John Bruggeman

What About VxWorks?

There was quite a bit of media coverage about Wind River’s announcement last week, but if you haven’t already read about it, we acquired the intellectual property for RTLinux, the industry’s only commercially-available, hard real-time Linux technology developed by FSMLabs. You can learn more about it here.

Now while the technology we acquired will be integrated with our Linux platforms, this asset acquisition could raise bigger questions about our future commitment to VxWorks - which for the record, continues to grow and is the most widely adopted commercial real-time operating systems in the device software industry. So in an effort to eliminate any doubt or confusion, I want to make sure I address these questions directly.

So what about VxWorks? If any conclusion regarding Wind River’s product strategy is to be drawn from our acquisition of the RTLinux technology, it should be about our commitment to offering device manufacturers a robust portfolio of real-time capabilities: soft real-time, hard real-time, Linux or VxWorks. Any conclusion that this acquisition was about perceived lack of or scaled back commitment to VxWorks, would be extremely inaccurate.

I have said it often and will continue to say it because platform choice is a very important component in the value delivered by a comprehensive DSO strategy: Wind River is committed to the VxWorks platform – VERY committed, in fact. We will continue to make long-term technology and ecosystem investments in the VxWorks platform. VxWorks customers can expect continued and ongoing improvements in the platform such as better performance, reduced memory footprint, additional support for multiprocessing, additional leading-edge networking features and specific requirements for markets with strict certifications that Linux can’t meet.

In a nutshell, acquiring hard real-time Linux technology was merely a move to broaden the company’s real-time expertise and deliver additional choice and flexibility. Our customers make platform choices for a variety of reasons – some reasons are based on requirements and others are based on preference. What are some of the requirements and preferences VxWorks customers care about? VxWorks was architected from the ground up to be a real-time system and will continue to have smaller footprint capabilities than Linux – some customers care about footprint. Also, due to its large size, Linux is not easily certified for certain government and A&D certifications – a requirement only certain segments of customers care about. Not all customers have embraced open source and the idea of GPL. Others have already made huge investments in VxWorks. The point here is that different customers have different requirements and preferences which means they demand choice.

Wind River will continue making investments in technology innovation that meets the needs of our customers – on BOTH the VxWorks and Linux platforms. These investments should not be interpreted as anything other than our ongoing efforts to deliver the choice and flexibility device manufacturers have long asked for. They want to build the best devices they can, using the platforms and technology they prefer and that meet their specific requirements.

We call it DSO.

Even Bacteria Form Communities

After several days at CES, I was so OVER IT by my final night in Vegas. OVER the crowds, OVER the lines and especially OVER the abundance of cool, new gadgets I live to try out every January. I was “gadgeted-out.” As amusing as it was to hang out in front of the Taser booth where grown adults were waiting in a VERY, long line to voluntarily sacrifice their neuro-muscular systems and experience total bodily incapacitation from the uncontrollable jolt of an extremely painful electrical current device…I was still OVER it. 

With a grueling show schedule behind me, all I wanted to do was hideout in my hotel room, order room service, plow through the 3000+ emails in my inbox and maybe channel surf my way to something interesting on TV that wasn’t a repeat episode of CSI, Law & Order or Cops. As I powered up the laptop and simultaneously started clicking through the channels, I stumbled onto a Nova special about bacterial research and actually watched the entire segment completely engrossed. So much for the 3000 emails…

The segment was about a scientist, Dr. Bonnie Bassler, a Princeton geneticist and Howard Hughes Medical Institute investigator. While researching how bacteria “talk” with one another and act as groups, Dr. Bassler discovered that bacteria communicate with a chemical language called quorum sensing. Prior to this discovery, the ability of bacteria to communicate with one another was considered to be an anomaly that occurred only among a few marine bacteria. Based on Dr. Bassler’s research, it is now clear that “group talk” is the norm in the bacterial world.

During the segment, a couple of examples were highlighted to demonstrate quorum sensing. An experiment working with Vibrio Harvey, harmless marine bacteria that glow in the dark, caught my attention. The experiment demonstrates how quorum sensing allows bacteria to act like an enormous multicellular organism. This particular type of bacteria makes bioluminescence - meaning it basically lights up like a firefly. According to Dr. Bassler, this glowing trait is controlled by quorum sensing.

