Ron Guest

Lytro has developed a technology to allow you to select the focus point in photographs after taking a photo. This isn’t a sharpening technique but instead is a fundamental capture of light field information that allows amazing post-processing results. What’s really cool is the company states this technology can be applied to virtually any camera technology including smartphones and point & shoot cameras. They’re planning to introduce their first camera by the end of the year and ultimately license the technology. They’re claiming cameras with their technology will be competitively priced. It will require a new image format for the “raw” data and specialized post-processing software. Still, this will be some amazing tech if it lives up to the hype. Visit the Picture Gallery for a demonstration. 

Light Field camera

As I discussed in a recent post, the relative performance and stability of WiFi to 3G/4G/LTE is going to dictate an ever-expanding role of WiFi for smartphone and tablet users. Building out cellular data networks is just too costly and slow (largely due to permitting) to keep up with demand. GigaOm has a great article today on Japan’s KDDI plan to use of WiFi to supplement  their cellular network.

Wi-Fi: It’s the other cell network — Broadband News and Analysis.

Refreshment

Some of my favorite things…

UFOs

A prior post talked about a lightweight and reliable method for sending push notifications from an Arduino to a mobile phone. In this post I’ll to describe my application for the notifications.

When looking for project inspiration I saw a couple of designs for monitoring a postal mailbox and sending a notification when the door was opened (e.g. when you get new mail). Those didn’t seem very useful but they did tickle a notion I’ve had of knowing if my garage door was left open for a period of time. I see this happen regularly in our neighborhood and it sometimes results in the loss of bicycles, tools and once even a car. In the spirit of truth-in-posting we once left ours open all night. Fortunately nothing was taken. Now on to the build…

The parts list, other than a basic pullup resistor and some wire, is pretty short and consists of

• " target="_blank">Arduino Uno
• WiFly WiFi Shield
• Arduino Project Enclosure
• Arduino Wall Adapter
• Magnetic Proximity Sensor, and
• A pair of Screw Terminals

In simple terms the goal of the project is to send me an alert if the door is open longer than a specified period of time (e.g. 15 minutes) and to send me an update if the door closes after that time period. Quite simple eh? I also implemented a “health check” that enables the Arduino to ping my server periodically. A script runs on the server and if it doesn’t hear from the Arduino it too will send an alert letting me know there has been a monitoring failure. After all what good is a security device that fails silently?

If you’re interested in a project like this let me know and I can give you the source code (sketch).

[slideshow]

Although I’m surely in the camp of folks who’d love unlimited high-speed Mobile data (3G/4G/LTE/…) for the rest of my life I recognize that isn’t a viable business model.  Although Wifi service is also becoming metered the offered capacities and performance at a given cost are superior. It simply costs less to provide a specific level of service using this type of technology than using Mobile spectrums. Not only is the up front cost lower but the ongoing costs of keeping up with new technology is lower as well (i.e. upgrade costs). Time to market is another issue as AT&T has publicly stated multiple times one of the reasons they can’t keep up with demand is the inability to gain rights to add new cell sites or upgrade existing ones. We should ultimately expect a Wifi descendant to be our primary “un-wired” technology with Mobile spectrum providing service when we are “in the gap” between Wifi connections.

Wifi investment would be climbing rapidly now except for one thing - much of the public Wifi infrastructure is owned by, yes you guessed it, companies like AT&T who are presently milking Mobile users for as much money as  they can. It’s the Mobile data users who are propping up their revenue stream as legacy technologies erode. They don’t have much incentive to wean users from expensive Mobile data plans hence they currently only favor Wifi in places like Times Square and Super Bowl venues where Mobile spectrum is utterly unable to deal with the demand.

But as I stated in the opening, the relative infrastructure costs will ultimately cause them to switch horses.

Facing data caps, consumers keep turning to Wi-Fi — Broadband News and Analysis.

If you haven’t dry roasted peanuts at home you should. Delish!

The Wii Trance

Happy Elephant

Blue Angle

Eyes on Daddy

Modern Thinker ?

