They're not 100% perfect, but they're much much much more accurate than they need to be for this sort of analysis.
For example; I picked out a sample mostly at random from an internal dataset. In this shallow sample of 4 million reads, over 90% of reads have all 36 bases with quality > 30. (Quality is in Phred score: q30 = the sequencer estimates a 99.9% chance that this base is read correctly). Even if 3-4 bases are incorrect, it's still usually possible to map the read to the genome with high confidence (even if the aligner discards the base-quality information!). This combines with the fact that you'll have multiple reads mapping to each target, so even if several reads are mapped to the wrong target, the right calls will be made. Finally, I'm sure there's some heuristic for "we found 98% of the expected targets for E. xamplis, and trace amounts of T. atoeba, we're pretty sure it's the former."
The false positive rate is obviously not zero - it's irresponsible of the article to claim that. But it's much lower than other competing methods, and I wouldn't be at all surprised if in their testing they found 0 false positives.
In slightly more technical terms, they've designed a system that selectively targets & amplifies ~2 million DNA sites; chosen from the genomes of all known infectious viruses. The scientists basically apply this assay to the infected cells (I'm assuming they take a blood sample or something), leaving them with DNA that matches those targets. Then, they run those DNA fragments through a sequencer, and see what they got. From there, they can deduce which virus was present in the original sample.
Only if "forget everything about me" includes the fact that it's been asked to forget you. I can see cases where people can say "Write down in your book never to store information about me" and have that be useful. Yes, the datapoint that you requested to be forgotten is not of no value, but it's likely better than them remembering what kind of weird porn you're into.
Alternately, it's not a hassle to keep up this loop if John Doe has a passive signal to the robot to keep it from verbally asking him each time - say, a little card on the table, or a DoNotTrack bit set in his browser.
And even if they aren't malicious enough to hack you, it's not a hard decision for google to say "Oops, we no longer support URLs longer than 200 characters," or just drop everything after the anchor tag, so they aren't stuck storing some million cat gifs in their database.
"You say that a zygote isn't a child. But I ask: if we kill all the zygotes will there be any children in a few months?"
Of course there will, unless you're planning on slaughtering all existing children. Children generally remain children for about 12 years, which is much longer than a few months.
And yes, if we destroy every single existing acorn/minnow/egg/child (wow), all of those things will be around in the 30-200 year time frame you mention. Oak trees produce acorns every year, as do chickens with eggs, and people (can) with children. If you killed every single zygote that was conceived for the next twenty years, we'd still have people 200 years from now.
Even your straw men don't make the point you think they do.
Huh, I wonder why a 9mm penetrating an empty soda can wouldn't transfer much momentum to it?? Oh, it comes out the other side? Funny that your expert understanding of physics missed that.
Exactly - 90+% of programmers are only writing straight-forward business programs. You don't *need* to understand the intricacies of, say, finite fields unless you're one of the hundred people in the world researching new crypto.
Even the math that is very important for many programmers (say, relative time complexity of algorithms, or set unions/intersections) is pretty dissimilar to what we Americans teach as "math" to our kids, and can largely be intuited.
I say this with a minor in math - it's really not that important for coding.
It's kinda screwed up how you're blaming the women for "facilitating cheating" and causing dudes to lose their community standing and commit suicide. As opposed to blaming the cheating men - you know, the ones with an actual obligation to their wife and kids.
Like I hear what you're saying - it's not good to get involved and be "the other woman." But their fault is minuscule next to the cheater's.
My worry with this thinking (determining who "really needs it") is that no test is perfect, and it needs to be really really good before it's okay to use it as diagnosis (and refuse care to those who don't pass the test).
Even if it's 95% effective, I'd hate to be the one-in-twenty person who is wrongly told "Nope, you're faking it, no brain medicine for you."
Again, that's energy consumption, not electricity consumption. The claim was " it takes more electricity (yes, actual electricity, not "energy" in general) "
That actually sounds pretty sensible. It seems like much of her frustration is from people blindly running static analysis tools on their code, finding false-positive vulns, and wasting Oracle's time and making it more difficult to identify legitimate security vulnerabilities.
