O'Reilly Radar Podcast - O'Reilly Media Podcast

Latest Episodes:

Eric McNulty on real-time disaster response and leadership beyond control Jan 28, 2016

Show notes

The O'Reilly Radar Podcast: FEMA's Innovation Team and practicing leadership as if it's an Olympic sport.

O'Reilly's Jenn Webb chats with Eric McNulty, a consultant, writer, speaker, and catalyst for positive leadership. McNulty talks about real-time disaster response, the connections between disaster response and organizational leadership, and how today's leaders can achieve order beyond control and influence beyond authority. McNulty will talk more about instituting effective leadership at the Cultivate leadership training at Strata + Hadoop World in San Jose in March.

Here are a few highlights:

Right after Hurricane Sandy, I was here in New York and New Jersey, and FEMA deployed their first ever Innovation Team, which meant they were trying to innovate in the midst of disaster response. ... They were coming together and building mesh networks in some cases. They were crowdsourcing evaluation of photographs. They would get the Civilian Air Patrol to do overviews of the effected areas. They'd upload that and people anywhere in the country could look at it and give it a basic evaluation of severe, moderate, or mild in terms of the damage. They were able to get situational awareness very quickly the way they never would have been able to otherwise. ... They were able to innovate in real time in the field and see what worked and what didn't; the then deputy administrator, who is now a colleague of mine at Harvard, said it fundamentally changed the way FEMA operates.

What we have found over time is that being on the ground watching leaders in high stakes, high-pressure situations like the Boston Marathon Bombing response, like Deep Water Horizon, like Hurricane Katrina or Sandy, it's like the Olympics of leadership. It's when things are absolutely at their toughest and people may die if you make the wrong decision, which is not what most of us face in our organizational lives. Like the Olympics, you don't start practicing the day you get there. You practice for a long time before you get there. The fundamental skills are ones you can use every single day. You can use them at work, you can use them at home. There are things that we teach, some people, they're saying go home and use it with your kids. ... There's a lot of applications for these things beyond a 'leadership setting.'

Right now we're doing a lot of work around cyber. The cyber threats are growing and they're going from disruptive to destructive. There's an increased need for public and private entities to work well together. ... Fundamentally, it's a leadership problem. Finding a way for different parties who have a shared interest that don't want bad things to happen to come together and be able to collaborate and work well together.

There's a fundamental shift from leadership being thought of as a senior management function to an independent function. What I mean by that is leadership is really about human factors. ... When you're leading, how do you maintain the clarity of your voice, the clarity of your vision, and motivate people to inspire them to get them to where you need to be amidst all the craziness? ... How are you thinking beyond control? [Maintaining] order beyond control and influence beyond authority? Which is a different way of functioning for a lot of people.

Mark Burgess on a CS narrative, orders of magnitude, and approaching biological scale Jan 14, 2016

Show notes

The O'Reilly Radar Podcast: "In Search of Certainty," Promise Theory, and scaling the computational net.

Aneel Lakhani, director of marketing at SignalFx, chats with Mark Burgess, professor emeritus of network and system administration, former founder and CTO of CFEngine, and now an independent technologist and researcher. They talk about the new edition of Burgess' book, In Search of Certainty, Promise Theory and how promises are a kind of service model, and ways of applying promise-oriented thinking to networks.

Here are a few highlights from their chat:

We tend to separate our narrative about computer science from the narrative of physics and biology and these other sciences. Many of the ideas of course, all of the ideas, that computers are based on originate in these other sciences. I felt it was important to weave computer science into that historical narrative and write the kind of book that I loved to read when I was a teenager, a popular science book explaining ideas, and popularizing some of those ideas, and weaving a story around it to hopefully create a wider understanding.

I think one of the things that struck me as I was writing [In Search of Certainty], is it all goes back to scales. This is a very physicist point of view. When you measure the world, when you observe the world, when you characterize it even, you need a sense of something to measure it by. ... I started the book explaining how scales affect the way we describe systems in physics. By scale, I mean the order of magnitude. ... The descriptions of systems are often qualitatively different with these different scales. ... Part of my work over the years has been trying to find out how we could invent the measuring scale for semantics. This is how so-called Promise Theory came about. I think this notion of scale and how we apply it to systems is hugely important.