Little brother doesn’t want to be left out

One proud daughter

The article from TechCrunch details the findings of a British developer who inadvertently discovered a great way to exploit expired domains to gain access to a person or company’s email accounts and other online services. This very simple technique, which requires no hacking at all, can easily yield access to Amazon accounts and banking services. Google and other companies really need to raise the bar to make it more difficult for people to claim email accounts and service accounts.

Security Breach: Here’s How Expired Domains Expose You To Embarrassment And Theft.

Fire!!! Nah, but junior liked seeing the truck

What’s keeping me going

These days she reads to me at bedtime:)

It is good to know there is a Bluetooth solution to the odd lack of GPS functionality on Wifi-only iPads. The device is a great size and the review is pretty positive. But at $99 list it would make more sense to buy a Wifi+3G iPad model instead. However if you bought the Wifi model not knowing GPS didn’t work I suppose this would be a slightly cheaper solution than  trading up to a 3G model.

Getting GPS on a WiFi iPad with New Sky.

Flowers for my wife - happy Mother’s Day!!!

Nice write up on CNET about the agencies and they tools they are likely using to gather up and crack the data on Bin Laden’s computers. If extreme care was taken by the terrorists this could be an extremely difficult job. But maintaining absolute security of data over long periods of time is extremely difficult. The human factor usually provides some easy doors for cracking even encrypted data so I suspect they can process the data pretty quickly.

Bin Laden’s computers will test U.S. forensics | Privacy Inc. - CNET News.

I needed to spend some time last night exploring the source to the library that comes with the SparkFun WiFly device. In the process I discovered why it wasn’t possible to use the library’s code to join a WiFi network if the SSID has a space in the name (even if you used a $ substitution). It is not quite the fault of the library code as the root cause is that one form of the join command on the chip won’t accept a space/$. The good news is by changing the code in the library slightly I could get the $ substituted SSID name to work just fine. I’ve posted details on the SparkFun forum.

My current annoyance is that the Arduino's official Ethernet library derives from Stream but the WiFly library derives from Print meaning third party libraries and code that work with Ethernet can't be used with the WiFly without quite a bit of work. I need to dive in further to understand if the WiFly library could be modified. I'm not sure why the library's author deviated in such a fundamental way from the standard library. 

Junior used it for first time today. All by himself. Happy days for mom and dad.

As I previously mentioned I got a WiFly shield from SparkFun in order to connect my Arduino to the Internet. It is a nifty little board but I immediately ran into problems getting it to join my wireless network. I’ll skip the fits and starts in my journey to diagnosis and troubleshoot a solution and instead jump to my conclusion.

But first some relevant bits about my WiFi network. I’m using WPA for security and my router publishes two SSIDs. One is the main network and another is a “guest” network with limited access and a different SSID. The main SSID has one space in the name while the “guest” SSID has two and is longer. In fact the “guest” SSID is 26 characters in length.

I initially tried to connect the WiFly to the “guest” SSID using example code included with the WiFly library. This failed with no error - in fact the example code hung at the point where it called “join”. After sending commands directly to the chip I found I could not join when using the “guest” SSID name. However I could join if I did a “scan” command and then sent a “join” command specifying the index number in the scan output. This was my main clue that the SSID name was the problem. Conclusion: either 26 characters for the SSID is too long or the chip cannot handle two spaces in the SSID name.

I next tried to join the main SSID using the example code and this too failed. However in this case I could successfully join this SSID by sending the commands to the chip and so I could conclude one space in the SSID was no problem (you do substitute a $ for the space). Reading the source code for the WiFly library I found a “TODO” comment to support spaces and hence I’m confident the library does not support spaces. I added code to set the SSID, bypassing the library’s code, and could then use the library to execute the join and access the internet successfully.

The code I used was a snippet from a post by Gary on this SparkFun Forum thread. If you have questions feel free to comment. In spite of the problems with the library, and potentially the chip having limitations on the SSID, I think this is a nifty little shield and I’m anxious to start some new projects involving connectivity.