Monday, December 28, 2015

A Luminary

Actually a few luminaries.  It has been a tradition in our family to have luminaries lining our driveway and entrance during Christmas and often also for new year.  A luminary is basically a  brown lunch bag with about two inches of sand and a candle.  It usually works quite well and is a nice touch to holiday decorations.  However bad weather can make this challenging and it sometimes is difficult to get the correct candle at a reasonable price. 

This year while teaching my STEM classes I had an idea of how to solve the candle problem.  The first exercise in my STEM class is for folks to make a throwie.  A throwie is basically a LED and a coin battery (usually a CR2032) taped together being mindful that the longer positive lead of the LED touches the positive side of the coin battery.  My STEM course manual has directions if you cannot figure it out from this description.  It provides a sense of accomplishment for the students and demystifies touching wiring.  The history of the throwie was that makers also attached magnets and they would throw it at metal structure to decorate them.  We usually place tack on the back and they place it on their id badges.  I wondered if such an arrangement would help with the candle problem in the luminaries.

The first  hurdle was that any LED would not do.  It would have to be large 10mm and would have to flicker like a candle.  I experimented with many a blinking, flickering LED and discovered the best candle like LED at http://evilmadscientist.com, a great site for LEDs and other geekie stuff.  I experimented with a few and selected a 10mm white flicker.  It was perfect.  I tested it one night with a paper lunch bag and sand.  I secured the LED to the battery with electrical tape and shoved the leads into the sand until only the bulb was showing.  It flickered nicely but simply was not bright enoungh.  Next night I used two CR2032 coin batteries in series, second battery positive (+) side was against first battery negative (-) side.  Fortunately this did not blow the LED but made it considerably brighter.  Please be careful when you first test yours - since LEDs may differ - wear adequate eye protection.   Here is a picture of them in our driveway.
Not a great picture but you get the idea.  Although they dim every night a bit, they are still glowing enough to see from the street after 4 days.

I am working on a bunch of new posts including benchmarks for the UDOO Neo and the CHIP computer. Hopefully I will do better in the new year.  The best to all of you in the new year!  Later!

Monday, June 8, 2015

We will be the Cylons

There was a fantastic recent issue of Nature that had several great articles on Artificial Intelligence.  I highly recommend the Insight articles which review most current hot issues in AI, specifically Deep learning, Reinforcement Learning and Evolutionary computation.

However the section  that gave me pause and was the inspiration for the title was a series of articles from researchers expressing their concerns on the risks of intelligent machines.  In particular, the article by Stuart Russell stands out.  He expresses concerns over Lethal Autonomous Weapons Systems, LAWS for short.  He sees them as being feasible in a matter of years, not decades.  Once I started thinking about the technologies necessary to make this happen, I agreed and feel that the results would be as accurate or as inaccurate as aerial missiles and bombs.  With advances in vision recognition and  local processing power, it is possible.  Robots without the human in the loop, explicitly ignoring Asimov's Laws of Robotics.  Russell provides some remedies, but nothing that would prevent rogue elements from employing them.  We must prepare for a world with these autonomous entities and consider controls equivalent to those of nuclear weapons.  Unfortunately, unlike nuclear weapons the technology necessary to construct them will be available to most groups.

So have I given our technology too much credit?  Please comment if you have any opinions and as always thanks for reading!  Sorry it has been so long between posts!  Later.

Friday, January 16, 2015

Bots Arrive

On Monday I received an Amazon Echo.  I am impressed.  "Alexa" is the keyword that gets the Echo's attention and it works near flawlessly.  You can barge-in, get Alexa's attention,  any time, even when the Echo is playing music.  The speech recognition works well in our large kitchen, even with  water running!  It is limited in its range of tasks at this time, but they are sufficient to make it worthwhile.  The music quality is fine, especially for an aging tin ear.  Although Siri does much better in search, I am really impressed by the Echo's speech recognition which is better than Siri's due in part, I am sure, to the 7 microphones.  I also am interested to see how Amazon uses the echo app in conjunction with the Echo device.  Currently the app logs your interactions with the Echo. Also  folks have hacked the API!  If you come across it in a friend's house, Alexa deserves a talking to!

  

My experience with Alexa has me thinking more about home bots.  Not all robots need to be mobile and my suspicion is that Alexa like entities will become more common in our homes earlier than Rosie like entities. It seems to be a natural transition to go from our smart phone based, social media and search engine interactions to an ambient bot that provides the same functionality by saying "Alexa."

I am still struggling with my copter but I did get software updates into it and learned a bunch about the arduino.  May have to set for a while as I crank up my iBeacon and other Bluetooth low energy devices. More on that in a future post.  Later!
 

Monday, January 5, 2015

Anyware

I just completed reading an article from O'Reilly by Mike Barlow, "When Hardware Meets Software."  It is short and free and does a great job of framing perhaps one of the next technological revolutions.  It makes at least three key points:  (1) hardware is beginning to achieve the same speedup in product realization that software has experienced using similar techniques, (2) hardware and software will be more equal partners in the future, and (3) software has  to raise the bar on quality, security, reliability, (name you favorite "ility"), internet quality of service is simply not good enough for the combination.

What does this say about the next generation of software engineers and especially hackers? They have to be as well versed in hardware too!  I think the hacker culture still retains it to a certain extent but not as completely as early Ham Radio enthusiasts or the Byte generation that had no choice but to pick up a soldering iron in order to do their software.

This next wave is a good thing.  Just as the rise of social media increased how connected we were to each other.  The rise of the Internet of Things/Industrial Internet/ Internet of Everything/Cyber Physical Systems will provide us with an appreciation of the  things around us, how they work and what they provide.  Great reading for my train ride to Hoboken!

I spent the weekend building some lego and trying to coax a mini-quadcopter to life.  Sadly it is not connecting to my iPhone so a new version of the software is in order. Double sadly I am struggling with earlier versions of Java related to earlier version of the arduino IDE. Hopefully progress this week and the next post will have a video of my success. Later!


Friday, January 2, 2015

Culture Mapping of Higher Education

We are all struggling to understand how we can convey the joy and excitement of technology to our children.  I have been involved at modest efforts focusing on grade school and high school students, some of which is chronicled at my aarphacker site.

Tim Stock and Marie Lena Tupot are focusing on using technology to understand what is available for students and using culture mapping and college syllabi to map and interrelate courses.  What a concept, to understand what is available and how it relates. They have a kickstarter to accelerate this mapping and I just stumbled across it with less than twenty hours to go.  So I am asking you if you read this on January 2nd would you please check out their kickstarter and hopefully back it and hopefully let others know.  Thanks!

I did not meet my resolution of last year for more posts but in 2015 I hope to average a post a week.  It has some probability of success since much of what I am doing in my new gig (Industry Professor at Stevens) will hopefully be of interest.  I am redoing my previous udoo benchmarks and am working on some items from kickstarter, so hopefully more soon.  The best to all of you in 2015, later!