上海419论坛最新

Some people might think that online privacy is a, well, private matter. If you don’t want your information getting out online, don’t put it on social media. Simple, right?

But keeping your information private isn’t just about your own choices. It’s about your friends’ choices, too. Results from a study of a now-defunct social media site show that the inhabitants of the digital age may need to stop and think about just how much they control their personal information, and where the boundaries of their privacy are.

When someone joins a social network, the first order of business is, of course, to find friends. To aid the process, many apps offer to import contact lists from someone’s phone or e-mail or Facebook, to find matches with people already in the network.

Sharing those contact lists seems innocuous, notes David Garcia, a computational social scientist at the Complexity Science Hub Vienna in Austria. “People giving contact lists, they’re not doing anything wrong,” he says. “You are their friend. You gave them the e-mail address and phone number.” Most of the time, you probably want to stay in touch with the person, possibly even via the social media site.

But the social network then has that information — whether or not the owner of it wanted it shared.

Social platforms’ ability to collect and curate this extra information into what are called shadow profiles first came to light with a Facebook bug in 2013. The bug inadvertently shared the e-mail addresses and phone numbers of some 6 million users with all of their friends, even when the information wasn’t public.

Facebook immediately addressed the bug. But afterward, some users noticed that the phone numbers on their Facebook profiles had still been filled in — even though they had not given Facebook their digits. Instead, Facebook had collected the numbers from the contact lists innocently provided by their friends, and filled in the missing information for them. A shadow profile had become reality.
It’s no surprise that a social platform could take names, e-mail addresses and phone numbers and match them up with other people on the same platform. But Garcia wondered if these shadow profiles could be extended to people not on the social platform at all.

He turned to a now-defunct social network called Friendster. A precursor to sites like MySpace and Facebook, Friendster launched in 2002. In 2008, the social site boasted more than 115 million users. But by 2009 people began to jump ship for other sites, and in 2015 Friendster closed for good. Millions of abandoned public profiles vanished into the ether.

Or did they? The Internet Archive — a nonprofit library — has an archive of more than 200 billion web pages, including Friendster. Garcia was able to use that repository to get data on 100 million public accounts from the social media site. Garcia dug through the records in a process he calls Internet Archaeology, after a satirical video from The Onion in which an internet archaeologist announces that he has, ironically, discovered Friendster. “The time scale of online media is very fast, but it’s still studying things in society that don’t exist anymore,” he explains.

Garcia hunted for patterns in the data. Most people don’t have a random assortment of friends. Married people tend to be friends with other married people, for example. But people also have connections that complicate the ability to predict who’s connected to who. People who identified as gay men were more likely to be friends with other gay men, but also likely to be friends with women. Straight women were more likely to be friends with men.

Using this information, Garcia was able to show that he could predict characteristics such as the marital status and sexual orientation of users’ friends who were not on the social media network. And the more people in the social network who shared their own personal information, the more information the network received about their contacts, and the better the prediction about people not on the network got.

“You are not in full control of your privacy,” he concludes. If your friend is on a social platform, so are you. And you don’t have a choice in the matter. Garcia published his findings August 4 in Science Advances.

This does not mean that social platforms are creating shadow profiles of your social media–averse friends, Garcia notes. But with the information people give to social networks and with the platforms’ computing abilities, they certainly could. To prevent the data being used this way, Garcia only used the most basic, public information. He didn’t predict anything about specific people. He only checked to see if it was possible. Garcia also kept the power of his predictions low and very general. And he was careful to not construct an algorithm that could actually build a shadow profile, to make sure that others cannot misuse the findings.

But the results do show that information from your friends on a social network could accurately predict your marital status, location, sexual orientation or political affiliation — information that you may not want anyone to know, let alone in a social network you’re not even on.

“It’s a good illustration of an issue we have in society, which is that we no longer have control over what people can infer about us,” says Elena Zheleva, a computer scientist at the University of Illinois in Chicago. “If I decide not to participate in a certain social network, that doesn’t mean that people won’t be able to find things about me on that network.”