You're always trying to find the balance between the forces of destruction and the forces of repair.

There are two ways you can repair a system. One is that you can just wait until it fails and then repair it very fast, and try to maintain an equilibrium like that. We do that when we break a leg or when we do large-scale things. There's another way that biology does it, and that is to simply have an abundance of resources and let some things just die. Kill them off and replace them. The disposable cell version of biology, which is, if you've got enough containers, enough redundant cells, it doesn't matter if you scrape a few off. There's plenty more. If you scratch yourself, you don't bleed usually. You have enough skin left over to do the job. That's the thing that we're seeing now. Back in the 90s, it wasn't very plausible, because we had hundreds of machines and killing a few of them was still a significant impact. Now, when it's tens of thousands, hundreds of thousands, millions of computers, we really are starting to approach biological scales.

As these, what today are toys, become actually integrated parts of our lifestyles and technologies—maybe the new homes are built with things with things all over the shop and industrial-strength controllers to manage them. Once that happens, the challenges of managing them and keeping them stable, and keeping them under our control, become paramount. It's a different order of magnitude, again, than we're used to today. This idea of centralized data centers is going to have to break up. We're going to need Cloud substations. In the same way we scale the electrical net, we're going to need to scale the computational net, and storage as well.

Subscribe to the O'Reilly Radar Podcast: Stitcher, TuneIn, iTunes, SoundCloud, RSS

Charles Fracchia on a new breed of biologists Jan 12, 2016

Show notes

The O’Reilly Hardware Podcast: The merging worlds of software, hardware, and biology.

In this new episode of the Hardware Podcast—which features our first discussion focusing specifically on synthetic biology—David Cranor and I talk with Charles Fracchia, an IBM Fellow at the MIT Media Lab and founder of the synthetic biology company BioBright.

Discussion points:

The blurring of the lines between biology, software development, hardware engineering, and electrical engineering
BioBright’s efforts to create hardware and software tools to reinvent the way biology is done in a lab
The most prominent market forces in biology today (especially healthcare)
How experiments conducted using Arduino or Raspberry Pi devices are impacting synthetic biology
Pembient’s synthetic rhino horns

This week’s click spirals

Studies on the effects of oxytocin
Dafen, a Chinese village where copies of artwork are mass-produced
Gorgeous Swiss- and Japanese-made calipers. See a brand comparison here. If you’re looking for a good way to pass a long flight, download the PDF versions of the Mitutoyo, TESA/Brown & Sharpe, and Starrett catalogs.

Katie Dill on heading up experience design at Airbnb Dec 30, 2015

Show notes

The O'Reilly Radar Podcast: A triforce company structure, the power of storyboards, and designing business strategy.

O'Reilly's Mary Treseler chats with Airbnb's head of experience design Katie Dill about the values that drive design at Airbnb, the triforce structure of the company, and the process of journey mapping their users' experience.

Here are a few snippets from their conversation:

That triforce of product management, engineering, and design, working together from point zero on the process of what problems we are trying to solve, and how we might solve that, and why we might solve it, and what the road map should be in getting there, is a process that is facilitated through design thinking. It's a process that includes all those voices in a way that we think gets us to some solutions that are a little bit more creative than we otherwise would have gotten to, but also thoughtfully considered in terms of the technology and the business impact.

We literally use that storyboard for everything. Right now, we're mapping out how we email—how and when we email people. We've got the storyboard up on the wall as a reminder of the steps in the journey, and then we're printing out every single email we send to people and putting it where it happens in the journey. What's so great is then you take a step back, and it's almost like a Monet. The impressionistic viewing, oh, my goodness, we're sending 12 emails to people when they're booking, but we're ignoring that opportunity later in the journey.

It comes down to, in some ways, a shift in thinking about when design is a part of the process in creating a product, and the shift is to think of design as a way of also helping to define business strategy.

