1. Lawyers. Lawyers are in their element when they are enforcing the status quo. They are the "sticky" in "sticky spaghetti".
2. Bureaucrats. We all know what a bureaucrat is. I think the most direct definition is this: it is someone who succeeds by effectively making others fail.
3. Fear of failure. This is a hard one. In fact we probably all fear failure to some extent (yes, I acknowledge that the rare exception exists, and we need these people too...) This is probably an important survival trait, but in modern society subverting it can be a challenge.
4. Paralysis by analysis. This is common in business. We need analysts. People who can sift through the data, add up the columns, and come up with the decision that maximizes benefit while minimizing risk, etc., etc. I just don't want these people responsible for actually making the decision. That is just plain dangerous.

On Monday Location Labs announced the initial release of our Geofence Library for the iPhone platform in private beta. In many ways we believe this solution will set the standard both for how geofencing services are provided on the iPhone and other smartphone platforms, and more generally how location data streams will be derived from the mobile device.

Wikipedia defines a geofence as a "virtual perimeter for a real-world geographic area". In the mobile context, when a mobile device enters or exits the geographic area, the relevant caller (typically a mobile-resident or network-based application) is notified. Sounds simple enough, but of course the devil is in the details.

There are two key factors that complicate the implementation of a geofence solution in a mobile context. The first is location technology. Deriving location from a mobile device is not a new problem, dating back at least to the FCC E911 mandate, and as it turns out there is no single one-size-fits-all solution to this problem. As with many important engineering problems, there are trade-offs. The second, and related factor is battery consumption. For example, GPS, an important enabling technology for mobile location, is a battery hog, typically drawing 300-350 mA on a modern smartphone device (as reference, typical smartphone batteries have a capacity in the 1,200-1,500 mA hour range.)

On Monday, Apple released its latest version of the iPhone operating system, iOS4. One of the important advances Apple provided as part of this release is the ability to support background processes. This rectified an important deficiency in the iPhone platform, as other modern competing platforms have supported this for some time (Android, RIM/Blackberry, Symbian, etc.) It is also essential for supporting a geofencing capability.

The iPhone provides two separate capabilities relevant to the problem of background location access and geofencing. The first option, available in all versions of the iPhone operating system, is the standard location service. This is a callback-based API that will notify the caller when location changes based on configurable accuracy and distance change filters. The particular location technology used by the underlying OS is determined by Apple, and may involve GPS, WiFi triangulation or cell/sector triangulation. The second option is the newly introduced "significant change location service", which is a low-battery consumption option that presumably updates only on cell/cell-sector change, and will report even if the application has been suspended.

As it turns out, the iPhone significant change location service is simply not adequate for geofencing applications. In in-house testing on 3G and 3Gs devices, we see notification delays of up to multiple hours. It appears that Apple is too conservative here in terms of battery consumption.

A second, extreme approach would be to run GPS continuously (this can be accomplished using the standard location service with optimal accuracy settings.) Of course, if GPS runs continuously, you can trigger notifications very promptly when a geographic boundary is crossed. Along with this extremely low trigger latency behavior comes dramatic battery consumption which renders the battery drained within less than six hours, depending on the health of the battery and other service use.

Location Labs has taken a significantly more sophisticated approach. With the iPhone, we employ a combination of techniques and heuristics involving both the standard and significant change location services, intelligent interaction with the iPhone backgrounding and suspending logic as well as local awareness of proximity to the geofence boundaries. Together these allow us to offer a high quality firing latency guarantee (measured in minutes) while keeping impact on battery life to a minimum.

The release of our Geofence Library for iPhone is available in private beta, follow this link for details on participating.

- @sahotes

There's no alternative "lifestyle" for Holden Caulfield or the members of the Glass family to retreat to, as there is for the Beats, no group of like-minded adventurers. Salinger's characters aren't after thrills. Their quest is for an impossible purity that drives them away from the workaday world, toward a dangerous, self-burying seclusion. "We're...freaks with freakish standards," says Zooey Glass to his sister Franny. "We're the Tattooed Lady, and we're never going to have a minute's peace, the rest of our lives, till everybody else is tattooed too."

