Archive for June, 2009

Role of privacy in protecting political freedoms

Tuesday, June 30th, 2009

Josh Hall brings to our attention an article in Technology Review by Simson Garfinkel. While the article focuses on online privacy, below are some excerpts that may be useful for our purposes here:

Privacy gives us the right to meet and speak confidentially with others—a right that’s crucial for democracy, which requires places for political ideas to grow and mature. …

Collectively, we made things worse by not building strong privacy and security guarantees into our information systems, our businesses, and our society…

Another law, the Video Privacy Protection Act of 1988, makes it illegal for Netflix to disclose the movies you rent…

A Nixon administration advisory committee then developed the Code of Fair Information Practice, a guiding set of principles that underlies the majority of U.S. privacy laws passed since.

This code is surprisingly straightforward. There should be no secret data banks; individuals must be able to view their records; there must be a way to correct errors; organizations maintaining data banks must make sure they’re reliable and protect them from unauthorized access; and information collected for one purpose must not be used for other purposes…

Congress, however, opposed TIA on the grounds that it treated everyone in the country as a suspect, and because it feared that a massive data surveillance system might be used for purposes other than catching terrorists. This prospect was not so hypothetical: in 1972 Richard Nixon had ordered the IRS to investigate his political opponents, including major contributors to George McGovern’s presidential campaign…

For more than 100 years, American jurisprudence has recognized privacy as a requirement for democracy, social relations, and human dignity. For nearly 50, we’ve understood that protecting privacy takes more than just controlling intrusions into your home; it also requires being able to control information about you that’s available to businesses, government, and society at large. Even though Americans were told after 9/11 that we needed to choose between security and privacy, it’s increasingly clear that without one we will never have the other.

Simson’s proposal for a government-issued online identity will be very controversial—see Bruce Schneier for the opposite view—but his discussion of the role of privacy in protecting freedom is useful separately from that.

I increasingly feel we need a different term than “privacy” for what we are trying to protect. Privacy has negative connotations. Any ideas? —Chris Peterson

Sensors at Google Internet Summit: see video

Monday, June 29th, 2009

Increasingly when we look at sensing we need to look at wireless technology as well. The topics were covered in one session at this year’s Google Internet Summit. Speakers were Craig Partridge (BBN), Larry Alder (Google), Sumit Agarwal (Google Mobile), Kevin Fall (Intel), and Deborah Estrin (UCLA).

For those of us who missed the conference, videos are posted. This session lasts 90 minutes; I haven’t tackled it yet but the talks by Estrin, Agarwal, and Partridge look relevant to our interests, and maybe the others as well. Partridge refers to a 10-20 year timeframe, by when this whole field should be doing amazing things (for good and maybe ill as well).

One problem with the “video-ization” of the world, as opposed to accessing text, is the difficulty of skimming to see if something is worth viewing/reading. I look forward to a future feature on YouTube to enable seeing a few seconds’ clip every five minutes or so. Thanks to for the pointer.

Slashdot discusses Open Source Sensing

Thursday, June 25th, 2009

Jeff Ubois brings our attention to a discussion over at Slashdot of our project, including this response to a comment that open source hardware is much more expensive than open source software:

Your point is valid, in that hardware has a higher cost of entry than software (and the relative levels of maturity of OSS vs. Open Hardware reflect this); but I’m not sure that it applies as much as you suggest.

Designing, fabbing and installing specialized sensors is one aspect of “sensor technology” and one that OSS is, as you say, arguably of limited use as a model. However, co-ordination of sensor values, turning the data points into some meaningful picture of the world, is more or less completely a software problem.

Also of note is the idea of building distributed sensor networks out of what already exists, which is largely a problem of software, creativity, and social structuring. For instance, consider the sheer number of cameras, accelerometers, and RF receivers, all connected to programmable computers and radio modems, that are running around the streets in the form of cell phones. For that matter, think of the giant crowds of happy-snapping tourists in most tourist destinations as constituting a sort of camera network.

The business of actually putting hardware into the field will, as you say, likely remain a more or less closed commercial enterprise, with some open source/DIY projects here and there (just as most software, even OSS, runs on commercial hardware); but there is a lot of room for OSS models in the systems and software that tie the sensors together, and make something useful of their output.

Welcome, Slashdotters, to the debate and (ideally) the project! —Chris Peterson

Sensors get more intimate

Wednesday, June 24th, 2009

The LA Times reports on work by the USC School of Medicine and School of Engineering:

The team has devised a wireless body area network that includes an accelerometer, a heart rate monitor, a GPS device and a sensor that measures the electrical conductivity of skin. All the data is relayed to a Nokia phone, which then transmits it to a secure server.
The device will become so attuned to its wearer that the researchers have dubbed this the KNOWME Networks study, said Donna Spruijt-Metz of the Keck School of Medicine, who described the project last week at the Childhood Obesity Conference in Los Angeles.

“We’ll train the sensors to guess pretty well what you’re doing,” she said.

I bet she’s right! Speaking as one who is somewhat addicted to wearing a pedometer, I predict that this kind of thing will catch on. So it would be a good idea for us to make clear that we’d like to know what the systems we’re wearing are sensing, and where that data is going… —Chris Peterson

Vancouver moves to Open Data…but what data should be open?

