic of China was founded and Chiang moved to Taiwan. So Chiang held on to it. And until 1971, Taiwan’s government held China’s seat at the UN.
It was a perilous situation though. Both countries strenuously called for there to be just One China. With the US and USSR facing off with nuclear weapons at the height of the cold war, it seemed unwise for a huge communist country like China to have no seat at the biggest diplomatic table in the world. It was counterproductive to world stability.
All countries wanted a better solution to this. I’m way oversimplifying her of course but that’s the general situation that led to US president Nixon secretly sending National Security Adviser Henry Kissinger to China’s Premier (under Mao) Zhou Enlai. He was often seen as the successor to Mao, and his ally Deng Xiaoping went on to govern China in the 1980s. In a talk between Zhou and Kissinger on July 9, 1971, Kissinger made clear that “we are not advocating a `two Chinas’ solution or a `one China, one Taiwan’ solution.” Zhou said “the prospect for a solution and the establishment of diplomatic relations between our two countries is hopeful”
That was good progress. Way to go Hank.
And on July 15, 1971 President Nixon announced he would visit the People’s Republic of China the following year. Remember that Nixon’s US has been fighting a proxy war against China in Vietnam. This is a huge shocking announcement.
Then, on October 25, 1971 a coalition of Soviet bloc and non-aligned countries, along with the UK and France, voted to give the People’s Republic of China the UN seat in place of Taiwan. The vote was initiated by Albania – you know Albania, the one about as big as Taiwan–
The US acted upset. But Nixon already said he would go to China. And he did. February 21st, 1972, US President Nixon began a 7 day visit to three cities in China, including a meeting with Chairman Mao Zedong. Mao told Nixon, “I believe our old friend Chiang Kai-shek would not approve of this”.
US TV audiences get a new look at China and we got the phrase “Only Nixon could go to China.”
The visit changed things for Taiwan too and got us closer to the odd situation we’re in now. The meetings resulted in the Shanghai Communique. The US acknowledged that “all Chinese on either side of the Taiwan Strait maintain there is but one China” but for now to set aside the “crucial question obstructing the normalization of relations.” Clever little diplomatic sidestep that let them be friends or at least friendlier, with both Chinas.
In fact, the US maintained formal relations with the Republic of China in Taiwan until 1979.
Because in 1978 China’s Communist Party really sweetened the deal. It declared that China was in a united front with the US, Japan and Western Europe against the Soviet Union. It supported US operations in communist Afghanistan against the USSR-supported regime there, and China conducted a military expedition against the US’s old nemesis Vietnam. China did all that.
So what are you going to do? On January 1st, 1979, US President Jimmy Carter and Zhou’s old friend Chairman Deng Xiaopeng issued the Joint Communiqué on the Establishment of Diplomatic Relations. This ended US recognition of the Republic of China in Taiwan and established formal relations with the People’s Republic of China. It also ended the Mutual Defense Treaty with the government on Taiwan.
So the US just up and abandoned Taiwan? No.
Not everybody was pleased with the President just ending the defense pact with Taiwan.
You see, the Mutual Defense Treaty with Taiwan had been passed by the Senate in 1954, and the Senate, particularly Senator Barry Goldwater, figured they were the only ones who could undo that. So Senator Goldwater brought a case to the Supreme Court as Goldwater v. Carter. But the court basically said our name’s Paul this is between y’all.
It issued a dismissal based on the fact that the case was a political matter not a judicial one and would not rule on it. The legislative and executive branches need to work this out amongst themselves. In fact, Justice Powell wrote in a concurrence that if the Senate had issued a resolution objecting to the dissolution then it would become a matter for the courts. The Senate had drafted a resolution but did not vote on it.
So that’s what the US Congress did. It went to work on making some laws. And on April 10, 1979, the US enacted the Taiwan Relations Act.
It defined how the US sees Taiwan separately from the People’s Republic of China and has shakily guided international relations around the two countries for decades.
The act refers to the “governing authorities of Taiwan” avoiding the whole issue of who gets called Republic of China. It did not restore diplomatic relations with Taiwan nor did it recognise its government. Doing either of those would have undone the last decade of warming relations instantly.
So no, we will not recognise Taiwan’s government. Instead the Act said Taiwan would be treated under U.S. laws the same as “foreign countries, nations, states, governments, or similar entities”. And the American Institute in Taiwan will not at all be an embassy but it can do anything embassies do. And all agreements made with Taiwan’s Republic of China before 1979 stay in effect with the governing authorities of Taiwan. Except the mutual defense pact.
Which you’re probably thinking was Senator Goldwater’s whole sticking point right? Yes. So here’s what the Act did do. It said “the United States will make available to Taiwan such defense articles and defense services in such quantity as may be necessary to enable Taiwan to maintain a sufficient self-defense capability”, and “shall maintain the capacity of the United States to resist any resort to force or other forms of coercion that would jeopardize the security, or social or economic system, of the people on Taiwan”