Subscribe to the O'Reilly Radar Podcast: Stitcher, TuneIn, iTunes, SoundCloud, RSS

Patrick Wendell on Spark's roadmap, Spark R API, and deep learning on the horizon Dec 23, 2015

Show notes

The O'Reilly Radar Podcast: A special holiday cross-over of the O'Reilly Data Show Podcast.

O'Reilly's Ben Lorica chats with Apache Spark release manager and Databricks co-founder Patrick Wendell about Spark's roadmap and interesting applications he's seeing in the growing Spark ecosystem.

Here are some highlights from their chat:

We were really trying to solve research problems, so we were trying to work with the early users of Spark, getting feedback on what issues it had and what types of problems they were trying to solve with Spark, and then use that to influence the roadmap. It was definitely a more informal process, but from the very beginning, we were expressly user driven in the way we thought about building Spark, which is quite different than a lot of other open source projects. … From the beginning, we were focused on empowering other people and building platforms for other developers.

One of the early users was Conviva, a company that does analytics for real-time video distribution. They were a very early user of Spark, they continue to use it today, and a lot of their feedback was incorporated into our roadmap, especially around the types of APIs they wanted to have that would make data processing really simple for them, and of course, performance was a big issue for them very early on because in the business of optimizing real-time video streams, you want to be able to react really quickly when conditions change. ... Early on, things like latency and performance were pretty important.

In general in Spark, we are trying to make every release of Spark accessible to more users, and that means getting people super easy-to-use APIs—APIs in familiar languages like Python and APIs that are codable without a lot of effort. I remember when we started Spark, we were super excited because you can write a k-means cluster in like 10 lines of code; to do the same thing in Hadoop, you have to write 300 lines.

The next major API for Spark is this API called Spark R that was merged into master branch [early in 2015], and it's going to be present in the 1.4 release of Spark. This is what we saw as a very important part of embracing the data science community. R is already very popular and actually growing rather quickly in terms of popularity for statistical processing, and we wanted to give people a really nice first-class way of using R with Spark.

We have an exploration into [deep learning] going on, actually, by Reza Zadeh, who's a Stanford professor who's been working on Spark for a long time and works at Databricks as well, as a consultant. He's starting to look into it; I think the initial deliverable was just some support for standard neural nets, but deep learning is definitely on the horizon. That may be more of the Spark 1.6 time frame, but we are definitely deciding which subsets of functionality we can support nicely inside of Spark, and we've heard a very clear user demand for that.

Subscribe to the O'Reilly Radar Podcast: Stitcher, TuneIn, iTunes, SoundCloud, RSS

Leah Busque and Dan Teran on the future of work Dec 17, 2015

Show notes

The O'Reilly Radar Podcast: Service networking, employees vs contractors, and turning the world into a luxury hotel.

O'Reilly's Mac Slocum delves into the economy with two speakers from our recent Next:Economy conference. First, Slocum talks with Leah Busque, founder of TaskRabbit, about service networking, TaskRabbit's goals, and issues facing the peer economy. In the second segment, Slocum talks with Dan Teran, co-founder of Managed by Q, about the on-demand economy and the future of work.

Here are a few highlights from Busque:

As a technologist myself, I became really passionate about how we mash up social and location technologies to connect real people, in the real world, to get real things done. I'd say in the last two years, it's become real time, and that's really the idea about where service networking was born.

It's certainly our job to create a platform where demand is generated so that our tasker community, our suppliers, can find work, but I think even more than that, it is about building a platform and tools for our taskers to build out their own businesses.

I remember in 2008 when the iPhone just came out, people thought it would be crazy to jump in a strangers car and take a ride with someone. People thought it would be crazy to have a neighbor or handyman come into their house and hang shelves, so in the early days, there was a big trust barrier to entry. Now, as the consumer mindset as evolved and changed over the course of the last five, seven, eight years, trust has been able to be bridged utilizing technology and creating trust between users is sort of a challenge that's, of course always going to be and is still there, but it's not the main challenge anymore. I'd say in the last couple of years the consumer mindset has shifted into, 'How can I get something I need—whether it's transportation, goods, or services—in real time?'