Monday, June 22nd, 2009

Jennifer Kavur of Infoworld brings great news from Vancouver:

Open source activists are praising the Open Data, Open Standards and Open Source motion passed by the City of Vancouver last month.

Proposed by City Councillor Andrea Reimer, the motion encourages the adoption of open standards, promotes distribution of open data and places open source on equal footing with commercial software during procurement cycles.

Vancouver is the first municipality in Canada to pass a motion that embraces the “open” city concept. But “we took some of our lead from Toronto, who did a 1.0 version of a motion last fall and is looking at rolling some stuff out,” said Reimer.

There is no question but that this is the right direction to go. Next, we need to help visionary cities and other governments think clearly about Open Data: what should be open, and what shouldn’t. Given that they are still struggling with the concept of open source software, this is going to take some serious work. (Credit: Open Source Technology) —Chris Peterson

Closed-source sensors: how good are the best ones today?

Friday, June 19th, 2009

For Open Source Sensing to succeed, we need to be familiar with the best proprietary sensors available today. Check out the “Best of Sensors Expo” Award Winners. Examples:

Gold Awards

Category: Sensors
DTS’ 6DX rugged, ultra-small 6 degrees of freedom sensor
Category: Data Acquisition Products
MicroStrain Inc.’s HS-LINK high-speed wireless sensor node
Category: Sensor Components
PowerCast Corp.’s P2100 Powerharvester energy harvesting module
Category: Sensors
SBG Systems’ IG-500N miniature Attitude and Heading Reference System (AHRS)

How can open source sensing interact with proprietary sensors? Should we focus instead on data-handling, from any sensor? Or start there, and then expand to software, and then hardware? —Chris Peterson

Monitoring the rainforest: pluses and minuses

Wednesday, June 17th, 2009

Rodney Gedda of ComputerWorld in Australia brings news of rainforest monitoring:

The CSIRO will install some 200 solar-powered, sensor network nodes over the next two years to helping rehabilitate rainforest in the Springbrook World Heritage precinct in south-east Queensland, the research group announced today…

They can measure biodiversity indicators, like bird and frog calls, and physical characteristics like leaf wetness, soil moisture and temperature.

Since May 2008 a network of 10 wireless sensor nodes has been sampling physical parameters, including soil moisture and the amount of available light inside the forest, every five minutes…

CSIRO’s Sensor Network Technologies research director, Dr Michael Bruenig, said the Springbrook project demonstrates that real-time data can be streamed back from open and covered rainforest using a low-bandwidth, wireless sensor network.

From an environmental perspective, this is absolutely wonderful; let’s get these deployed ASAP. From a political perspective, it makes clear that the day is rapidly approaching when there will be no place to go to talk about forbidden matters. Australia is a democratic nation, but what does this mean for less-free countries? (And even good countries can go bad.) Let’s design environmental sensors that provably don’t record, say, human conversations.

Open source personal robot platform with sensors

Monday, June 15th, 2009

I will be out of the country when this talk happens: can someone go and ask whether the sensor suite is open source? —Chris Peterson (Info forwarded by Ravi Mistry)

“An open platform for robotics research”

Personal Robotics research and development are accelerating. A
growing community of Open Source developers is creating a platform
called ROS that anyone can build on to make the breakthroughs that
will lead to new applications. Willow Garage is building a personal
robot platform, PR2, which has two arms, a mobile base, and a rich
sensor suite. This talk will review the current status of ROS and
PR2, and discuss opportunities to join the ROS Open Source community. (more…)

‘Towards privacy-sensitive participatory sensing’

Wednesday, June 10th, 2009

From the Fifth IEEE International Workshop on Sensor Networks and Systems for Pervasive Computing, three authors from Australia bring us Towards Privacy-Sensitive Participatory Sensing:

The ubiquity of mobile devices has brought forth the concept of participatory sensing, whereby ordinary citizens can now contribute and share information from the urban environment. However, such applications introduce a key research challenge: preserving the location privacy of the individuals contributing data. In this paper, we propose the use of microaggregation, a concept used for protecting privacy in databases, as a solution to this problem.

More of this kind of creativity, please. —Chris Peterson

Korea starting ‘Ubiquitous Sensor Network’

Tuesday, June 9th, 2009

Kim Tong-hyung of The Korea Times brings news of a new sensor project in that country:

The idea is to combine the sensors and closed-circuit camera (CCTV) information under a broadband convergence network (BcN), or a planned massive IP providing connection speeds between 50 to 100 megabytes per second (mbps), the Internet protocol version 6 (IPv6) and other next-generation Internet platforms.

The gathered information would be accessible anytime and anywhere by mobile, through the country’s advanced second and third generation (3G) wireless networks and WiBro, the local variant of mobile WiMAX…

Seoul will look to explore the possibility of using IP-USN for providing real-time traffic information, and monitoring road conditions and bus movements, while also evaluating air pollution levels. The information will be managed through the WiBro network.

Chuncheon is planning to use IP-USN to construct a “smart well-being leisure city,” using sensors to provide pulse monitoring and other health information to joggers through mobile devices, and the network for monitoring air pollution.

That does indeed sound pretty thorough. Maybe we could get this project to collaborate with us here on openness and data-handling issues? —Chris Peterson