This means we might need to think differently about what privacy means. “We’re used to thinking of having a private space,” Garcia says. “We think we’ve got a room with keys and we let some people in.” But a better image, he argues, might be to imagine ourselves covered in the wet paint of our personal information. If we touch someone else, we leave a handprint. “The more you touch other people, the more you leave on them,” he explains. Touch enough people, and anyone who looks at those people and their paint-covered sleeves will be able to pick out your personal shade of teal.

And because we are no longer in full control of our privacy, Garcia notes, it also means that protecting privacy isn’t something any one person can do. “In some sense it resembles climate change,” he says. “It’s not something you can solve on your own. It’s everyone’s problem or it’s no one’s problem.”

Earthworms are great for soil, right? Well, not always. In places where there have been no earthworms for thousands of years, foreign worms can wreak havoc on soils. And that can cause a cascade of problems throughout an area’s food web. Now comes evidence that invader worms in the Upper Great Lakes may be stressing the region’s sugar maples.

There are native earthworms in North America, but not in regions that had been covered in glaciers during the Ice Age. Once the ice melted, living things returned. Earthworms don’t move that quickly, though, and even after 10,000 years, they’ve only made small inroads into the north on their own.

But people have inadvertently intervened. Sometimes they’ve dumped their leftover bait in worm-free zones. Or they’ve accidentally brought worms or eggs in the soil stuck to cars or trucks. And the worms took up residence as far north as Alberta’s boreal forests.

Earthworms “are not really supposed to be in some of these areas,” says Tara Bal, a forest health scientist at Michigan Technological University in Houghton. “In a garden, they’re good,” she notes. They help to mix soil. But that isn’t a good thing in a northern forest where soil is naturally stratified and nutrients tend to be found only in the uppermost layer near the leaf litter. “That’s what the trees have been used to,” Bal says. Those trees include sugar maples, which have shallow roots to get those nutrients. But worms mix up the soils and take away that nutrient-rich layer.
Bal didn’t start out studying worms in northern regions. She and her colleagues were brought in to address a problem that sugar maple growers were experiencing. Some of the trees appeared to be stressed out. They were experiencing what’s called dieback, when whole branches die, fall off and regrow. This is worrisome because if enough of the tree dies off, “it’s a slow spiral from there,” Bal says — the whole tree eventually dies.
To investigate, the researchers collected data on trees and anything that could be affecting them, from soil type to slope to insects. They looked at trees in 120 plots in Michigan, Wisconsin and Minnesota. And they compared trees that were on growers’ land with those on public land, thinking that how the trees were managed might have some effect.
When the researchers analyzed the data, “the same thing that kept coming up over and over again was the forest floor condition,” Bal says. “That is directly related to the presence of earthworms.” They didn’t go out to look for the worms, but they could see signs of them in the amount of carbon in the soil and in changes in the ground cover. Wildflowers, for instance, were replaced by grasses and sedges, the researchers report July 26 in Biological Invasions.

Bal and her team can’t say what this means for maple syrup production. It might not mean anything at all. But “worms are ecosystem engineers,” she notes. “They’re changing the food chain.” Everything from insects to birds to salamanders could be affected by the arrival of worms.

Even if the sugar maples take a hit, though, there could be an upside, Bal says. These trees are often grown with few other types of trees around. Such a grove is naturally less resilient to climate change and extreme weather. So replacing some of those sugar maples with other trees could result in a healthier, more resilient forest in the future, Bal says.

Zika’s damaging neurological effects might someday be enlisted for good — to treat brain cancer.

In human cells and in mice, the virus infected and killed the stem cells that become a glioblastoma, an aggressive brain tumor, but left healthy brain cells alone. Jeremy Rich, a regenerative medicine scientist at the University of California, San Diego, and colleagues report the findings online September 5 in the Journal of Experimental Medicine.