1. The User Profile. This would typically include some kind of identity information (such as reliable contact information), personal information about the user, and personal media, such as photos and videos.
2. The Social Graph. That is, some definition of who the user's friends and family are, or more generally, who they wish to communicate with.

1. It's not at all clear why sharing location information with my service provider would imply a willingness to share it with the government.
   
   
2. In order to provide cellular service, the service provider also has access to a variety of other information, including who the user communicates with, and the information communicated. They certainly need to know the former, and for practical purposes have access to the latter. Does this then imply that there is no reasonable expectation of privacy regarding this information?

1. Appropriateness: Is the collection of location information appropriate given the context of the service or application?
2. Minimization: Is the minimum necessary granularity of location information distributed or collected?
3. User Control: How much ongoing control does the user have over location information? Is the user a passive receiver of notices or an active transmitter of policies? Are there defaults? Do they privilege privacy or information flow?
4. Notice: Can requesters transmit information about their identity and practices? What information is required to be provided to the user by the requesting entity? What rules can individuals establish attach to their location information and transmit? Is there a standard language for such rules?
5. Consent: Is the user in control of decisions to disclose location information? Is control provided on a per use, per recipient or some other basis? Is it operationalized as an opt-in, opt-out or opt model?
6. Secondary Use: Is user consent required for secondary use (a use beyond the one for which the information was supplied by the user)? Do mechanisms facilitate setting of limits or asking permission for secondary uses?
7. Distribution: Is distribution of location information limited to the entity with whom the individual believes they are interacting or is information re-transmitted to others?
8. Retention: Are timestamps for limiting retention attached to location information? How can policy statements about retention be made?
9. Transparency and Feedback: Are flows of information transparent to the individual? Does the specification facilitate individual access and related rights? Are there mechanisms to log location information requests and is it easy for individuals to access such logs?
10. Aggregation: Does the standard facilitate aggregation of location information on specific users or users generally? Does the specification create persistent unique identifiers?

My cousin just sent me a link to The Story of Stuff, a wonderful little flash animation detailing our consumer-driven society and the problems it presents. Extremely creative with a message that is tough to swallow. The kind of thing that would be great to put in front of school children, and other people with open minds...]]> http://sahotes.com/blog/2008/01/the-story-of-stuff.html http://sahotes.com/blog/2008/01/the-story-of-stuff.html environment Thu, 31 Jan 2008 10:31:24 -0800 recent op-ed piece in the NY Times, Paul Davies stirs up the long-standing argument that the reliance on "immutable" mathematical laws by physical science is an act of faith, and thus (or in this way) science is no different than religion.

More interesting, perhaps, is this series of rebuttals to this tired assertion by The Reality Club at edge.org. Choice quotes include:

Alas, Davies also brings up the anthropic principle, that tiresome exercise in metaphysical masturbation that always flounders somewhere in the repellent ditch between narcissism and solipsism.

  -- PZ Myers

Science is, indeed, founded on the working hypothesis, one amply borne out by four centuries of scientific practice, that the world, or at least some aspects of it, is ordered in a stable and intelligible way. But that tentative and partly testable working hypothesis is a far cry from religions' reliance on sacred texts and personal revelations. To characterize these radically dissimilar endeavors as both based on "faith" is to point out a superficial commonality while obscuring the fundamental difference. And at a time when humanity is wracked by conflict between incompatible versions of faith, in the genuine sense of the term, to muddy the distinction between religion and science is worse than philosophically misguided: it is irresponsible.

  -- Alan Sokal

A few years ago, when I read the recently released On Intelligence by Jeff Hawkins, I was, to say the least fascinated, if not downright inspired. Jeff Hawkins, inventor of the Palm Pilot and Handspring Treo, had codified and clearly explained recent results in our understanding of how the brain works, and in particular how memory could be modeled in the neocortex.

