This post explores how to run great experiments in your company, based on recent conversations with my friend Astro Teller, Chief of Moonshots at “X” (formally Google X, Google’s R&D factory).

X’s mission is to invent and launch “moonshot” technologies that could make the world a radically better place…dare I say, help create a world of Abundance.

Astro leads a team of brilliant engineers, scientists and creatives developing solutions to dozens (perhaps hundreds) of the world’s toughest problems. Some of their publicly known projects include: the self-driving cars, the smart contact lens, high-altitude wind-power generation, and Project Loon, just to name a small fraction.

All of these projects started as a series of experiments.

Today’s most successful companies, the ones that are “crushing it,” started as a series of crazy ideas, followed by experiments to test just how viable those ideas might be.

Experimentation is a crucial mechanism for driving breakthroughs in any organization.

By the way, you may also want to check out Astro’s 2016 TED talk which was just released.

What Are Experiments and Why Do They Matter?

Experiments help you test a hypothesis about your product or service and help you find answers to your most difficult questions.

Good questions are questions that, if answered, fundamentally change (and improve) the way you operate.

Understanding how to focus your team on asking good questions and turning good questions into experiments is critical.

So what makes a good experiment?

The Three Principles of a Good Experiment

Astro explains that the following three principles describe a good experiment:

Principle 1: Any experiment where you already know the outcome is a BAD experiment.

Principle 2: Any experiment when the outcome will not change what you are doing is also a BAD experiment.

Principle 3: Everything else (especially where the input and output is quantifiable (i.e., measurable)) is a GOOD experiment.

Seems simple enough, right?

You need to be asking questions to which you don’t know the answer but such that if you did know the answer, you’d change the way you operate.

If you already know the answer, or if you are testing an insignificant detail that doesn’t matter, you’ll just be wasting time and money.

How to Ask Good Questions and Design Good Experiments

In any organization, you get what you incentivize.

In order to get good questions/experiments, you have to create a culture that incentivizes asking good questions and designing good experiments.

Astro describes a very unique approach to doing just this:

“At X, we set up a ‘Get Weirder Award.’ The whole point of the Get Weirder Award was to focus the team on experiments and to drive home why they needed to think in terms of experiments.”

Teams would be challenged to ask “weird” questions — to put forth crazy ideas around framing problems differently and to design experiments that really push the limits.

Critically, Astro gives out the Get Weirder Award before the experiments are run.

“If you give out the award before you’ve run the experiment, then people start to really feel that you don’t actually care about the outcome. You care about the quality of the question. So every two weeks, we would give out an award for the best experiment.”

Doing so constantly (and viscerally) reinforced the behavior of asking good questions — and as such, at X, they’ve built a culture around celebrating the questions themselves.

How to Manage Experiments

Once you’ve designed a good experiment and assembled an intellectually diverse team to tackle it, what are the best management principles to keep from screwing it up?

Management Principle 1: Don’t Be a Bottleneck!

As a CEO or manager, it is critically important that you don’t get in your team’s way by micromanaging them or by demanding to be the sole decision maker.

If you need all of the information, all of the time, your team will never get their work done.

Astro explains, “Your job as a manager is to give your team your recommendations and empower them to do whatever they think is right.  Allow them to learn.”

He continues, “I work super hard for me not to be the bottleneck at anything that goes on here. Ironically, that’s a full-time job.”

One funny story involved two employees who had a major strategic conflict.  They wanted Astro to make a decision as to which one was right.

Rather than do so, he said, “I believe I already know which of you is right and which of you is wrong. I can just make that decision right now, but I’m not going to. The problem is: If I tell the two of you who’s right and who’s wrong, in my opinion, the next time you have conflict, you will come back and ask me to do it again, and that does not scale. I will spend however as much time it takes to either train the two of you to work well together or figure out that you can’t.”

Not being a bottleneck means deliberately letting your team learn. Sometimes it’s hard to do, but it’s a necessary step if you want to derive the most value from your experiments.

Management Principle 2: The Value of Secrecy

In a previous blog, I mentioned the notion that ambitious entrepreneurs (and companies) need to get comfortable with being misunderstood.

Interestingly, experiments are one of the most publicly misunderstood domains within a company.