Here are a few highlights from Teran:

It's funny, we get that question [about why we opted to go the employee route] a lot. For us, it was a very simple decision. We wanted to provide the best service to our customers, and we found that in order to do that we needed to have the best employees. To have the best employees, we needed to be the best employer, and the only way we could do that was by providing things that you can only do as an employer. Things like training, benefits, career progression. It just didn't make sense for us, really, to be able to deliver at the service level that we wanted, to use contractors.

We're seeing a lot of companies now struggle or go out of business, and it's not really a one-size fits all. Uber has an amazing business model and an amazing business, but the 'Uber for X' paradigm is broken if you blindly apply it to any industry.

I actually had this thought this morning getting out of bed that the on-demand economy is just turning the world into a luxury hotel.

Subscribe to the O'Reilly Radar Podcast: Stitcher, TuneIn, iTunes, SoundCloud, RSS

Dave Zwieback on learning reviews and humans keeping pace with complex systems Dec 10, 2015

Show notes

O'Reilly Radar Podcast: Learning from both failure and success to make our systems more resilient.

O'Reilly's Jenn Webb chats with Dave Zwieback, head of engineering at Next Big Sound and CTO of Lotus Outreach. Zwieback is the author of a new book, Beyond Blame: Learning from Failure and Success, that outlines an approach to make postmortems not only blameless, but to turn them into a productive learning process. We talk about his book, the framework for conducting a "learning review," and how humans can keep pace with the growing complexity of the systems we're building.

When you add scale to anything, it becomes sort of its own problem. Meaning, let's say you have a single computer, right? The mean time to failure of the hard drive or the computer is actually fairly lengthy. When you have 10,000 of them or 10 million of them, you're having tens if not hundreds of failures every single day. That certainly changes how you go about designing systems. Again, whenever I say systems, I also mean organizations. To me, they're not really separate.

I spent a bunch of my time in fairly large-scale organizations, and I've witnessed and been part of a significant number of outages or issues. I've seen how dysfunctional organizations dealing with failure can be. By the way, when we mention failure, it's important for us not to forget about success. All the things that we find in the default ways that people and organizations deal with failure, we find in the default ways that they deal with success. It's just a mirror image of each other.

We can learn from both failures and success. If we're only learning from failures, which is what the current practice of postmortem is focused on, then we're missing ... the other 99% of the time when they're not failing. The practice of learning reviews allows for learning from both failures and successes.

In the practice of learning reviews—and of course, this is also present in the "blameless postmortems"—we don't focus on a single root cause, but we focus on a bunch of conditions. That comes not out of anything other than the realization of the complexity of the systems that we work with.

In the current practice of postmortems, we talk about accountability, but really what that version of accountability means is, who's throat are we going to choke. Who are we going to punish? ... In a learning review, we go beyond blame to achieve real accountability. ... If there's blame, or there's punishment, then you're not going to get the full account. You really cannot fully hold people accountable.

The other sort of lineage of removing blame and punishment actually comes from a normal non-restorative or punitive justice system, where in certain situations, we give people immunity. Why? So that they can give us the full account of what happened. We do that sometimes with people we know have done bad things. In mafia cases. In those cases, what we are saying or doing by giving people immunity is that we value the information they provide to us more than pushing them.

Why do we want to go beyond blame, why do we want to go beyond bias? Those two are the short tickets to learning. ... More importantly, it's not a one time thing. We continually have to be learning about our systems and feeding that knowledge back into that system to make it more resilient.

Subscribe to the O'Reilly Radar Podcast: Stitcher, TuneIn, iTunes, SoundCloud, RSS

Jeff Jonas on context computing, irresistible surveillance, and hunting astroids with Space Time Boxes Dec 03, 2015

Show notes

The O'Reilly Radar Podcast: Context-aware computing, privacy by design, and predicting astroid collisions.