In other words, don’t call it a country, but treat it like a country, don’t call it an embassy but use it like an embassy and don’t call it a defense pact, but make sure Taiwan is defended.
And crucially through all of this, never once has the US recognized the People’s Republic of China’s sovereignty over Taiwan.
This approach has been called Strategic Ambiguity.
And it worked. Sort of. It still pissed off China. China’s official position is that the Taiwan Relations Act is “an unwarranted intrusion by the United States into the internal affairs of China.” Deng Xiaopeng viewed the US as insincere. A feeling carried on and amplified by subsequent leaders. And over the years, the PRC drifted away from being united with the US against the USSR to aligning with developing nations.
But the US has not backed off of the strategic ambiguity of the TRA.
It reaffirmed the TRA with a nonbonding-resolution in the 1990s, a Congressional Research Service Report in 2007, and a concurrent resolution in May 2016.
And for its part Taiwan has pursued its own strategic ambiguity. You’d think it would have declared itself an independent country but it hasn’t. never declared its independence. In its early days this made sense. You don’t declare independence from something that doesn’t exist. In Chiang Kai-shek’s view, his was the legitimate government of China. There was nothing to declare independence from,
However since the US recognition of the People’s Republic of China, Taiwan’s insistence on this point–the one china policy– has lessened. If it ever abandoned that policy, theoretically Taiwan could just declare itself Taiwan, not China, and seek recognition from world governments.
China would not be OK with that.
Worried about the rising possibility of Taiwan admitting reality, China passed a law on March 14, 2005, restating that there is only one China, Taiwan is part of it, it’s illegal for Taiwan to secede from China, all means to peaceful reunification should be pursued, that under unification, Taiwan would get a lot of autonomy, BUT if Taiwan declares itself independent, or is taken over by another country, OR if all possibility of peaceful unification is lost, China will take non-peaceful actions. The law also states if it does go “non-peaceful” it must do so while protecting Taiwanese civilians and foreigners as much as possible, as well as Taiwanese interests in the PRC. Pay attention to that last. Because that includes Foxconn and TSMC plants.
Yeah about that. Why are so many Taiwanese companies operating in China.
Relations between Taiwan and China cooled off quite a bit in the 1990s and the two decided to ignore their diplomatic differences and focus on economic ties. By 2002, China was Taiwan’s largest market for export.
China hosts around 4,200 Taiwanese enterprises and more than 240,000 Taiwanese work in China. This dependence on China’s economy has been described as a blessing and a curse. On the one hand it has made Taiwan dependent on China, which gives the People’s Republic leverage over it. On the other hand, close economic ties make military intervention more costly.
Taiwan’s economic success is largely down to tech. The Taiwan Semiconductor Manufacturing Co. or TSMC founded in 1987 has a market cap equal to 90% of Taiwan’s GDP. It is in the top 10 largest companies in the world by Market Cap and a bigger semiconductor manufacturer than Intel or Samsung.
TSMC’s customers include Apple, Qualcomm, Nvidia, Broadcomm, AMD, Ampere, Microsoft, MediaTek and Sony. It makes about 60% of the world’s semiconductors.
Other major tech companies headquartered in Taiwan include Acer and Asus which make devices like Phones, laptops, PCs and more. And Foxconn- which also lists on the stock market as Hon Hai, and is famous for assembling Apple products in its mainland China-based factories, but also makes products for Microsoft, Amazon, Google and Huawei with factories located in Brazil, India, Vietnam and all over Southeast Asia.
Taiwan makes the most important part of arguably the most important devices for the world’s economy.
OK that’s not even a very deep look at Taiwan but it’s still a lot, so let’s summarize.
Taiwan’s current government originated on mainland China as one side of a civil war. Taiwan operates under the fading narrative that it is the true government of China. Only 12 countries, mostly in Micronesia and the Caribbean, have full diplomatic relations with Taiwan.
However it’s de facto treated like a country by the US and others but not fully recognised, as a way to placate mainland China which asserts that Taiwan is just a breakaway province that needs to be reunified.
Since the 1990s, economic interests have superseded diplomatic disagreements to the benefit of pretty much everybody. China got Taiwanese investments. The US got a cheap place to buy parts and assemble electronics. And Taiwan became dominant in the chip industry.
Not to oversimplify the country’s economy but Taiwan is the engine that drives chipmaking. If Taiwan’s companies suddenly disappeared, it would be a LOT harder to make electronics ANYWHERE in the world.
And the US has been able to pull off a magic trick keeping mainland China happy while sheltering Taiwan.
BUT the “strategic ambiguity” is beginning to wear thin. A stricter regime in China is pressing the issue more and is less placated by economic benefits.
From here, you need experts in international relations to explain things to you. But hopefully you have a good grip on the basics with which to understand what’s going on.
In other words, I hope you Know a Little More about Taiwan.
CREDITS
Know A Little More is researched, written and hosted by me, Tom Merritt. Editing and production provided by Anthony Lemos in conjunction with Will Sattelberg and Dog and Pony Show Audio. It’s issued under a Creative Commons Share Attribution 4.0 International License.

