Sure. Someone who has their data in 23andme was someplace where something horrible happened. Law enforcement has no leads, so they process the DNA, and find no matches. They subpoena 23andme (or just look at the leaked data, who knows), and that person is now a person of interest. If they don't know they should have a lawyer on their side with them when being questioned, they might talk themselves into prison.
I know it's not exactly the point, but the answer is you always have a lawyer if you're being questioned by the police, especially if you're not the one who called them. This is in no way specific to any DNA related situation. If any law enforcement shows up and has questions for you, you say nothing until you have a lawyer with you.
Doesn't matter. If they're talking to you say you need your lawyer present -- that will end the interrogation, and you might never need a lawyer. But if you don't say that you might end up convicted.
I'm not saying it's fair or right. As with most things in life, the cost/benefit makes it likely you'll have to take some risks, but people need to understand that literally any interaction with any law enforcement without a lawyer present is a risk of things going horrifically badly. This is especially true if they show up and you do not know why. That is a massive set of alarm bells that should be going off.
If you can’t afford a lawyer just say nothing. Nothing you say will help you; anything you say can be used against you. If you say nothing you force them to use other evidence which they likely don’t have.
Police (in the U.S. anyways) are allowed to use ruses almost without limits, so you might not even know you're being interrogated, and your answers can be used against you (and only against you -- never to exculpate you!) in a court of law.
IMO the hearsay rule is way too biased in favor of the government.
Worse! Say you know it's police asking you questions, and the questions are all very harmless, so you answer them, but then you start to get an inkling that you are a suspect, that they like you for some crime, so now you shut up and/or lawyer up, but guess what: your disposition's change from cooperative to non-cooperative can and will be used against you in a court of law!! What, you say? Yes, the SCOTUS in the 2010s (see Salinas vs. Texas, from.. 2014 IIRC) greatly reduced the 5th Amendment's protections in this way. If you're talking then clam up, the fact that you clammed up -and at what particular question- can be used against you. And if you never said a word to them, that too can be used against you! The only thing that works is to tell them very early on that you will only talk to them with your lawyer present [and since you don't have a lawyer yet you might never talk to them] then follow through.
If the police are talking to you it's either a) they think your testimony can convict someone else, and/or b) they like you for some crime and want to give you ample opportunity to convict yourself of it even if you didn't commit it (they may not know that, but they may like you so much for it that their bias is too strong to see that you're innocent) and even if you have no idea what the heck they're talking about (because they don't even have to tell you).
See all of professor James Duane's videos on this topic, starting with Don't Talk to Police (this one is pre-Salinas), and then the later post-Salinas reprise(s) of it.
I mean, assume any time you're talking with the police, you're being interrogated. If you called them yourself for something, then you've decided to take that risk, but that's also a situation where you're less likely to be in trouble yourself, though there's still a risk.
The company seems to be in rough condition. Say they go bankrupt and an ad-tech data broker buys their assets. Now DraftKings can laser focus their ads to folks genetically predisposed to addiction.
You’d have to work out what variants predispose for that which is no easy task. And once you did that you don’t even really need individual dna data. You might find say a swedish population tends to have the variant and you just target swedes in general.
> Say they go bankrupt and an ad-tech data broker buys their assets. Now DraftKings can laser focus their ads to folks genetically predisposed to addiction.
Can an insurance company deny claim based on your DNA? They deny claims for pre-existing condition that you hid from the, which would be the wrong thing to do on your part. They cannot deny claim based on pre existing disposition. Practically everyone is predisposed for getting cancer by merely being human, you might even have cancerous cells in your body right now, that you body will destroy in a couple of minutes.
> Can even one single person here articulate their specific fears of using 23andme
Sure. I find out my competitor for the top role has a degenerative disease, e.g. Parkinson’s. It’s not relevant for many years. But I use it, subtly, to shape opinion.
More pointedly: we are in an era of mass disinformation. The simple fact that somebody used 23andMe makes any lie about it somewhat credible.
Well for one, English is not my first language, so take it with a grain of salt. Also: when I use genocide as as a verb, it means there is a concerted effort to kill of a group of humans, that you happen to be a part of.
I know we live in an individualist society, but when you are murdered as part of a genocide that has nothing to do with you as an individual, which is a significant part of the horror of the whole thing. You are then murdered because someone thought you belonged to a group that should be wiped of the face of the earth. Whether you really belonged to that group, whether you share the ideology of the group or of those doing the genocide, wheter you are a really nice individual has nothing to do with it.
If I've committed a crime and gotten away with it for several decades, I don't need a relative to NARC on me by giving 23 and me and the feds a DNA sample, thank you very much.