We talked in great depth in the last blog about why you have to focus on killing your ideas — in line with that discussion, an experiment that proves an idea won’t work is as successful of an experiment as proving that one will work!

The problem is: the public (and especially the press) doesn’t understand this.

They see “failed experiments” and think, “failed company” — which adds enormous pressure and stress to your employees and investors.

Thus, it can be useful to keep your experiments secret.

Astro explains, “The main value of secrecy is not ‘to hide our awesome ideas’, it’s to make it easier to kill ideas. Secrecy offers air cover. It allows the team to have the emotional space to solve the problems without outside pressure.”

Management Principle 3: Don’t Worry about Success — Worry about Progress & Learning

Finally, if you are worrying about success, you are going to be paying attention to all of the wrong indicators and misguiding your team.

You need to focus on progress and learning — and success will follow.

Rather than creating a culture that only celebrates big wins, create one that celebrates progress on tough projects and running good experiments.

You’ll be amazed by the difference in your organization with such a small mindset and cultural shift.

Image credit:

As much as the headlines proclaim “virtual reality is here,” it has also only just arrived.

Samsung Gear VR was released in November; this month is the initial launch of the HTC Vive and Oculus Rift; and Playstation VR is just a couple months out.

The initial release has not been without its hiccups, but VR is most certainly here to stay.

In the coming years, big moves from the likes of Google, Facebook, and possibly even Apple will cement the fate of VR for years to come. While we have the best, most affordable consumer VR technology ever today, there are undoubtedly opportunities that have yet to be fully actualized and will be even greater than what we’ve seen up to this point.

In February, as a part of our 2016 Virtual Reality Industry Report, Greenlight VR conducted an online survey with VR developers and studios to understand their own expectations for the industry and their opportunities within it.

Here are three of the many takeaways from our research.

Multiple Use Cases and Verticals Are Being Developed Beyond Gaming

As expected, gaming dominates the sample’s current focus, but there is a surprising diversity of applications, content, and experiences being developed. These reported professional and enterprise verticals ranged from education, travel, corporate training, healthcare, and even financial services. If this data proves to reflect the industry at large, we should expect to see a significantly diverse landscape in the coming years.

Use Case Table for SUArticle (1)

This is promising for an industry that is currently characterized as gaming-centric. Gamers will pave the way, but we think the larger market opportunity lies in the summation of multiple, highly-specialized use cases for which VR offers a compelling proposition.

Business Model Experimentation Is Underway

Although profitability is expected, a near majority of the sample is still figuring out key aspects of a revenue model. For instance, 41% report they plan a mixed revenue model, which could include direct-consumer pricing or sponsorship support.


We expect developers will respond to what consumers are most willing to buy as the market matures and a sizable consumer base forms. For gaming, this could look like micro-transactions and standalone purchases. For non-gaming, it could resemble subscription services, pay-per-view, or a contract model.

Developers Have High Hopes for 2016

Based on our sample*, independent developers overwhelmingly believe (76%) that their companies will make a profit in 2016. Considering how lean some early market leaders can be, it is not inconceivable that some will find sustainable profits. However, we believe the data reflect a community that is still aspirational rather than indicative of the overall market in 2016.

greenlight-developers-survey-2As we’ve discussed with industry leaders over the years, we’ve noticed this aspiration can prove to be well founded, but with the caveat that VR will require many more years, iterations, and development breakthroughs before HMDs are as ubiquitous as smartphones.

We believe the latter to be true, but the expectations of profits to be optimistic. 2016 could quite possibly be the year VR has to survive rather than the tipping point.

Although Robust Growth Is Expected, the Ecosystem Is Still Fragile

As we continue to monitor and analyze the industry, the one thing that appears certain today is that new market opportunity is still under development.

The rapidly growing community of developers is experimenting. However, they are not alone.

We are all learning, researching, and creating in an effort to advance the medium together toward a sustainable future. Although this nascent industry is burgeoning, there is still quite a bit of ground to cover before mainstream appeal truly sets in. Most certainly we will see both dramatic leaps and stumbles as the industry and related technologies develop.

