Day: 9 January 2019

Kenichi Yamada, Nikkei Asian Review:

Samsung Electronics’ disappointing fourth-quarter earnings report comes less than a week after Apple cut its revenue estimate, underlining the global repercussions of the Chinese economic slowdown.

Both of the South Korean company’s core businesses — memory chips and smartphones — are facing downturns this year. With no other growth business to fill the hole, Samsung finds itself scrambling to trim the fat.

Samsung said Tuesday that operating profit plunged 29% in the three months ended in December, to 10.8 trillion won ($9.6 billion). The preliminary guidance represents the company’s worst quarter since July-September 2016, with the drop exceeding analyst expectations by 2 trillion won to 3 trillion won.

That second quoted paragraph is particularly interesting: Samsung is doubly exposed because it makes its own consumer electronics and sells components for other companies’ products.

Anyway, it wouldn’t surprise me to see reports like these from several other companies in the coming weeks.

Aliya Ram and Madhumita Murgia, Financial Times:

Data brokers mine a treasure trove of personal, locational and transactional data to paint a picture of an individual’s life. Tastes in books or music, hobbies, dating preferences, political or religious leanings, and personality traits are all packaged and sold by data brokers to a range of industries, chiefly banks and insurers, retailers, telecoms, media companies and even governments. The European Commission forecasts the data market in Europe could be worth as much as €106.8bn by 2020. 

“The explosive growth of online data has led to the emergence of the super data broker — the ‘privacy deathstars’, such as Oracle, Nielsen and Salesforce, that provide one-stop shopping for hundreds of different data points which can be added into a single person’s file,” says Jeffrey Chester, executive director of the Center for Digital Democracy based in Washington. “As a result, everyone now is invisibly attached to a living, breathing database that tracks their every move.”

Over the past five years, the data broker industry expanded aggressively in what amounted to a virtual regulatory vacuum. The rise of internet-connected devices has fuelled an enhanced industry of “cross-device tracking” that matches people’s data collected from across their smartphones, tablets, televisions and other connected devices. It can also connect people’s behaviours in the real world with what they are doing online. 

The reluctance in virtually every country to restrict the purchase and sharing of user data without explicit consent is a complete regulatory failure. Nobody would tolerate someone asking them to submit a list daily of everything they’ve bought, every page they’ve seen online, every ad they’ve viewed, and everywhere they’ve been — not because that would be a lot of work, but because it would feel invasive. There shouldn’t be a “data market” at all.

I love this collection of examples by Dr. Drang of how Siri handles date and time questions. In some instances, its understanding of context is remarkably good; in a few, it’s disappointing. The key point, for me, is something Drang writes at the end:

The upshot is that Siri can be good at date and time math, but it needs the right syntax. Not surprising for a computer program, but not how Siri has been promoted by Apple.

I was chatting with a friend about this. Make no mistake: Siri is far better at parsing different variations of sentence construction than traditional voice control systems. But users must still learn some syntactical tricks to ensure Siri understands the precise intention and context of what is being said.

I also wonder how much of this is because of the English language; many other languages create more rigorously-structured sentences.