Lots of gambling involves some skill. There's skill in playing poker. Sure, it involves some luck of the draw, but it also involves playing the odds and manipulating opponents. Lots of games have some skill, but they're still generally banned as illegal gambling when money is involved.

Fantasy sports are still legal, just not these sites. I can see that playing over the course of a full season is very much a game of skill. Drafting well for the season, making trades, and dropping and adding players all involve skill. Over the course of a season, the performance of a player is generally fairly predictable. It's far from perfect, but it's predictable. And that makes it more a game of skill than a game of chance.

Daily fantasy sports are more a game of chance, with much more limited skill. The performance of players is highly variable from game to game and week to week. A lousy hitter in baseball can have a great week and put up big numbers. A really good QB in football can throw four interceptions in a game and have an awful week. There's a lot of luck involved. As they say in the NFL, any given Sunday...

These sites focus on daily or weekly leagues rather than full season leagues. For that short amount of time, there's a huge amount of chance involved. Over the course of a season, the random luck usually averages out and it becomes much more a game of skill.

There's also the issue that the statistics about draft decisions aren't generally available to the public. But an employee of a daily fantasy sports site can get access to that information. Those statistics are likely to be similar with other daily fantasy sports sites. The employee has insider information that they can use to gain an edge over competitors on other sites.

I don't think jargon in scientific writing is always a bad thing. It's important to be precise. When I conduct an experiment, I need to be precise about my procedure. If I'm precise, readers can identify caveats in my methods that would affect the outcome of my work. It also means someone else can duplicate my experiment. It's important for scientific experiments to be repeatable. It's also important to be precise about conclusions. I work in meteorology, a field that's next of kin to climatology. A lot of research about global warming is misunderstood or exaggerated when the general public hears about it. If I write a paper about global warming, I need to be precise in my conclusions so I don't contribute to this problem. If jargon helps me be precise, it's a good thing. If jargon exaggerates the importance of my work or obfuscates its meaning, it's a problem.

I think scientific writing is difficult to understand because so much is written in the passive voice. It was once acceptable to use first person pronouns, so writing has more active voice. First person pronouns fell out of favor in scientific writing about a century ago. Writers should be free to use first person pronouns if they make the writing easier to understand.

I also don't like how so many papers try to exaggerate their importance in the introduction. The first paragraph describes a very important problem while the rest of the paper only addresses a tiny part of that problem. It's done to persuade editors that a paper is of interest to more of their journal's audience. But it also contributes to misunderstanding.

Actually, there are many examples of FCC enforcement against transmitters on certain 5 GHz bands interfering with terminal doppler weather radars: https://www.fcc.gov/encyclopedia/weather-radar-interference-enforcement. This is actually a real issue.

No, it isn't especially frequent, but it does take place. There are two reasons it isn't more frequent:
1) Most transmitters aren't located in buildings that are high enough to be in the line of sight of airport weather radars. Generally the enforcement actions are against operators of transmitters in or atop tall buildings. Your transmitter a couple of floors above ground is highly unlikely to ever interfere with a radar. And if the radar beam was refracted severely enough for this to occur, there would almost certainly be a lot more interference from ground clutter than your wi-fi transmitter. This is more of an issue in tall buildings. The actual buildings are normally pretty unlikely to cause problems because they are stationary point targets that get filtered as ground clutter. Wi-fi, however, would probably contaminate an entire radial, similar to a sun spike.
2) Transmitters operating on either of the 5.25-5.35 GHz and 5.47-5.725 GHz bands are required to use dynamic frequency selection. They are supposed to listen for the signals transmitted by weather radars and, upon detection, switch to a frequency that does not cause interference.

The transmitters are locked into a frequency set. However, this is shared with the frequencies used by terminal doppler weather radars (TDWRs). Wi-fi equipment is supposed to detect when a radar is operating on a particular frequency and, upon detection, switch to a frequency that's not in use by a radar. Normally this isn't an issue if you're operating a transmitter near the ground. However, if it's located higher up in or atop a building that happens to be within the line of sight of a radar, it can cause interference. Locking wi-fi transmitters out of any frequencies used by TDWRs would greatly reduce the available spectrum and it isn't necessary. Most wi-fi in the 5 GHz band won't interfere with TDWRs because it either a) detects the radar and switches frequencies or b) is located such that it is out of the line of sight of the radar and can't interfere.

These radars are used to detect hazardous weather, not aircraft. There are about 45 terminal doppler weather radars installed in the US, which share the 5 GHz band with wi-fi. They are primarily used to detect wind shear and microbursts, which are dangerous to aircraft.

Terminal Doppler weather radars (TDWRs) are installed at about 40 major airports. They operate in the 5 GHz range and are used to detect things like wind shear and microbursts, which are dangerous to aircraft. The higher resolution than the WSR-88D (Nexrad) radars probably makes it easier to detect these features. Also, if the nearest WSR-88D is a significant distance away from the, the beam will be significantly above the ground over the airport. Interference from wi-fi isn't an issue for the WSR-88D radars because they operate around 2.7-3.0 GHz.

As I understand it, the 5 GHz band used for wi-fi systems is shared with TDWRs. This allows wi-fi to operate on TDWR frequencies in areas where there isn't a TDWR using that frequency. Wi-fi equipment is required to detect when a TDWR is operating and, upon detection, switch to another frequency to avoid interference. There have been quite a few enforcement actions by the FCC for wi-fi interfering with TDWR operation: https://www.fcc.gov/encyclopedia/weather-radar-interference-enforcement.