*Research conducted in February, 2016 utilizing a respondent base of executives solicited directly by Greenlight VR from opt-in lists of companies already in correspondence. All respondents reported they were involved in a decision making capacity at their respective companies (a requirement for survey participation). Of the 90 initiated surveys, 52 were completed from self-identified “independent” developers. These findings present the opinions of 52 “developers.”

Due to the small sample size of the survey, the findings are not statistically significant and should not be taken as representative of the independent developer community. However, based on ongoing interaction with the developer community (including industry conferences, analyst briefings, consulting) we believe these findings are likely to be representative.          

Banner image credit:

Maybe the Belgians known something we don’t. The country has just decided to give everyone iodine pills for the event of nuclear catastrophe.

Read more…

There’s a small British app that allows people to take pictures of trash on the street then report it to local authorities so that they can do something about it. But Instagram thinks that its name—Littergram—is garbage.

Read more…

Después de muchos retrasos y un aparente sinfín de denuncias por corrupción y malversación de fondos
, Rusia lo ha logrado. El cosmódromo Vostochny ha lanzado su primer cohete, poniendo en órbita tres satélites, aunque no sin otro retraso de última hora que hizo esperar al presidente Vladimir Putin.

Read more…

Instragram sigue en fase beta en su versión para Windows Phone y apenas se ha actualizado desde que llegó a la tienda de aplicaciones de Microsoft en 2013. Hasta ahora los usuarios de Lumia tenían que recurrir a aplicaciones de terceros como 6Tag para subir vídeo a su perfil, pero con esta nueva actualización el vídeo ya es oficial.

Read more…

On April 12th, 1961 Yuri Gagarin launched into space on a Vostok rocket from the Baikonur Cosmodrome, becoming the first person ever to leave the planet.

Here’s the crazy thing: today’s astronauts travel to space on a nearly identical rocket, the Soyuz, which went into operation only five years after Gagarin’s historic flight.

Why? While we’ve seen innovations in rockets over the last half-century, the underlying physics of how we get people and things into space has not changed. Chemical propulsion technology is still at the heart of all rockets. It’s what makes rocket science, rocket science.

Getting to space is hard. Just 2% of a rocket is the thing we want in space, the other 98% is the “rocket-stuff” that accelerates the payload and gets it out of Earth’s powerful gravity well.

For a long time, the paradigm of launching big expensive rockets worked well enough because very few things actually needed to get launched into space, and the things that we launched were big and expensive themselves. So the space industry could afford the long lead cycles and absorb the high barriers to entry.

But all that has begun to change over the past several years as new satellite capabilities have emerged and new players have entered the space arena. The industry now has startup companies in Silicon Valley launching more satellites to space than traditional satellite markets.

For the history of space exploration, we have been trying to make rockets bigger, better, faster, cheaper, and reusable to beat the supply chain problem. There is another way to approach this problem, however. What if we didn’t need to launch anything at all?

Manufacturing satellites is a huge business. Annually, it’s a $15 billion market. Currently, the only option for getting these expensive, complex systems into orbit is to launch them on a rocket.

Made In Space, Inc. is approaching this market in a different way that doesn’t involve the traditional hurdles of the current supply chain to space. Over the past six years, the team has been putting together the technological building blocks to make manufacturing of satellites in space possible.

AMF-halo (1)

Made in Space’s Additive Manufacturing Facility (AMF), currently on the ISS.

Today, Made In Space is taking the first steps towards manufacturing satellites in space by using the company’s 3D printer on the International Space Station to make small customized satellites (called CubeSats) on-demand.

Through a program called  Stash and Deploy” a stockpile of satellite components that cannot yet be 3D printed will be stored on the ISS.

As a satellite is needed, specific components will be assembled with 3D printed structural components. The end product? A custom on-demand satellite assembled in space and deployed into orbit.

XManSummit_750x100Banner_slower_discount (1)

A future stage of stash and deploy will be the robotic manufacturing and assembly of entire satellite systems in space. Recently, NASA funded Made In Space’s Archinaut Program which will begin the development of a set of capabilities to enable large-scale spacecraft, satellites, and space assets to be manufactured in space.

Making satellites in space unlocks fundamentally new design possibilities. Today’s satellites are designed for launch, but when they are manufactured in space, we will be able to let the conditions of space alone drive the design of our satellites.