The experiment starts out with a jar placed in a dark room. The jar contained Vibrio Harvey bacterium that was dilute. Because there were only a few cells around, the jar did not light up and glow. But when the bacterium grew to a particular cell number, all the cells turned on light together and the jar started to glow. By isolating the bacteria into different situations – alone versus together – the glow of the light indicated when the cells were talking and working as a community.

Now, while this experiment kicked off the discovery of quorum sensing, learning how bacteria communicate can be crucial to fighting deadly strains of bacteria and for understanding communications between cells in the human body. From a perspective of community effectiveness versus individual contributions, quorum sensing is another example of the power that can be derived from coordinated group efforts. If even bacteria naturally form communities to accomplish activities and are more effective as a group, then certainly it is an approach that should be widely encouraged and embraced in the device software industry.

Come to think of it, as I head into the final budgeting process for Wind River’s upcoming fiscal year, Dr. Bassler’s research and the Vibrio Harvey experiment merely reminded me why we invest so heavily in standards bodies and other community-related activities. I believe the group efforts of the device development community will eventually lead us to glowing jar.

Next-Generation Applications and The Device Food Chain

2006 is in the home stretch. Only a week or so left and we can ring in 2007. December was a busy month for me: sales training, customer meetings in Europe and finally, the release of our full product line updates delivered late last week. We’ve accomplished a lot this year and I look forward to continuing the momentum when we kick off FY08 in February.

As Chief Marketing Officer of Wind River, one frequently-asked question I get...A LOT...is: “How big is the DSO market?” This question has become one of my favorites to answer because there are so many mind-shattering things happening in just about every segment of device category available to end users. I try to explain how the potential size of the DSO market is directly linked to the growth of the unquestionable explosion of demand for next-generation applications and services. As more sophisticated applications are desired by end users, next-generation devices must be developed to support these applications. This demand triggers what I call the “device food chain” and all the technology providers up and down this chain have the potential for significant growth. Smartphones alone represent a huge opportunity for Linux device software providers and this category doesn’t even factor in RTOS-based devices.

Internet Video is another great example of an application that will increasingly drive the demand for next-generation devices.Consumers are showing early indications they will readily consume applications that enable access to video content via many different types of delivery mechanisms such as iPods, mobile phones and PCs. In a BusinessWeek article this week, about broadband access and Internet video, JMP Securities Financial Analyst, Samuel Wilson, used a common analogy for this market by comparing it to an arms race. According to Samuel, the place to invest in an “arms race” market like Internet video is with the arms merchants or in my analogy, the players within device food chain. Several technology providers in the Internet video food chain were mentioned in the article such as Juniper Networks and Netgear, Inc – both networking equipment manufacturers - and NDS Group, a UK company working on security software for “hybrid” set-top boxes that stream TV shows from the internet to satellite users.

How does this article relate to the DSO market? Being a leading, broad-based business publication, I suspect most BusinessWeek readers aren’t familiar with embedded device software and probably don’t really care to read about it. Therefore, in the interest of not boring readers to death, this Internet video article didn’t go far enough down the food chain to include examples of device software providers that enable next-generation devices...the devices that ultimately enable next-generation applications and services. The point here is that the types of “arm merchant” companies mentioned in the BusinessWeek article are either the EXACT types of customers who have bought or are looking to buy a DSO solution OR the EXACT types of technology vendors that want to partner on joint development of a DSO offering.

In this particular market example, as consumers continue to demand next-generation Internet video applications, they will continue to buy the devices that enable these applications. As demand for these devices explodes, technology providers within the device food chain will also experience an increase in demand for products, technology and enabling partnerships.

Regardless of the device segment, I see the same cause and effect situation occurring throughout the industry. Demand for new applications leads to demand for supporting devices and therefore, the underlying technology as well. Over time, this linkage will increasingly drive growth of the DSO market as a whole and will also foster a broad ecosystem that supports the demand.

Using the Internet video market as merely one example, it becomes pretty clear why it’s difficult to predict or pinpoint an exact numerical figure on the size of the DSO market. It seems to me that the accurate response to the market-size question is just, “Too large to count.”