There’s a lot of fear uncertainty and doubt being spread about OpenAI. So let’s help you straighten out what it is and what it isn’t.

Featuring Tom Merritt.

MP3

Please SUBSCRIBE HERE.

A special thanks to all our supporters–without you, none of this would be possible.

Thanks to Kevin MacLeod of Incompetech.com for the theme music.

Thanks to Garrett Weinzierl for the logo!

Thanks to our mods, Kylde, Jack_Shid, KAPT_Kipper, and scottierowland on the subreddit

Send us email to [email protected]

Episode transcript:

OpenAI is BIG in the news these days what with ChatGPT, GPT-4, its partnership with Microsoft and mounting criticisms from multiple corners.
You may have heard it’s a non-profit. Or that it used to be and now it’s not. Or that it was supposed to open source things and now it’s not.
There’s a lot of fear uncertainty and doubt being spread about Open AI. So let’s help you straighten out what it is and what it isn’t.
Let’s help you Know a Little More about OpenAI.

OpenAI was founded Dec. 10, 2015 with funding donated by Sam Altman, Greg Brockman, Reid Hoffman, Jessica Livingston, Peter Thiel, Elon Musk, Amazon Web Services (AWS), Infosys, and YC Research.
It’s other founding members were scientists and engineers, research director Ilya Sutskever, as well as Trevor Blackwell, Vicki Cheung, Andrej Karpathy, Durk Kingma, John Schulman, Pamela Vagata, and Wojciech Zaremba.
Its advisors were Pieter Abbeel, Yoshua Bengio, Alan Kay, Sergey Levine, and Vishal Sikka.
And its co-chairs were Sam Altman and Elon Musk.
OK. That’s a lot of names. I can summarize by saying they mostly make up AI researchers from academia and places like Google and Facebook, or in some cases went on to work at Google and Facebook. Some are still there, some are not and some don’t seem to list their time at OpenAI.
The point being OpenAI made an effort to find to people in the field from all parts of the industry, that were really good at this. And the two driving visionaries of them were the last two names I mentioned. Sam Altman and Elon Musk. We could spend a lot of time talking about Reid Hoffman and Peter Thiel and their ties to Musk being former PayPal folks. And Greg Brockman is an interesting guy from North Dakota and joined Stripe as a founding engineer in 2010 and became CTO there in 2013. He was the first CTO at OpenAI and is now its President.
But I want to focus on Musk and Altman.
Elon Musk you probably know. Born in South Africa, founder of the first federally-insured online bank X.com which in 2000, merged with Confinity, makers of PayPal. Oh right. You might know him more for companies he invested in and bought like Tesla and Twitter. Or companies he founded later like SpaceX.
You might know less about Altman. Born in St. Louis. Went high school at Burroghs out in Ladue. Founded the social networking app Loopt in 2005 and sold it for $43.4 million in 2012. He was then president of Y Combinator in 2014. And he was the CEO of Reddit for 8 days in 2014 between Yishan Wong and the return of Steve Huffman.
Why these two? Well Altman is CEO of OpenAI. And Musk? He is the magnet and Altman’s the steel. [Brief Walter Egan music?]]
Let me explain, While CEO of Y Combinator, Altman began having conversations with Musk, sometimes recorded for the public, about AI. They both shared a concern that it was expanding too rapidly and companies in charge of it weren’t paying enough attention to the risks and to responsible development. They both believed AI could be one of the greatest benefits to humanity but also one of its greatest threats.
They weren’t the only ones thinking along these lines so they gathered together some like-minded folks I mentioned earlier. People concerned with ethics and responsibility. And from the beginning it leaned toward idealism.
OpenAI Incorporated was founded as, and still is, a 501(c)(3) nonprofit. From its beginning it reflected the concerns of Musk and Altman, writing on its website “It’s hard to fathom how much human-level AI could benefit society, and it’s equally hard to imagine how much it could damage society if built or used incorrectly.” OpenAI said it wanted to “to advance digital intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate financial return.” It took $1 billion in investment and said it expected to only need to spend a small amount of that over the next few years.