What Could We Do If We Didn’t Need to Launch?

Archinaut-progression (1)

Archinaut concept design: building large scale satellites in space.

Imagine for a moment a large space structure, like an antenna, that will never see any environment other than the microgravity of space—it could be incredibly sparse and wispy.

Now, imagine a satellite that never had to be constrained within the cylindrical diameter of a rocket body or survive the loads of launch—it could be perfectly optimized for its mission, maybe being as thin as a sheet of paper to absorb and radiate solar radiation more evenly.

Satellites and space structures like these have never existed because they couldn’t be launched. More importantly, today’s space assets can be drastically improved when the satellite is manufactured in space.

Remote sensing satellites can be made over a hundred meters long to enable synthetic aperture radar that collects new information on our changing planet. Satellites depending on reliable communications, such as GPS and telecommunications, can be made more capable with larger, more customized antennas. A large space-based phased array antenna platform can address new communication markets, such as connecting the world to global broadband from space.

Just as we will build large antennas for satellite communications, we can also build very large space telescopes for observing the universe. Radio telescopes with apertures larger than the Arecibo Observatory in Puerto Rico could be positioned in space for more optimal observations.

In addition, satellites will one day be made far larger than today’s satellites. There will be unique applications for satellites that are miles in length. The traditional term for such objects is megastructures—but none have existed in space yet. That said, space is the optimal place for megastructures. It’s easy to support the weight of a mile long satellite when it is weightless!

Much of a satellite’s mass goes to its structure, allowing it to be built and tested on Earth and launched on a rocket. Soon, when we only have to launch the raw materials to build satellites in space, that mass will go much further because the structural components of the final satellite will never need to experience gravity or the large force loads of rocket launch. And one day, we may start using the resources of space for construction and thus removing the need to launch anything from Earth at all.

Space is a wonderful place for industrial activity for other reasons too.

Unfiltered solar energy can drive even the most power hungry manufacturing methods. Weightlessness allows huge objects to be manipulated with the bump of an actuator. We even have unlimited, clean vacuum for manufacturing without oxidation or corrosion.

And beyond specific applications, the whole process can be more efficient—3D printing the satellite components themselves in space enables a faster, more responsive ability to deploy a new satellite.

Why All of This Is Important

We may not always recognize it, but satellites provide value to us everyday. Our “birds” in the sky do everything from remote sensing of weather patterns and crop yields, to GPS and telecommunications connecting the people of the world. Satellite-based telescopes, like Hubble and Kepler, are even helping us understand some of the age-old questions of the universe.

In every one of the examples described above, the satellite and the satellite systems that are manufactured in space are far more optimized for their operating environment than any satellite currently in space today.

We have come a long way since Gagarin’s first spaceflight thanks to our daring efforts to reach further and further into the cosmos. Looking back on human history, we recognize key moments of creation: The creation of fire, the first tools, industrial processes, the internet.

We have now entered a moment in our shared timeline in which humanity can create things not just on Earth, but off Earth as well. Making satellites in space will change the way all satellites are operated in the future. But if we are truly successful, satellites will be just the first in a long line of new products and industry made in space.

Interested in learning more about the future of manufacturing? Join Jason Dunn and other leading manufacturing experts at Singularity University’s inaugural Exponential Manufacturing Conference May 10-11, 2016 in Boston.

Image credit:; Made In Space

En Captain America: Civil War
llegará al universo cinematográfico de Marvel uno de los superhéroes más queridos de todos los tiempos: Spider-Man
. Esta vez el actor que hará de “arácnido” es Tom Holland, y el joven británico asegura que hay una gran diferencia entre su versión del héroe y las anteriores: su inocencia.

Read more…

La oficina del sheriff del condado de Santa Clara ha confirmado que el cuerpo fue hallado sin vida junto a un arma y que no hay otras personas involucradas. Podría tratarse de un suicidio por un disparo en la cabeza.

Read more…

The recent announcement that a young paralysed man in Ohio in the US named Ian Burkhart managed to regain the use of his fingers after having a chip implanted in his brain is an exciting step forward for science and healthcare. In fact, you may now be wondering how long it will be before we can unlock a door, turn on a kettle, or even send an email simply by thinking about it?