-John

Services and Support Become the Killer Linux Apps

Yet another foreign government organization announced their intent to dump Windows for Linux and open source. Starting in mid-2007, the French parliament will equip the PCs in their deputies’ offices with a Linux operating system and open-source productivity software.

This announcement follows a September update from the city of Munich on the progress of its Linux project initiated back in 2003. If you happen to remember, the Munich project was one of the first to shoot right at the heart of Microsoft’s monopoly and helped fuel an industry debate around the viability and adoption rate of desktop Linux. At the time, the blogosphere lit up with potential speculation around the city’s decision ranging from payback for Microsoft licensing 6.0 to anti-American sentiment surrounding the Iraq war. It all seems like ancient history now, doesn’t it?

Beyond the historical appeal, the Munich project itself exposed some of the classic Linux deployment issues customers often face when trekking along the open source migration path. While undoubtedly we will continue to watch these government-sponsored open source projects roll out, they do highlight the potential for another form of open source contribution that as a community, I hope we are committed to embracing. Now that the early-adoption projects are going live we should begin to focus not only on sharing code, but also identifying and sharing best practices for the successful deployment of open source solutions.

Having made this call-for-action, I know someone out there is going to ask me what best practice I would contribute…so here it goes:

In the device software market, I keep an ongoing, direct dialog with Wind River’s customers about the open source issues that concern them and the best practices they feel have helped them be successful with Linux adoption. Based on my discussions, if I had to relay only one piece of advice relevant for any customer (enterprise or device manufacturer) considering a Linux solution, it would be to prioritize and carefully select options for professional services, support and training as part of an overall migration plan. Incorporating globally accessible professional services, support and training offerings from open source experts is a key learning that our customers tell us again and again has led to their project success. 

Apparently I’m on to something because a “special sauce” Linux distribution was not the “Killer App” Oracle pulled out of its powerful arsenal to take on Red Hat. Nope, as we learned a few weeks back, services and support became the “Killer Apps” Oracle is relying on to entice Red Hat customers away. Oracle’s Linux distribution will be a clone of Red Hat’s making services and support the differentiator based on price. It's getting interesting - the competitors are starting to place their bets and tip their hands.

Enough about Oracle and Red Hat though. My point is really about broadening our approach for contributing to the community. If, by any chance, you have any thoughts on best practices for open source adoption, please share. I know there are lots of us out there who’d be interested and grateful for the input.

Open Source: The Universal Model for Innovation?

I came across an interesting article yesterday from the Harvard Business School Working Knowledge site. Thearticle is about an assistant professor,  Karim Lakhani at HBS who conducted some research on how an open source “broadcasting” model taken from software development scenarios, might work among scientists trying to solve scientific problems. According to his research results, practices in the open source software community do in fact offer a model for encouraging large-scale scientific problem solving. The research team discovered that when a problem resides in one domain of expertise, the solution may reside in another domain because innovation happens at the intersection of disciplines. By broadcasting a problem to outsiders, solutions were often found much faster than situations that did not incorporate open source practices.

As an active proponent of an open source model and all of the benefits that have been realized by this approach in the software industry, the results of this study are quite fascinating. Just like the software community where the roadblock to embracing open source has been fear over IP protection from companies competing for software sales, in this science scenario, the same kinds of concerns exist but in this case amongst individuals competing for research publication. The lesson here is that the broadcasting model of sharing problems and solutions, still worked in the science environment and significant benefits were achieved.

Wind River is a company committed to the open source model and contributing to the community in a variety of ways from code contributions to resource investments. We believe in active community participation and the positive results we deliver to our customers from the industry standards we support. After all, the software community is THE open source research study demonstrating on a daily basis what the impact can be from lots of different people working together to solve the problems facing this industry. The results of this study prove there should be no doubt that open source is a real, valid and very effective model for successfully driving innovation regardless of what the proprietary vendors try to position in the market.

I guess if you read this HBS article and get nothing else from it, the results of this study certainly support the theory that the open source approach of collaborating on problem solving is truly a universal model for innovation. Pretty phenomenal if you think about some of the global problems we face as a civilization…cancer, hunger, war, etc. The impact of global adoption for open source problem solving could significantly improve the world as we know it today.

Would You Pay $1200 for a Game Console?