O'Reilly's Jenn Webb sits down with Jeff Jonas, an IBM fellow and chief scientist of context computing, Ironman triathlete, and contributing author to Privacy in the Modern Age: The Search for Solutions. Jonas talks about applications of context-aware computing, his new G2 software, and astroid hunting with astronomers at the University of Honolulu.

Here are a few highlights from our conversation:

The definition I'm using of context is this: to better understand something by taking into account the things around it. Context computing is taking a new piece of data that arrived in the enterprise as a puzzle piece and finding other pieces of data that had been previously seen and see how it fits. Instead of using algorithms staring at puzzle pieces, you end up with whole chunks of the puzzle and it's much easier to make a high-quality prediction.

The purpose of G2 is to be able to take structured and unstructured data from batch or streaming sources. Think of it as new observations across a virtually unlimited number of data points. You could think of this as internet of things feeding it or transactional systems or social data or mobile data. It's about weaving all those puzzle pieces together and then using the puzzle pieces as they land to figure out what's important or not and use these system to help focus people's attention.

I'm super interested in something called privacy by design. It's an interesting question, when you ask people if they even looked at the terms of use and people say, "No." Here's my prediction: the surveillance society is inevitable; it's irreversible; but the most interesting thing is, it's irresistible and you're doing it. Organizations keep creating these irresistible services. … the general rule is you have to tell your employees and your customers what data you're collecting and why, and then use it for that purpose. Then this whole field of privacy by design, and one of my goals with this G2 project, is I wanted to bake the privacy in before we even started. So, I took every privacy principle that I'd come up with in the prior years, and I baked it into this new thing.

There are 600,000 known asteroids, but none of them hit Earth. Now and then, they hit each other. It's only been seen twice. The first time, about five years ago, the Hubble telescope just took a picture. In the middle of the picture is a giant X. It's because two asteroids hit each other. Total accident; it was not predicted. I asked them, 'Well, why didn't you compute if the asteroids were going to hit each other?' They said, 'You silly fool. That's multi-body orbit math, which means expensive, and it's an N-squared problem, which means 10 million computer hours.' I said, 'But if you use these Space Time Boxes, you can figure out if they're ever going to be near each other and then only use the heavy compute if they're going to be near each other.' We did a 25-year forecast, and now they're pointing the telescopes at places in space and actually watching asteroids get close to each other.

Subscribe to the O'Reilly Radar Podcast: Stitcher, TuneIn, iTunes, SoundCloud, RSS

Kristian Hammond on truly democratizing data and the value of AI in the enterprise Nov 25, 2015

Show notes

The O'Reilly Radar Podcast: Narrative Science's foray into proprietary business data and humanizing machines to bridge the data gap.

O'Reilly's Mac Slocum chats with Kristian Hammond, Narrative Science's chief scientist. Hammond talks about Natural Language Generation, Narrative Science's shift into the world of business data, and evolving beyond the dashboard.

Here are a few highlights:

We're not telling people what the data are; we're telling people what has happened in the world through a view of that data. I don't care what the numbers are; I care about who are my best salespeople, where are my logistical bottlenecks. Quill can do that analysis and then tell you — not make you fight with it, but just tell you — and tell you in a way that is understandable and includes an explanation about why it believes this to be the case. Our focus is entirely, a little bit in media, but almost entirely in proprietary business data, and in particular we really focus on financial services right now.

You can't make good on that promise [of what big data was supposed to do] unless you communicate it in the right way. People don't understand charts; they don't understand graphs; they don't understand lines on a page. They just don't. We can't be angry at them for being human. Instead we should actually have the machine do what it needs to do in order to fill that gap between what it knows and what people need to know.

