Zoom executive charged with disrupting meetings commemorating Tiananmen Square

Yeah - this is the real risk with TikTok, not the data (or who hosts the data).

Moving the cloud hosted data to Oracle doesn't solve the underlying risk.

The real risk is CCP manipulating what's surfaced via the algorithm to sway public perception or to censor things they dislike. They've already done this explicitly with the protests in HK and Tiananmen Square, they'll do it in other ways that suit them.

The worst part is with TikTok the manipulation is more subtle, harder to see - are you seeing something because it's popular? or because the CCP is interested in you seeing it?

Are you not seeing something because it's not popular? or because the CCP has told Bytedance to kill it?

Moving the data to Oracle misses the point. The data is not important in this case, but the ability to control what the public sees is. Letting the CCP have control over that lever when they abuse it elsewhere: that's the risk.

This can be an issue anywhere, but the CCP has a terrible track record and no interest in a free press or free speech. What they're doing to the Uyghurs and what they've done to HK and how they treat people talking about Tiananmen Square. They don't take responsibility for any of it and they do what they can to suppress all of it.

It's a very machiavellian view and their soft influence on the US and western companies tied economically to China to self-censor or otherwise defer to the CCP's interests is alarming.

That recent article about Tim Cook killing an Apple TV show written by a couple ex-gawker editors doesn't bother me at all. Gawker was awful and we're better without it, shows glorifying it, or supporting the editors that worked there. This is Tim Cook deciding for himself he doesn't want to support those people, it's a principled position.

The other bit in that article about an Apple rep saying the two things Apple TV would never do are 'nudity and China' was a lot worse. The South Park 'band in china' episode was right. This is Tim Cook knowing what's expected of him to be able to operate in China - it's not a principled position, his hand is forced.

I’m not pointing to page 47 I’m pointing out that there is 47 pages of the government describing the evidence it has. That’s typical for an indictment. The government doesn’t “in general” refuse to show the evidence. That’s just not how it works.

Grand juries aren’t “known to be tools of the prosecution.” That’s just something defense lawyers and public interest people say. A grand jury is just a group of ordinary people tasked with deciding whether there is enough evidence to even bring a case. The secrecy is there to protect the defendant as much as it’s there for the government. If the prosecutor’s case is weak, then the allegations don’t even become public. Grand juries do vote to indict at a very high rate. That’s because prosecutors only bring cases they know they can win (and indeed they have an ethical obligation not to bring cases they don’t think they can win).

There is a lot of misinformation about how the justice system works. I have seen it from the inside, working for a judge. These cases are buttoned up top to bottom. They’ll arrest a guy for robbing a store. They have him on security camera. They have tracking information from his cell phone. They have his text messages with the guy who he fenced the goods to. They searched his car and found stolen items.

Wrongful convictions are rare. The legal clinics that help wrongfully convicted people actually spend tremendous amounts of time screening cases to find ones that are meritorious. And a lot of those are from the era before DNA testing came into routine use. See: https://dc.law.utah.edu/cgi/viewcontent.cgi?article=1130&con...

> Based on a careful review of the available empirical literature, it is possible to assemble the component parts of a wrongful conviction rate calculation by looking at error rates at trial, the ratio of wrongful convictions obtained through trials versus plea bargains, and the percentage of cases resolved through pleas. Combining empirically based estimates for each of these three factors, a reasonable (and possibly overstated) calculation of the wrongful conviction rate appears, tentatively, to be somewhere in the range of 0.016%–0.062%

I was inclined to agree with you, but then I remembered the case of Saudi Arabia's guys inside Twitter: https://www.wired.com/story/twitter-insiders-saudi-arabia-sp...

Ah yeah, one feature Zoom had was the multi-person grid view which Meet has nowadays but before, Meet would only show primarily the presented content or one speaker at a time