It's bad enough they took my fingerprints when I worked for a school district.
Besides the obvious examples of gathering a nice database to use for genocidal purposes... (sure lot's of idiots like to say that's overblown or not really a worry, while being alive in a world where there are on-going 'ethnic cleansing' campaigns).
There's also things like - the terms of service include the boilerplate "these terms are subject to change at any time", and I don't want those terms to suddenly change to "we will provide your PII to all insurance companies proactively in exchange for a kickback every time they are able to use it to reject a claim".
I already get hassled by the law somewhat frequently because my house used to be the residence of a criminal (2 owners ago it was used as a rental and that owner evicted said criminal). I don't want to add getting hassled by a bunch of people who came in below the max IQ requirements over someone I've never met because they're from "that side" of the family.
To be fair, Brave seems to have genuinely tried, but, seems to be failing to reach sustainablility judging by their various pivots. They've tried to walk the line of leaning toward the user while middlemanning revenue ideas to said user. BAT and crypto, VPN, messaging sales, now search+ads. The market seems to be saying no thanks from what I can see from the outside.
Orion, in straight up exchange for cash via donation and Kagi subsidization, offers to be entirely the user's agent.
I use a number of different search engines for different purposes. Yandex offers sites duckduckgo doesn't. Duckduckgo offers sites Google doesn't Google is good overall and at image searching.
You seem like someone who has used a number of search engines. How does it treat sites Google blacklists like nsfw sites or porn sites or torrent sites? Does typing in movie/book/game torrent bring up listings the way Google previously did? How does it treat older non-mobile friendly sites? Is the image search comparable to Google?
I'm on the same boat. Used to g! most of my queries with ddg. Since switching to kagi, I've fallen back to Google for only a few queries, and I think only one or two had better results
That doesn't mean it was an accident, it could just mean that they realized it's a problem. Especially as this issue was opened after this HN submission got posted.
Issue opened 3 hours ago, thread posted 2 hours ago. I look at Brave with skepticism as one should with a company that flirts with crypto, but in giving them a fair shake, it's important to keep the facts straight. That said, in keeping the facts straight, an issue opened a week ago citing this as a problem was closed in favor of the one opened by the Brave VP.
It’s a publicly viewable GitHub issue that people can still navigate to or comment on, it’s not like it was scrubbed. It’s surely a PR move to show they’re on top of it but not anything to read into in my opinion.
Most voiced complaints are (as neutral as possible):
- past CEO personal belief on LGTB rights
- If opted in, blocking ads while showing their own ads as system notification (ads are not being replaced in page)
- hijacking typed-in domains in the address bar by adding their affiliate code on select domains related to crypto (reverted after backslash)
- Tips (using their crypto coin BAT) to Content Creators not using Brave being held indefinitely until withdrawn by Creator (behavior changed after backslash)
and now:
- VPN services being installed automatically if browser has admin rights on Windows.
Edit: I'm just listing what I have been hearing the most on HN.
First is irrelevant to whether the product is good, I don't care about purity tests.
The rest of the list, minus the end, is them trying to make money without selling data. That actually makes me more confident about their belief in their stated values. Last one is an "accidental" oversight, probably to boost install numbers. They decided to change it without backlash first. Again, good move.
I am biased because this is my research area, but I have to respectfully disagree. Actor models are awful, and the only reason it's not obvious is because everything else is even more awful.
But if you look at e.g., the recent work on task-based models, you'll see that you can have literally sequential programs that parallelize automatically. No message passing, no synchronization, no data races, no deadlocks. Read your programs as if they're sequential, and you immediately understand their semantics. Some of these systems are able to scale to thousands of nodes.
An interesting example of this is cuNumeric, which allows you to take sequential Python programs that use NumPy, and by changing one line (the import statement), run automatically on clusters of GPUs. It is 100% pure awesomeness.
If you really want to dig into it you can read up on the tutorials and/or papers from the Legion project: https://legion.stanford.edu/
But briefly, these task-based programs preserve sequential semantics. That means (whatever the system actually does when running your program), as long as you follow the rules, the parallelism should be invisible to the execution of the program.
While not for clusters AFAICT, Rayon (Rust lib) auto parallizes loops. It is used in Programming Rust by O'Reilly in a great example. Forgive the RESF response please.
It is a pity Concurrent ML didn't take off. F# has a great library called Hopac that implements it, but it is 50 times less popular than its closest competitor Rx.
Also seconding that other post. Actor models and async concurrency are only useful if you need to send messages between machines, but otherwise you want to use synchronous concurrency as it is easier to deal with.
I'm not a geneticist so I could be totally off the mark; but, to my understanding, the painful part is that is a big disclosure right there