The point of the technology is to humanize the machine so we don't have to mechanize people. I always think it's a sad, sad state of the world where technologists keep demanding that everyone become data literate. What they mean is that everyone needs to have the analytical skills needed to look at a data set and figure out what's going on. I always see that as technologists saying, "We failed. We could not figure out how to explain to you what's going on, so you have to have our skills." While I think it's a noble notion that everybody has these skills, it's not going to happen. At the end of the day, it's not democratizing data to say we're going to do that. It's meritocratizing data. It's saying, "The only people who are allowed to understand what's happening in the world, based upon this data, are the people who have these high-end skills." It's incumbent upon us as technologists to move that data into information that is absolutely accessible to regular people. If we don't do it, we have failed.

I think over the next two years, we're actually going to see a shift in the business attitude toward artificial intelligence. Right now, businesses are really struggling with, "What's going to be my AI or cognitive computing strategy?" That's going to shift into, "I have particular problems, are there particular AI systems that can solve these problems?" What we're going to get is a much more rational approach to the introduction of AI into the business world. It's not, "We need machine learning," it's, "We actually need to understand Churn." It's not that we need predictive analytics, it's that we actually need to know when our supply chains are going to break down.

Subscribe to the O'Reilly Radar Podcast: Stitcher, TuneIn, iTunes, SoundCloud, RSS

Mike Kuniavsky on the tectonic shift of the IoT Nov 19, 2015

Show notes

The O'Reilly Radar Podcast: The Internet of Things ecosystem, predictive machine learning superpowers, and deep-seated love for appliances and furniture.

O'Reilly's Mary Treseler chats with Mike Kuniavsky, a principal scientist in the Innovation Services Group at PARC. Kuniavsky talks about designing for the Internet of Things ecosystem and why the most interesting thing about the IoT isn't the "things" but the sensors. He also talks about his deep-seated love for appliances and furniture, and how intelligence will affect those industries.

Here are some highlights from their conversation:

Wearables as a class is really weird. It describes where the thing is, not what it is. It's like referring to kitchenables. 'Oh, I'm making a kitchenable.' What does that mean? What does it do for you?

There's this slippery slope between service design and UX design. I think UX design is more digital and service design allows itself to include things like a poster that's on a wall in a lobby, or a little card that gets mailed to people, or a human being that they can talk to. ... Service design takes a slightly broader view, whereas UX design is — and I think usefully — still focused largely on the digital aspect of it.

I have a deep, long-seated love for appliances and for furniture because they are the tools of our everyday lives, and if anything becomes the content of this new Internet of Things thing first, it's them. What's interesting to me is that they have an already existing set of affordances, which means people know what to do with them and how to do it. They have a set of expectations, and how this set of things can now utilize this amazing set of sensing and actuation and meaning-making and statistical analysis technologies that are available up in the cloud, to do the things that they have always done, but do it better. I'm really interested in how intelligence affects the appliance industry.

There are ways to spin the IoT as an Orwellian cyberpunk anti-future, things that spy on you from every corner. They will do that, but I'm not that interested in that aspect of it. I think, actually, that humans are pretty good at negotiating their technologies, even though it sometimes takes a while.

The thing that is happening right now is that by connecting all of these different sensing devices, you turn that sensor input from this very simple gas gauge-like thing that might be useful to somebody in one situation, to a sequence of knowledge that can be modeled and can be much more broadly useful, especially when you have many, many different sources of information that are coming together. That, to me, is a tectonic shift because now you can essentially reason on a giant quantity of information, but the end points that are collecting this information or acting on it can be incredibly small and thin. You get the full power of these enormous artificial intelligence systems, machine learning systems, but without any of the computational overhead or cost, locally. That is really powerful. Every single little thing becomes as powerful as the most powerful computer on earth, and can then anticipate, compensate, and work together with other things in ways that were inconceivable before this shift.

What I'm interested in, broadly speaking, is predictive analytics — I should say, machine learning, statistical modeling, but specifically in predictive statistical modeling, predictive machine learning. I think, really, that is the new super power.

Subscribe to the O'Reilly Radar Podcast: Stitcher, TuneIn, iTunes, SoundCloud, RSS

Our TOPPODCAST Picks

Follow Us

Stay Connected

O’Reilly Radar Podcast – O’Reilly Media Podcast

Related Podcasts

Links

Stay Connected