Interesting development in Steve Ballmer’s departure–I wonder what happened inside the boardroom after the articles criticizing stack-ranking were published.

Microsoft axes its controversial employee-ranking system | The Verge

We live in the future! This video is almost unimaginable–it’s like something out of a gritty sci-fi movie. It gets me fired up when new technology is not just about having another gadget, another line item on the wish list.

Here it is [warning: not for the squeamish!]:

Two thoughts: 1) I pray that I can contribute to a technology some day that will help people or save lives like this. 2) I’m glad I didn’t choose to become a doctor (and you should be too). That made me really queasy.

[Hat tip to my awesome wife Robin for sending this to me!]

My advisor, Dr. Eric Johnson, provided a little bit of insight recently into software development. We work on software a lot, both in our simulator and in our autopilot. Unfortunately, I have to paraphrase here because I couldn’t write down his exact words fast enough, but he said something like this:

Let’s say you have an idea, and it takes X amount of time to formulate the concept in your mind.

Then it will take 10 times that long to write out the idea (the equations, algorithm, etc.) on paper.

Then it will take 10 times longer to convert that algorithm into usable code that it took to write it down.

Then it will take 10 times longer to test and debug the code than it took to write in the first place.

Putting that into some nice round numbers, and assuming 8 hour work days, if it takes you 5 minutes to come up with an idea:

• Then it will take you the rest of the hour to put it down on paper.
• Then you’ll spend the whole day tomorrow writing the code to implement it.
• Then plan on spending the next two weeks to debug and test the code to actually do what you wanted it to.

What’s the moral of the story? Focus on simple ideas and concepts, well-executed.

I finally got a chance to sit down and read through the whole UAS Roadmap, and wanted to share some nuggets I found and general impressions:

• The FAA looks at this process in three phases: Accommodation (“we’ll make exceptions to allow some to fly”), Integration (“UAVs in the airspace are normal”), and Evolution (“the NAS and UAS technology will grow together”).
• Publishing this roadmap was required by law, and I kind of get the impression that this was published just because they had to.
• The FAA will look to ICAO (International Civil Aviation Organization, a branch of the UN) to select UAS policies which are at least compatible with others in the world.  Though we aren’t and shouldn’t be beholden to the ICAO for making decisions that are right for the US, we also don’t need to re-invent the wheel here–I think this is a good thing.
• The FAA is acknowledging privacy concerns and is submitting that their test sites will help explore the privacy issues, but make clear that their first interest is in safety of operation.
• There is a focus on certifying whole systems–not just the airplane, but operator and flight crew, ground control station, and datalink as well.
• They acknowledge the regulatory difficulties coming from basic airplane assumptions being broken by UAVs.  The example they give is cockpit security regulations–do they not apply at all? Do they apply the operator in an office building?
• They also touch on the air traffic control interface with UAS and their operators–I think this is going to be a massive challenge, given that the ATC system is itself trying to transition to the NextGen system, and I don’t think UAS were considered in the initial designs of NextGen.
• They have some interesting assumptions about how UAS will operate once integrated into the NAS: –all UAS except sUAS in line-of-sight have to file IFR flight plans; –1 Pilot can control 1 UAS (which kind of ruins the efficiency you get by automating, right?); –Autonomous operations not permitted (what is autonomy, exactly? It’s kind of a continuous spectrum.)
• The FAA is creating a separate classification for small UAS (sUAS in their acronym-space), with a different rule-set which should be coming out earlier than rules for the bigger ones.  They’re defining sUAS as unmanned aircraft under 55 lbs.
• However, there’s a little hiccup here: the Notice of Proposed Rulemaking, or NPRM, is supposed to come out in early 2014 according to section C.6 Goal 1A.  But in Goal 1D, it says the NPRM is supposed to come out in 2013. Hmmm. . .
• The roadmap establishes a ton of time targets (most quarterly targets) for studies and conferences and draft proposals of rules–especially for use of sUAS by law enforcement and first responders–but what I got out of the whole thing is that the final rule allowing sUAS operations is targeted to published in 4th Quarter 2015 (I assume calendar year.) sUAS pilot certification standards are supposed to come out in 2014.

Ok, so I don’t think these will be available for purchase at Christmas time. But if we have the scratch when they do come out, I think I will be buying these for my kids. IEEE Spectrum reports:

These colorful robots are not only fun to play with—they can teach kids computer programming skills. That’s what Play-i, a Silicon Valley startup founded by engineers from Google, Apple, and Symantec, says about its robots, unveiled this week as part of a crowdsourcing campaign.

The idea of using robots to teach kids programming, math concepts, and problem solving is not new. In fact, it’s been more than 40 years since MIT educator Seymour Papert demonstrated the possibilities of hands-on learning with his Logo programming language and mobile machines known as “turtle robots.”

Over the years, numerous robotic toys and kits designed for kids came to market—most famously the LEGO Mindstorms set… But most of these products present a steep learning curve for kids (and parents!), and few are adequate for very young children.

Play-i, based in Mountain View, Calif., wants to see that change. It says its robots, Bo and Yana, can make programming fun and accessible for kids as young as 5 years old. The robots talk via Bluetooth LE with an iPad or other tablets, which Play-i says are the perfect interface for children to learn programming concepts in an engaging, intuitive way.

I’ve pointed out here, and here, and here how important programming skills are going to be for today’s kids and tomorrow’s workforce. It’s exciting and encouraging to see the ways to learn them continue to develop.

These Robots Will Teach Kids Programming Skills | IEEE Spectrum

The field of applications of small, cheap, easy to deploy UAVs continues to grow. Now ecologists are using them to collect “fly-on-the-wall” data that would have been nearly impossible or prohibitively expensive (or both) to gather. Previously, the available alternative would have been to use a manned helicopter, which is more dangerous, less responsive, could interfere with the subject of data collection, etc. etc.

Plan on starting a quadcopter company. When the US finally turns the key on small commercial UAV usage, demand is going to explode. I’m just excited to find out what applications are out there which no one has yet thought about . . .

ArduCopters mapping forest treetops in NSF-funded project | DIY Drones

The FAA posted its 2013 UAS Integration Roadmap to its website today. I have not read it yet, but here it is if anyone would like to take a look.

I’ll probably read it this weekend and offer some thoughts early next week.

According to IEEE Spectrum, University of Delaware researchers have discovered that staggering rows of windmills could increase the overall power output of wind farms.

Whether wind farms are on or off shore, there is a tradeoff between getting maximum energy from each individual turbine and packing a greater number of turbines into the space. As each turbine pulls energy from the wind, there is less energy in the downstream wind, which causes array losses for the entire wind farm.
…
But there may be a simple solution to boost energy production without expanding the boundaries of offshore wind farms. Researchers at the University of Delaware found that staggering turbines in the ocean can improve annual power capacity by 13 to 33 percent.

Ok, not to be too flippant, but . . . isn’t that kind of obvious? If the downstream rotor wake interferes with a downstream turbine, why would you put one right in front of the other?

Offshore Wind Farms Just Need a Little Stagger to Generate More Power | IEEE Spectrum

sUAS news reported on the purchase of a bunch of Prox Dynamics PD-100 derivatives by the US Army last week. I had the good fortune to meet and get an informal demonstration of one of these units from one of their salesmen a few weeks ago, and saw the reports of the UK Army using them in Afghanistan last year, and wished I had joined the Army fifteen years later. These units are amazing–the total size and weight of the control unit and two aircraft is roughly equivalent to a full 1-quart canteen, and it gives remote surveillance capability to squad or maybe even fire team size units.

Many kudos to Natick for equipping Soldiers with truly top tech.