Friday, November 14, 2014

Boot up: public Wi-Fi risks, four deadly numbers, YouTube payment queries

A burst of 9 links for you to chew over, as picked by the Technology team

Maurits Martijn:

In his backpack, Wouter Slotboom, 34, carries around a small black device, slightly larger than a pack of cigarettes, with an antenna on it. I meet Wouter by chance at a random cafe in the center of Amsterdam. It is a sunny day and almost all the tables are occupied. Some people talk, others are working on their laptops or playing with their smartphones.

Wouter removes his laptop from his backpack, puts the black device on the table, and hides it under a menu. A waitress passes by and we ask for two coffees and the password for the WiFi network. Meanwhile, Wouter switches on his laptop and device, launches some programs, and soon the screen starts to fill with green text lines. It gradually becomes clear that Wouter's device is connecting to the laptops, smartphones, and tablets of cafe visitors.

Total cholesterol, HDLcholesterol, blood pressure and smoking. That's all. (Many people probably only know one of those; most, only two.)

It's been several months since Gamasutra investigated the ethics of paying for YouTuber coverage for video games, and the disclosure that is required in these circumstances to meet legal requirements.

But while some YouTubers have taken this on board, many others have actively ignored the advice, or questioned whether they really need to provide the disclosure suggested.

Numerous YouTubers have been in touch with Gamasutra in the last few months, both to question exactly what sort of disclosure is required of them, and to provide plenty of examples where big-name YouTubers and YouTube networks are telling their clients to ignore the advice.

So Gamasutra asked the US FTC what disclosure was required. Answer: it's not that simple, because of embedding.

The company reported net income of $3.72bn in the period.

Breaking down Google's revenue by segment, you have the following:

• Sites revenue: $11.25bn. Up 20% from the preceding year's quarter.

• Network revenue: $3.43bn. Up 9% from the preceding year's quarter.

• International revenue: 58% of revenue. Up 2% from the preceding year's quarter.

The company reported that its average cost-per-click fell 2% in the period, a weakening in a key revenue source. However, pushing back against that decline was a 17% year-over-year increase in "aggregate paid clicks". So, while the amount of revenue that Google managed to extract from a click went down mildly, it sold more than one-sixth more compared to the year-ago quarter.

It is becoming more expensive to run Google. The company's earnings report details its increasing cost base: "Operating expenses, other than cost of revenues, were $6.10bn in the third quarter of 2014, or 37% of revenues, compared to $4.58bn in the third quarter of 2013, or 33% of revenues." Put another way: Building out new revenue streams, and investing in future products that won't monetize for some time isn't cheap.

Android's dominance of the smart watch market will come to an end in 2015 when the operating system's (OS') share of the sector will fall to under 50% for the first time as it succumbs to pressure from Apple.

ABI Research forecasts the share of Google Android powered smart watches will fall even as device shipments using the OS rise. The research company predicted that the number of Android smart watches shipped will grow from around 6m units in 2014 to 15m in 2015, but noted that shipments of Apple's Watch will begin to take share from Android rivals when it launches in 2015.

Apple's iOS is tipped to take a 35% stake in the smart watch market in 2015, and go on to establish at least a 50% share in the years after, according to ABI Research. In contrast, Android's share is tipped to fall from 67% in 2014 to 42% in 2015. The research company noted that Android accounted for 75% of smart watch shipments in 2013.

Something of a reverse of the usual story.

All that you could ever want to know (and slightly more you probably don't) about the denizens of Pornhub in terms of their operating system (and search) preferences. Useful for the data about Windows Phone and BlackBerry country distribution.

Tina Lu:

A few days ago, after checking the mails in my smartphone while waiting for a friend in a local café in Buenos Aires, Argentina, I raised my eyes and started to look what people around me were doing, a habit which I picked up during my time in the Consumer Insights team at Bestfood-Unilever. I realized that everybody around me was using some sort of electronic device!

Most people in groups have a smartphone, while half of the people that were alone have a tablet. I remember that a couple of years ago in the same situation, you were lucky to find one person with a tablet in a café. More users in Latam are moving to smartphones and tablets. Many are just adding one more device to their existing electronic devices portfolio. However, many users are leapfrogging directly into a tablet device, specially teenagers and kids.

The tablet market has reached around 17M units in 2013, with a growth rate of approximately 120% (Couterpoint Research estimation). Growth for 2014, might not reach three digits, but still should be higher than 50%. Growth rates differ substantially among countries, so are their drivers and constraints.

However, it's more subtle once you start digging into specific countries.

I see Twitter getting beaten up a lot for not deleting the spammers faster. Etsy gets beaten up for not deleting the "resellers" faster. Flickr used to get yelled at for not catching the photo stealers or porn spammers faster.

"It's so fucking easy, they're right over there, here, let me show them to you, what's your problem?"

This comes from not understanding the cost benefit ratio of false positives in identifying abuse of a social site at scale.

Imagine you've got a near perfect model for detecting spammers on Twitter. Say, Joe's perfectly reasonable model of "20+ tweets that matched '^@[\w]+ http://'". Joe is (presumably hyperbolically) claiming 99% accuracy for his model. And for the moment we'll imagine he is right. Even at 99% accuracy, that means this algorithm is going to be incorrectly flagging roughly 2 million tweets per day as spam that are actually perfectly legitimate.

If you've never run a social software site (which Joe of course has, but for the folks who haven't) let me tell you: these kinds of false positives are expensive.

Read on to realise just how expensive. (Hint: so much so that two sides of the business will effectively be at war.)

A large data breach at Home Depot Inc has started to trigger fraudulent transactions that are rippling across financial institutions and, in some cases, draining cash from customer bank accounts, according to people familiar with the impact of the hacking attack.

The fraudulent transactions are showing up across the U.S. as criminals use stolen card information to buy prepaid cards, electronics and even groceries, these people said. In some cases, the fraudulent transactions have been tracked to batches of cardholder accounts that are tied to specific ZIP Codes, they said…

The trends are all too familiar to thousands of the nation's financial institutions, which have spent much of the year trying to root out fraudulent transactions tied to breaches at merchants like Target Corp., Neiman Marcus Group Ltd., grocer Supervalu Inc. and Asian restaurant chain P.F. Chang's China Bistro Inc.

56m card details exposed. And the US is still not using chip-and-pin.

You can follow Guardian Technology's linkbucket on Pinboard

To suggest a link, either add it below or tag it with @gdntech on the free Delicious service.


View the original article here

No comments:

Post a Comment