Author: plain

How (Not) to Buy an Academic

By : Casey Fiesler Sadly this is not the life of a university professor. (“Sport of Tycoons” by Carl Banks, Flickr/CC-BY-NC-ND) Here’s a scary thought: “Google has been paying university professors for favorable policy papers.” The basic idea, put forth by the Google Transparency Project (GTP), is that Google is paying millions of dollars to academics in order to “buy” research that is favorable to their policy positions. Their report is called “Google Academics, Inc.” Catchy, I guess. It plays into dominant narratives that tech companies are evil. But it also plays into an increasingly troubling narrative that science can’t be trusted. In an attempt to discredit Google, this project’s report also implies that academics can be easily bought off. That research is for sale to the highest bidder. It is an easy leap from this concept to thinking that a political party is paying for climate change research, for example — and this is a narrative that is dangerous right now. To be clear, the issue of research funding not being disclosed is a big deal. And it definitely happens. The Wall Street Journal article based in part on GTP’s data seems to have uncovered some problematic individual cases. And we do already know that this is an issue in the pharmaceutical industry, for example. But the good news is, this is something that must not happen very often in...

Read More

Updating 8.2 million km² of high-resolution satellite imagery

By: Camilla Mahon We just purchased 8.2 million km² of high-resolution satellite imagery from DigitalGlobe: enough to cover 10,000 metropolitan areas the size of New York City all around the world. And it looks beautiful! Behind the scenes This new imagery is from DigitalGlobe’s ultra-high-res satellites. They collect data in state of the art sub–50 cm pixels. At that sharpness, people can see details of buildings, pathways, and parks, making it easy to orient in the landscape and find everything from shops to mailboxes. It also allows autonomous vehicles and machine learning algorithms to see lane markings, speed bumps, and other ingredients of HD maps How much of a difference does sharpness make for algorithmic work? Consider that a compact car takes up an area of fewer than 8 pixels in one-meter imagery — barely enough to see which way it’s pointing. In best-resolution satellite imagery, the same car is represented by more than 10 times as many pixels. Combining human and data insights Knowing where to update imagery at this scale is hard. We have users all over the world, and they have many different coverage priorities for their satellite views: roadways, new developments, commercial districts, farms, national parks, waterfronts, you name it. Even for the same point on the ground, people have different requirements. For example, recency is the main concern for some, while others would gladly see last year’s images...

Read More

What’s Proximity Computing?

An explanation and the three different types Edge computing is a means of processing data physically close to where the data is being produced, i.e. where the things and humans are — in the field area, homes, and remote offices. Since they don’t live in the cloud, we need to complement cloud computing with many forms of computing at the edge to architect IoT solutions. Discussions about edge computing often overlook how many types of “edge” computing there are. We’ll take a look at the fundamental drivers for edge computing and the many types of edges to consider. Since we’re referring to computing close to the source of things, data, and action, I will use a more generic term for this type of data processing: proximity computing. The Economics of Proximity Computing Events in our world need a timely response either for good user experience (“changing TV channels”) or to avoid catastrophes (“gas line leak”). As these events occur, we need to choreograph complex systems to Sense, Process, and Act (SPA). The cost of SPA’ing is a function of local vs. remote processing costs, network connectivity costs and remote systems management costs. Size and power aside, proximity computing balances the timeliness of responding to an event with the cost of that response. Legal restrictions on data traversing jurisdictions further drive the need for proximity computing. Many-Edged Landscape — Types of Proximity Computing When...

Read More