But it had to spend more than it anticipated. AI researchers are paid a lot. OpenAI persuaded talent on its mission, its ethics and responsibility, paying better than nonprofits usually did, but less than Facebook or Google. Founding engineer Zaremba told Wired he turned down offers two three times his market value to work at Open AI.
So the people weren’t cheap. Also the cloud computing wasn’t cheap. Reuters reported that OpenAI spent $7.9 million about 25% of its budget on cloud computing in 2017.
If they wanted to make more progress they needed money to attract top talent and to be able to run more complex experiments.
So you can imagine that after the first few years, OpenAI is starting to wonder about that non-profit status. It’s got to make some hard decisions about all that openness too. I mean they’ve done some impressive things training video game bots with OpenAI Gym and Universal, but is that going to move the needle. They’re the ones doing this responsibly, but what does that matter if the big companies stay so far out in front. If they really want to advance AI, if they really want to be the ones protecting humanity and pushing for responsible development, they’d need more, right? So how do you do that and stay true to your core principles?
This was clearly important for Musk. Not long before the founding of OpenAI, he had told students at MIT that AI was humanity’s biggest threat.
So what happened next was surprising if not shocking. So why did it get downplayed?
On February 20, 2018 Elon Musk announced he was leaving the board of Open AI. In a blog post announcing new donors to the non-profit, Open AI wrote, “Additionally, Elon Musk will depart the OpenAI Board but will continue to donate and advise the organization. As Tesla continues to become more focused on AI, this will eliminate a potential future conflict for Elon.”
Well OK. He still believes in the mission, but he’s got his own AI at Tesla to develop so he probably shouldn’t be a director at a competitor, even if it is a nonprofit one. I mean it’s no hard feelings right? Musk even spoke to OpenAI employees to explain that conflict of interest before he left.
But the employees didn’t really seem to buy it. And the line announcing it, was buried at the end of a long first paragraph in a three paragraph post about other funders. Seems like a bigger deal than that no?
Well, maybe it was.
You see, everybody had a solution for that problem OpenAI had of falling behind. Musk’s solution was himself. Put me in charge! Let Musk run the show and he’d catch up. Just look at what he did to the auto industry right?
You may have heard that Musk can be a little — enthusiastic. Maybe rubs people the wrong way sometimes. That seems to have been the case with Open AI’s other founders. Maybe they were also annoyed that Tesla had taken one of those founding engineers- Karpathy. The kind of engineer they were having a hard time convincing to leave higher paying jobs to get. So it’s not too surprising in retrospect that rather than putting Musk in charge, the board moved Altman into the role of President.
And Musk leaving had another effect. According to Semafor, he was supposed to keep contributing money to OpenAI, but he didn’t. That was about $1 billion that the company was expecting to get that it no longer had. At a time when it was scraping to make the funding meet its ambitions.
And right then Google Brain released its “transformer” model. The T in GPT by the way. It was a huge leap forward for AI models, but required a lot more data to train, meaning a lot more computer power, meaning a lot more cost. A cost Google, which ran its own cloud services, could afford to pay. OpenAI, which paid Google for cloud services, could not. If it didn’t want to see Google seriously outdistance it, OpenAI needed to do something.
It started by releasing a new charter in April 2018. It still “committed to avoid enabling uses of AI or AGI that harm humanity or unduly concentrate power,” and said that its “primary fiduciary duty is to humanity.” But now it also said, “We anticipate needing to marshal substantial resources to fulfill our mission, but will always diligently act to minimize conflicts of interest among our employees and stakeholders that could compromise broad benefit.”