Much more reasonable than the summary made it out to be, thanks.
A much bigger problem is how your actions split the vote up. Say you run a "referendum ticket" with yourself and your left-wing Real President.
If the voter has two tickets: "you + lefty" vs. "righty", then you conflate left-wing vs whatever you're referending on - you can't know who supported the referendum, and who simply didn't want righty. So you say, "I know, I'll run another ticket with me + right-wing candidate," at which point you're now splitting the right-wing ticket (because votes don't magically sum), and the left-wing candidate will likely win regardless of support for your referendum.
So you introduce a 4th ticket, with lefty alone, so any voter can vote for left, right, left+you, or right+you. That sounds fair, right? Unless there's any correlation between party & support for your referendum. Let's say the left party prefers the referendum 80-20, and right party support for the referendum is about 50-50, and overall voter preference is 60-40 for right party. The winner of the ticket would be you+lefty (with 32% of the vote), even though 60% of people voted for righty.
TL;DR - first-past-the-pole only barely works with two candidates. You can't co-opt it to do a referendum.
Your link is *also* not a law, but instead, a story about one football player punching another player in the face. I don't think it's an executive order either, but it's been a little while since my highschool civics class.
Slashdot Mirror
Guess what - anyone who sees your house on the street knows where you live.
Next up: Worrying about the existence of neighbors and pizza delivery men.
They're not 100% perfect, but they're much much much more accurate than they need to be for this sort of analysis.
For example; I picked out a sample mostly at random from an internal dataset. In this shallow sample of 4 million reads, over 90% of reads have all 36 bases with quality > 30. (Quality is in Phred score: q30 = the sequencer estimates a 99.9% chance that this base is read correctly). Even if 3-4 bases are incorrect, it's still usually possible to map the read to the genome with high confidence (even if the aligner discards the base-quality information!). This combines with the fact that you'll have multiple reads mapping to each target, so even if several reads are mapped to the wrong target, the right calls will be made. Finally, I'm sure there's some heuristic for "we found 98% of the expected targets for E. xamplis, and trace amounts of T. atoeba, we're pretty sure it's the former."
The false positive rate is obviously not zero - it's irresponsible of the article to claim that. But it's much lower than other competing methods, and I wouldn't be at all surprised if in their testing they found 0 false positives.
In slightly more technical terms, they've designed a system that selectively targets & amplifies ~2 million DNA sites; chosen from the genomes of all known infectious viruses. The scientists basically apply this assay to the infected cells (I'm assuming they take a blood sample or something), leaving them with DNA that matches those targets. Then, they run those DNA fragments through a sequencer, and see what they got. From there, they can deduce which virus was present in the original sample.
Only if "forget everything about me" includes the fact that it's been asked to forget you. I can see cases where people can say "Write down in your book never to store information about me" and have that be useful. Yes, the datapoint that you requested to be forgotten is not of no value, but it's likely better than them remembering what kind of weird porn you're into.
Alternately, it's not a hassle to keep up this loop if John Doe has a passive signal to the robot to keep it from verbally asking him each time - say, a little card on the table, or a DoNotTrack bit set in his browser.
Wipe it. Flash a new ROM; don't install any other app stores, don't download sketchy apps.
If you have malware, that's cause you (or someone with access to your phone) installed it. Don't do that.
And even if they aren't malicious enough to hack you, it's not a hard decision for google to say "Oops, we no longer support URLs longer than 200 characters," or just drop everything after the anchor tag, so they aren't stuck storing some million cat gifs in their database.
Correct, I was not familiar with that piece of cloth. But I guess you got a boner for it, which makes your religious motivations irrelevant.
Addressing someone with "O evolutionist!" is a pretty big giveaway of what you believe and where your motivations come from.
"You say that a zygote isn't a child. But I ask: if we kill all the zygotes will there be any children in a few months?"