The Ohio breakthrough used a technique called functional magnetic resonance imaging (fMRI) to identify the pattern of electrical impulses in the part of the brain that controls movement — the motor cortex — that was generated when Burkhart thought about using his fingers. The system learned to recognize this pattern when it appeared in his brain, and then instruct receivers to stimulate his arm muscles to make the appropriate movements.

This might produce life-changing results for people with disabilities, but it has limited potential outside the body. It is about sending movement instructions to parts of the body that cannot be reached in the usual way. We might be able to use it to make a robot reproduce our movements, but that may be the limit.

Having said that, we have already discovered ways of manipulating foreign objects. Three years ago, I demonstrated a modified Scalextric set at the Lancashire Science Festival that enabled people to make the cars go faster round the track simply by concentrating harder on them. Hundreds of people were able to try this using a Bluetooth headset called the Neurosky Mindwave, connected to nothing more than a laptop and a simple microcontroller.

Brain-powered Grand Prix. Image credit: Stephen Sigurnjak

The technology behind this shift from the telekinesis of sci-fi movies or comic books into the real world is electroencephalography, or EEG. This monitors the brain’s electrical activity using electrodes placed on the scalp. The data is then processed to see the underlying frequencies in these impulses, which are associated with different kinds of brain activities. The alpha frequency band is associated with wakeful relaxation with closed eyes, for example, while the beta frequency is associated with normal waking consciousness.

The headsets in my demonstration transmitted this information to the laptop, which used algorithms that recognize concentration as a combination of different impulses: increasing on several frequencies while falling on several other frequencies at the same time. When it detected this, it instructed the microcontroller to increase the amount of power going to the Scalextric. There is an art to making the system work well: sometimes people find the cars going faster even though they didn’t think they were concentrating. I found I made the cars go faster by doing the alphabet in my head; and could slow them down by looking at a blank wall. Everyone is a bit different.

There are now commercial toys available that use the same technology. One example is the Star Wars Force Trainer, where EEG — not Jedi power — enables users to elevate a ping-pong ball using just their mind.


There are also serious potential applications. To try to make computer programs easier to use, for example, researchers have studied EEGs to discern the amount of cognitive effort someone expends on different elements of a program. I have studied the brain activity of experienced archers and found a difference in impulses between “good” and “bad” shots. This might enable coaches to tell players when they are in the right mindset, while players might be able to train their minds to achieve better results.

The Trouble With Thoughts

These are promising developments, but they are all looking more at the “global” activity of the brain rather than someone’s thoughts. There is a very big difference. For instance, researchers have built an EEG-powered electric wheelchair, but it runs into problems when a hazard appears. The user is prone to start concentrating on the hazard, and because the system can’t tell one kind of concentration from another, the wheelchair keeps moving and the person could end up in danger. To get around this problem, researchers added a secondary control system that allows the user to touch a pad to allow the wheelchair to move and touch it again to disable it — with moderately successful results.


The brain is a very complex organ with multiple areas responsible for many different kinds of activity. It is a major challenge to unpick everything and isolate “thoughts” from the data. The limit of current technology is to attach numerous electrodes to the scalp and measure the activity in different areas of the brain at the same time. Because different areas govern different actions, this makes it possible to use algorithms to detect whether a person is thinking of, say, moving their left or right arm. This might allow a slightly more sophisticated means of brain-powered wheelchair control, for instance. But while this is starting to get closer to thought control, it is still rather global and has to be tailored to the individual subject since the exact patterns of brain activity vary from person to person.

In future, we may well gain a greater understanding of the structure and function of the brain. Together with more sensitive electrodes and more computer processing power, this might make it possible to further develop this brain-to-computer interface into a more accurate system that can adjust to the variations between one person and the next. This might make it easier for someone who would otherwise be paralysed to control a device or communicate.

Even then, that would still be quite a way from true “thought” control. It is already possible to switch on a kettle through concentration using EEG technology, but we are still some way from being able to think a range of different instructions to different objects attached to a single system. As for sending emails, it looks like we will be typing for some time to come.

Stephen Sigurnjak, Senior Lecturer in Electronics, University of Central Lancashire

This article was originally published on The Conversation. Read the original article.

Banner image credit:

« Previous posts Back to top