The FCC actually has investigated many complaints of transmissions from wireless devices interfering with terminal doppler weather radars. Here's a list of enforcement actions by the FCC: https://www.fcc.gov/encyclopedia/weather-radar-interference-enforcement. So, yes, there have been investigations of the source of interference and penalties for doing so.

I read the Nature article and, frankly, it's very troubling.

My DNA is unique to me, so there really is no anonymity to begin with. It's just not linked to my name, address, SSN, etc. This information would be a gold mine to insurance companies, doctors, pharmaceutical companies, and perhaps potential employers. There's so much potential for abuse and so little security involved. Furthermore, it doesn't seem like relatives have the option to opt out from being referenced here, which means my privacy might be at risk without my consent. Sure, it won't include my own DNA, but it's still a huge risk.

I understand the scientific value of such a data set, but there's just way too much risk involved.

There are versions for Windows, OS X, and Linux. Amazon supports the Snowball Client for Ubuntu 12+ and RHEL 6+, but no doubt it can run on other systems. https://aws.amazon.com/importexport/tools/

Also, as per this link, they're working on chain-of-custody tracking using GPS. Amazon has already considered the possibility of theft and it doesn't seem likely to be an issue.

Well, for one, it provides evidence of human activity in that region. It's not entirely clear when humans migrated into that area, so it provides information about human migration. It also provides some information about the climate of the area because the mammoth was limited to surviving in certain habitats. Yes, it's obvious that humans hunted the woolly mammoth, but it's new for that region.

I do think the woolly mammoth will eventually be created again by scientists. But I don't think it's quite as close as you'd think.

I think it should be tested on perhaps a simpler type of animal and one where there's less of a gap between generations. The passenger pigeon is a good candidate for this, in part because it might be easy to get funded and the timescale for animals to reproduce is much quicker. Of the 32 or so passenger pigeons that have had their genomes sequenced, there aren't a lot of differences. That means it might not be necessary to create huge numbers of them with large genetic diversity for them to survive. This also requires sequencing the genome of a similar animal, in this case the band-tailed pigeon. Then it's necessary to determine what genes made the passenger pigeon what it was, rather than another type of pigeon, and creating a chimera by inserting passenger pigeon genes into the band-tailed pigeon genetic sequence. Even if a passenger pigeon can be created through this process, there's still the challenge of teaching the passenger pigeon to behave like one. Simply creating an animal with similar genetic code to a passenger pigeon doesn't make it a passenger pigeon.

The same process could be done with the woolly mammoth by creating a chimera with the Asian elephant. Sequencing the genome is probably the simpler part, and that's underway. But it would be good to see the process done successfully with another animal like the passenger pigeon, which should be faster than the woolly mammoth. I'd guess it's probably 50 years away, but I think it'll eventually happen.

Sadly, I don't think another famous extinct animal, the dodo, will be so fortunate. I'm not sure there will be as good of a close relative since it evolved in isolation for a long time on Mauritius. It's also old enough and in a climate warm enough that genetic preservation is much less likely.

I agree it's a defective product and I started thinking about this after I made my post. My idea was to try to force manufacturers to honor their warranty, which is supposed to protect against defects. Also, because Samsung knew about this vulnerability in August 2014 and continued to sell the Galaxy S4, they're knowingly selling a defective product. That ought to be more serious than simply selling a defective product.

I don't understand why phone manufacturers and carriers don't get sued for things like this. Carriers have typically required two year contracts for phone subsidies, and normally it's possible to buy a phone two years old and get it free. At least that's how it is in the US. That means you can buy a phone that's as much as three years old and have a reasonable expectation to use it for two years because that's the contract with your carrier. That means manufacturers and carriers should provide support for a minimum of five years. That means a phone released in October 2015 should have support until October 2020. I think a customer has a reasonable expectation of this. If nothing else, that should be grounds for a lawsuit against manufacturers and carriers. There's also the issue of delays in fixing vulnerabilities both with the manufacturers and then the carriers. Again, I think there's a reasonable expectation for security updates in a timely manner. Also, when phones ship with locked bootloaders and customers can't choose to unlock them, it makes it very difficult to install a patched version of the OS. This also voids the warranty if you're able to do it. Customers are screwed no matter what they do in this situation, which is why carriers and manufacturers should be sued in the absence of specific laws to protect customers.

I can't help but wonder if the decision to not provide software updates to older phones is partly because people don't see a huge difference between models and this is one way to push people to buy newer and more expensive phones. I can't say it for certain, but it wouldn't surprise me if that's part of the decision process.

I'm actually less bothered by researchers getting access to data to answer specific questions than I am about the potential for data breaches.

Almost invariably, giving researchers access to administrative data means creating a copy of that data. While it may not be easy to obtain illegally in its original storage location, creating a copy of the data increases the potential for a breach by the mere fact that another copy exists. It's also easy to see how researchers might not safeguard the data to the extent that its original owner does.

I work at a university and do scientific research. Some of it involves running massive jobs on supercomputers. Access to those systems is tightly controlled, much more so than most other computing resources at the university, because it's been a frequent target of intruders. Within the past couple of weeks, I've gotten an email about phishing attacks specifically directed at getting passwords to those systems. Clearly the data on those machines is of value to criminals, otherwise they wouldn't be trying to obtain it. There are rules about how long data can be stored in what's essentially scratch space as well as prohibiting the storage of data protected by laws like HIPAA. I know for a fact that the rules about the use of scratch space as temporary storage are frequently broken by many users. I have no reason to trust that rules about what types of data can be stored on those systems are followed more frequently. Even if researchers aren't malicious, I generally believe that they are careless with data.

When these data are given to researchers, it's an inviting target for criminals. These are data breaches waiting to happen. Just say no to sharing administrative data with researchers.