Previous studies had shown that Zika kills stem cells that generate nerve cells in developing brains (SN: 4/2/16, p. 26). Because of similarities between those neural precursor cells and stem cells that turn into glioblastomas, Rich’s team suspected the virus might also target the cells that cause the notoriously deadly type of cancer. In the United States, about 12,000 people are expected to be diagnosed with glioblastoma in 2017. (It’s the type of cancer U.S. Senator John McCain was found to have in July.) Even with treatment, most patients live only about a year after diagnosis, and tumors frequently recur.
In cultures of human cells, Zika infected glioblastoma stem cells and halted their growth, Rich and colleagues report. The virus also infected full-blown glioblastoma cells but at a lower rate, and didn’t infect normal brain tissues. Zika-infected mice with glioblastoma either saw their tumors shrink or their tumor growth slow compared with uninfected mice. The virus-infected mice lived longer, too. In one trial, almost half of the mice survived more than six weeks after being infected with Zika, while all of the uninfected mice died within two weeks of receiving a placebo.

Using a virus to knock out cancer isn’t a completely new idea. Treatments that rely on modified polioviruses to target tumors such as glioblastomas are already in clinical trials in the United States, and there’s a modified herpesvirus approved by the U.S. Food and Drug Administration for treating melanoma.

These cancer-fighting viruses seem to work in two ways, says Andrew Zloza, head of surgical oncology research at the Rutgers Cancer Institute of New Jersey in New Brunswick. First, the viruses infect and kill cancer cells. Then, as those cancer cells split open, previously hidden tumor fragments become visible to the immune system, which can recognize and fight them.

“Right now we don’t know what kind of viruses are best” for fighting cancer, Zloza says — whether it’s more effective to use a common virus that many people have been exposed to or something more unusual. But now, Zika is yet another candidate.

Further testing is needed to determine whether the virus is safe and effective in humans. Since Zika’s effects are more harmful in developing brains, a Zika-based cancer therapy might be safe only in adults. And the virus would need to be genetically modified to make it safer and less transmissible.
Rich and colleagues are now testing in mice whether combining Zika with traditional cancer treatments such as chemotherapy is more effective than either treatment by itself. Because Zika targets the cells that generate tumor cells, it might prevent tumors from recurring.

Around the six-month mark, babies start to get really fun. They’re not walking or talking, but they are probably babbling, grabbing and gumming, and teaching us about their likes and dislikes. I remember this as the time when my girls’ personalities really started making themselves known, which, really, is one of the best parts of raising a kid. After months of staring at those beautiful, bald heads, you start to get a glimpse of what’s going on inside them.

When it comes to learning language, it turns out that a lot has already happened inside those baby domes by age 6 months. A new study finds that babies this age understand quite a bit about words — in particular, the relationships between nouns.
Work in toddlers, and even adults, reveals that people can struggle with word meanings under difficult circumstances. We might briefly falter with “shoe” when an image of a shoe is shown next to a boot, for instance, but not when the shoe appears next to a hat. But researchers wanted to know how early these sorts of word relationships form.

Psychologists Elika Bergelson of Duke University and Richard Aslin, formerly of the University of Rochester in New York and now at Haskins Laboratories in New Haven, Conn., put 51 6-month-olds to a similar test. Outfitted with eye-tracking gear, the babies sat on a parent’s lap and looked at a video screen that showed pairs of common objects. Sometimes the images were closely related: mouth and nose, for instance, or bottle and spoon. Other pairs were unrelated: blanket and dog, or juice and car.

When both objects were on the screen, the parents would say a sentence using one of the words: “Where is the nose?” for instance. If babies spent more time looking at the nose than the other object, researchers inferred that the babies had a good handle on that word.