In the book, Hawkins not only provides an overview of recent work, but combines and extends these ideas into a concrete hypothesis for effectively modeling neocortex brain function. These ideas have been formalized into a whitepaper. Hawkins published the book in 2004, and a year later formed a company called Numenta (along with long-time business partner Donna Dubinsky and technologist Dileep George) to help translate these ideas into reality.

When I first heard about Numenta, I was skeptical about its viability. It's certainly a noble goal to make practical the theoretical ideas outlined in the book, and nothing would lend more credibility to these ideas than commercial success. However, the ideas/concepts seem to be fairly low level from the perspective of building commercial solutions. Say, for example, that I build a commercially viable image recognition system based on the toolkit provided by Numenta. Is it likely that a great deal of IP would be attributable to the toolkit? Or to the concepts underlying Hawkins' program (something called HTM)? If it's the latter, then really Hawkins is commercializing the concept (I have doubts about the former.) HTM can be defined as a type of Bayesian network. It's not clear that patenting these concepts would not run afoul of previous academic research into brain function (recall that really these concepts were known previously, the contribution here was aggregating them into a practical whole.)

In any case, in a recent press release, DARPA has announced that it has awarded a $4.9M grant to Lockheed Martin to develop just such a program. This brings great credence to what Numenta is doing, and outlines a fairly clear path to success. Now more than ever, this book may be worth a read!]]> http://sahotes.com/blog/2007/12/on-intelligence-numenta-and-da.html http://sahotes.com/blog/2007/12/on-intelligence-numenta-and-da.html books AI Sun, 02 Dec 2007 23:36:28 -0800

Yesterday I saw the movie No Country For Old Men based on the novel of the same title by Cormac McCarthy. McCarthy is my favorite living author, so I had been looking forward to this, although after what Matt Damon and Penelope Cruz had done to "All the Pretty Horses", I had some concerns. "No Country For Old Men" was, however, directed by the Coen brothers, so I had some hope.

In short, the film did not disappoint. Most inspiring was the bringing to life of the character Anton Chigurh, the mass-murdering anti-christ that provides the vehicle for McCarthy to set the philosophical underpinnings of the story. Chigurh is played by Javier Bardem, and it would not be a surprise if he were to win a best supporting role Oscar for this work.

The film (and the novel) are not for everyone. The theater was nearly full the other night, and a good part of the crowd actually booed at the end (although my guess is that this was mostly for the abrupt ending, true to the novel to the letter, no hollywood, feel-good resolution here.)]]> http://sahotes.com/blog/2007/11/no-country-for-old-men.html http://sahotes.com/blog/2007/11/no-country-for-old-men.html books books Sun, 25 Nov 2007 23:32:57 -0800

A year or so ago, I read a fascinating book by mathematician Peter Woit entitled Not Even Wrong, subtitled "The Failure of String Theory and the Continuing Challenge to Unify the Laws of Physics".

Woit makes a compelling argument characterizing the current field of theoretical particle physics as a cult devoted nearly exclusively to following string theory despite the fact that the theory has failed over the past thirty years to make any real progress in going beyond the standard model. Thirty years later, it still appears that graduate students are advised that they leave the field at their peril.

This story is personally relevant as I earned my Ph.D. in string theory from the University of California at Berkeley in the fall of 1992, and have had similar suspicions since leaving the field. This is the first time, however, that I have seen the arguments so cleanly laid out. It certainly has been frustrating to sit by and watch as proponents of the field (such as Brian Greene and Michio Kaku) continue to speak overly optimistically about the progress being made in the field.

A great read!]]> http://sahotes.com/blog/2007/11/string-theory-not-even-wrong.html http://sahotes.com/blog/2007/11/string-theory-not-even-wrong.html books books physics Sun, 25 Nov 2007 23:10:13 -0800