A more public hint that things were changing was the announcement of GPT-2 on February 14, 2019. OpenAI’s Valentine’s Day gift was to not open source this release as it had for its previous releases. GPT-2 could take prompts and complete them. So give it a headline it could write the rest of the article. OpenAI justified the less open release by citing the risk that the tool could be used maliciously. Though a public interface was released. And eventually the full code was released in November.
But then it took a Serious step. On March 11, 2019 OpenAI pulled a move from Mozilla’s playbook. Mozilla had operated for decades as a non-profit that fully owned a for-profit subsidiary. This allowed it to make money on Firefox and attract talent and pay for development.
OpenAI was going to do a similar thing. OpenAI Incorporated, the non-profit, would form OpenAI Limited Partnership, a for-profit company, wholly controlled by OpenAI Inc. But OpenAI LP would be profit capped. Investors would receive up to 100 times their investment, and excess profits would go to the nonprofit OpenAI Inc. To assuage concerns about the move, Altman, the CEO of the new for profit company, took no equity in it.
So they had their solution. Sell non-controlling shares in the for profit company. Except nobody was buying. It was profit capped and the CEO didn’t even want a single share? Not for me.
Well, unless you’re Satya Nadella. In September 2019 OpenAI got its first big investment bite. Microsoft agreed to invest $1 billion, a nice replacement for the lost Musk donations. Not only would it invest but it had even better cloud resources than Google, so it would make its vast Azure infrastructure available. Money to pay talent AND bargain cloud computing. And Microsoft gets to become a bigger player in AI.
Microsoft and OpenAI built a supercomputer to handle the massive amount of data needed to train Large Language Models.
OpenAI was back in the race.
In January 2021, OpenAI released DALL-E a multimodal model that could create images based on a text description.
In August 2021 it launched Codex, which translates natural language to code, and powers Microsoft’s GitHub CoPilot feature.
And in November 2022 DALL-E 2 captured imaginations with much better performance and spawned multiple imitators like Craiyon and Midjourney.
But of the course the big leap also came with the launch of ChatGPT that same month. For OpenAI it was just the latest public demonstration of what its Large Language Model could do. Nobody got that excited when it launched DialoGPT in 2019. Why would this time be any different? Well. It was. But for whatever reason it captured the public imagination. Suddenly OpenAI wasn’t just staying in the race. It was leading it.
Google issued a code red. Microsoft and Google got into an AI announcement competition.
Altman was triumphant.
Musk was — not.
In December 2022, Twitter- now owned by Musk, pulled OpenAI’s access to Twitter data. Musk began tweeting criticisms of OpenAI.
On February 15, 2023, he sang his old 2015-era tune again to attendees at the World Government Summit in Dubai, United Arab Emirates, “One of the biggest risks to the future of civilization is AI.”
On February 27, 2023, The Information wrote that Musk was recruiting engineers and scientists to form a lab to compete with OpenAI.
And on Wednesday March 29 signed an open letter put out by a think tank he funded calling for all companies to pause their research into the next version of AI for 6 months in order to create a safety scheme.
Oh and the week before that, Shivon Zillis, the mother of Musk’s twins, resigned from the OpenAI board.
Altman on the other hand, talking on Lex Friedman’s podcast on March 25, 2023 described Musk as one of his heroes and said, “I believe he is, understandably so, really stressed about AGI safety.” [[find podcast sound for this?]]
So there you have it. OpenAI is a nonprofit AND a for profit company. It was co-founded by Elon Musk, but that’s not nearly the whole story. And whether it has remained true to the values from its founding or whether it engenders the fears it was formed to address. I’ll leave that up to you.
I just hope you Know a Little More about OpenAI.