When the babies were shown tricky pairs of closely related objects, like a cup of juice and a cup of milk, the babies spent nearly equal time looking at both pictures, no matter what word their parents said. But when the images were really distinct (juice and car, for instance) the babies spent more time looking at the spoken word.
These babies detected a difference between the “milk-juice” pair and the “juice-car” pair, recognizing that one pair is similar and the other isn’t, the researchers conclude November 20 in the Proceedings of the National Academy of Sciences.
To see whether this ability was tied to domestic life, the researchers sent the babies home with specialized gear: vests with audio recorders and adorable hats outfitted with small video cameras, one just above each ear. A camera on a tripod in a corner of the home also captured snippets of daily life. The resulting video and audio recordings revealed that babies whose caregivers used more nouns for objects in the room were better at the word task in the lab.

That means that babies learn words well when they can actually see the object being talked about. Hearing, “Open your mouth. Here comes the spoon!” as they watch the spoon come flying toward their face makes a bigger vocabulary impression than “Did you like riding in the car yesterday?”

A similar idea came from a recent study on preschoolers. These kids learned best when they saw one picture at a time (or when parents pointed at the relevant object). Babies — and older kids, too — like to see what you’re talking about.

The results are too early to provide advice to parents, says Bergelson, a cognitive and developmental psychologist. “But I think one thing suggested by our work is that parents should consider their young baby to be a real conversational partner,” she says. “Even young infants are listening and learning about words and the world around them before they start talking themselves, and their caregivers make that possible.”

There’s still lots to figure out about how babies soak up vocabulary. And as scientists come up with more ways to peer into the mysterious inner workings of a baby’s mind, those answers might lead to even more interesting conversations with our babies.

A new analysis is shedding light on drought in Mongolia, both past and future.

By studying the rings of semifossilized trees, researchers constructed a climate history for the semiarid Asian nation spanning the last 2,060 years — going 1,000 years further back than previous studies.

It was suspected that a harsh drought from about 2000 to 2010 that killed tens of thousands of livestock was unprecedented in the region’s history and primarily the result of human-caused climate change. But the tree ring data show that the dry spell, while rare in its severity, was not outside the realm of natural climate variability, researchers report online March 14 in Science Advances.
“This is a part of the world where we don’t know about the past climate,” says Park Williams, a bioclimatologist at Columbia University’s Lamont-Doherty Earth Observatory in Palisades, N.Y., who was not involved with the study. “Having this record is a great resource for trying to understand past droughts in the region.”

In recent years, many studies have sought to unsnarl the role of anthropogenic climate change from natural variability on extreme weather events (SN: 1/20/18, p. 6). Such work is necessary for more accurately predicting future climate trends and helping governments prepare for the most severe scenarios, says study coauthor Amy Hessl, a physical geographer at West Virginia University in Morgantown. This is especially true in countries like Mongolia that lack certain infrastructure, such as enough water reservoirs, to ease the impact of events like prolonged drought.

Hessl and her colleagues studied tree rings in hundreds of samples of Siberian pines, well-preserved by Mongolia’s naturally dry climate. A ring’s width indicates how much the tree grew in a year. In wet years, the rings are wider; in dry years, skinnier.
The recent dry spell was the severest in recorded history. But the rings showed that an even more severe drought took place around the year 800, long before anthropogenic climate change began.

Still, computer simulations suggest that about a third of the recent drought’s severity could have been caused by elevated temperatures linked to climate change, the researchers found. The finding is consistent with studies on how climate change has affected other recent droughts in South Africa and California.

Using computer simulations, Hessl and her colleagues conclude that droughts in coming decades may not be any worse than those seen in Mongolia’s past. The team predicts that as global temperatures rise over the next century, Mongolia will first become drier, then wetter. Increased heat initially will dry out the plains. But at a certain point, hot air holds more moisture, leading to increased precipitation.

Those climate patterns will likely guide how Mongolia develops, Hessl says, because they have in the past. In 2014, she and colleagues published a paper detailing how a 15-year period of unprecedented temperate and rainy conditions in 13th century Mongolia may have led to the rise of Genghis Khan (SN Online: 3/10/14).