Of course there will, unless you're planning on slaughtering all existing children. Children generally remain children for about 12 years, which is much longer than a few months.
And yes, if we destroy every single existing acorn/minnow/egg/child (wow), all of those things will be around in the 30-200 year time frame you mention. Oak trees produce acorns every year, as do chickens with eggs, and people (can) with children. If you killed every single zygote that was conceived for the next twenty years, we'd still have people 200 years from now.
Even your straw men don't make the point you think they do.
Huh, I wonder why a 9mm penetrating an empty soda can wouldn't transfer much momentum to it?? Oh, it comes out the other side? Funny that your expert understanding of physics missed that.
No.
And if you don't know anyone that'd be happy to get $6k, you need to step out of your tech bubble and meet somebody at the median level of income.
"Only $6000"
Maybe this is because I'm not an Apple/Google/Intel employee, but if I got $6k handed to me, I'd be psyched.
Exactly - 90+% of programmers are only writing straight-forward business programs. You don't *need* to understand the intricacies of, say, finite fields unless you're one of the hundred people in the world researching new crypto.
Even the math that is very important for many programmers (say, relative time complexity of algorithms, or set unions/intersections) is pretty dissimilar to what we Americans teach as "math" to our kids, and can largely be intuited.
I say this with a minor in math - it's really not that important for coding.
I recommend not eating the batteries, even if your diet is a little short on Iron.
I'm pretty sure Germany's had laws about denial of the holocaust since well before modern internet culture was around.
Watch what the ACs are (or will be) saying in this thread - their messaging is usually very consistent.
It's kinda screwed up how you're blaming the women for "facilitating cheating" and causing dudes to lose their community standing and commit suicide. As opposed to blaming the cheating men - you know, the ones with an actual obligation to their wife and kids.
Like I hear what you're saying - it's not good to get involved and be "the other woman." But their fault is minuscule next to the cheater's.
One of the bonuses of saving the info on the amiibo is that you can bring it to your friend's and use it there.
I dunno if it's the summary or the article that's trash, but wow. Terrible.
My worry with this thinking (determining who "really needs it") is that no test is perfect, and it needs to be really really good before it's okay to use it as diagnosis (and refuse care to those who don't pass the test).
Even if it's 95% effective, I'd hate to be the one-in-twenty person who is wrongly told "Nope, you're faking it, no brain medicine for you."
Again, that's energy consumption, not electricity consumption. The claim was " it takes more electricity (yes, actual electricity, not "energy" in general) "
That actually sounds pretty sensible. It seems like much of her frustration is from people blindly running static analysis tools on their code, finding false-positive vulns, and wasting Oracle's time and making it more difficult to identify legitimate security vulnerabilities.
Much more reasonable than the summary made it out to be, thanks.
A much bigger problem is how your actions split the vote up. Say you run a "referendum ticket" with yourself and your left-wing Real President.
If the voter has two tickets: "you + lefty" vs. "righty", then you conflate left-wing vs whatever you're referending on - you can't know who supported the referendum, and who simply didn't want righty. So you say, "I know, I'll run another ticket with me + right-wing candidate," at which point you're now splitting the right-wing ticket (because votes don't magically sum), and the left-wing candidate will likely win regardless of support for your referendum.
So you introduce a 4th ticket, with lefty alone, so any voter can vote for left, right, left+you, or right+you. That sounds fair, right? Unless there's any correlation between party & support for your referendum. Let's say the left party prefers the referendum 80-20, and right party support for the referendum is about 50-50, and overall voter preference is 60-40 for right party. The winner of the ticket would be you+lefty (with 32% of the vote), even though 60% of people voted for righty.
TL;DR - first-past-the-pole only barely works with two candidates. You can't co-opt it to do a referendum.
Which, you know, aren't laws.
Your link is *also* not a law, but instead, a story about one football player punching another player in the face. I don't think it's an executive order either, but it's been a little while since my highschool civics class.