The long-awaited Sony PlayStation 3 game console launched today in Tokyo. Man, I haven’t seen a line that long since I promised my son we’d camp out at Barnes and Noble to get him a copy of the hottest book around,  Harry Potter and the Goblet of Fire the day it released back in 2000. Well wait a minute - now that I think about it, maybe the line to get a Tickle-me-Elmo for my niece at FAO Schwartz on Black Friday might have been longer in the late 90’s. Oh and then there was that line for Halo 2 in 2004…

No matter, PS3 is just the latest example of how supply and demand is an economic principle alive and thriving thanks to today’s pop culture. Even though PS3 will retail at $499 - $599 U.S., given the limited quantity immediately available for the holiday rush, the wild supply/demand roller coaster will have the potential to cause a severe “loss of stomach” effect on American consumers – you know the feeling when you dream you are falling? News reports are predicting that for those of us who will be working on November 17, the day PS3 launches on our continent, we can expect to pay upwards of $900 - $1200 U.S. via online auctions – the only place the consoles will be available. As much as I like playing video games – and apparently standing in line, I just don’t know if I’m going to make the investment this time around. Ok, who am I kidding? There is definitely a PS3 destined for my game room in the near future.

Being in the device software industry, the availability of PS3 provides a unique glimpse into one of the industry’s key vertical markets and will demonstrate over time whether or not end-users will truly pay for premium functionality like I believe they will. While all the media hype at the moment is focused on the buzz factor of the launches themselves, moving forward, all three next-generation game consoles finally will be available and will offer varying price points to consumers to make a choice: Xbox 360, PS3 or Nintendo’s Wii.

It is no secret that Wind River’s business strategy is based on the premise that device manufacturers will increasingly choose a commercial device software platform in order to focus on developing functionality differentiation. By leveraging a commercially available platform, they can direct internal resources to build the features that enable them to compete. From a different angle though, device manufacturers can also win or lose their competitive battlefields on the basis of time-to-market – another significant benefit of choosing a commercial platform. If a device is first-to-market and has decent functionality, will it beat out later market entrants? The gaming console war is a fascinating case study to follow: will consumers pay the premium for bell and whistle functionality in the PS3; or did Microsoft take the power out of Sony’s punch by delivering Xbox 360 a year sooner and at a better price point?

Only time will tell, but I plan to have a lot of fun watching this battle play out.  Time-to-Market v. Premium Functionality…makes me think of Alien v. Predator…yet another line I stood in… 

“Hugging it Out” with Open Source…

Congratulations Microsoft. Responding to your community? Embracing an open source model in a non-trivial way? These are big, admirable steps I thought I would never personally witness. It always makes my day to see another industry leader make business decisions based on the same market premises I believe. Customers want choice and flexibility. For those of us in the industry that haven't learned the lesson by now, the key enabler has become community developed standards. By working together, we only make each other stronger.

Linux and Automotive: Reaching a Tipping Point

Linux has always been suspect when it comes to automotive applications.

The general belief was that safety-critical automotive applications required a deterministic, real-time operating system.  Linux just didn’t cut it for telematics or under-the-hood applications.  OEMs relied on proprietary OS companies for their solutions.

Now some of that belief is grounded in fact. Consequently, Linux adoption in automotive has moved at a snail’s pace.

But I believe moving forward, Linux adoption in the automotive industry will happen quickly, dramatically, and in ways we never expected.

We are now at a “tipping point” (Wikipedia) the point at which several key drivers align to make the uncommon common.

Here’s what I see:

First, there are the business drivers.  The automobile industry is hungry for a new business model. Gone are the days when a car company could charge a premium for anti-lock brakes or air bags.  Today, these are expected features in the base price of a car.  Those high margins have dried up.

The auto industry now needs a business model that looks more like the consumer electronics industry: faster development cycles, higher margins, and amazing new devices and features that consumers will willingly pay a premium for.

For the automotive business, this new model has come in the form of “car infotainment”: navigation systems, in-dash audio and video systems, etc.

Infotainment is where luxury car companies are making their high margins, and it’s moving down to mid and low-end cars.  And margins can stay high because Linux is an adequate solution for infotainment applications.  Nobody’s life was threatened when an audio system failed.

But Linux is now quietly creeping into another set of high margin automotive applications—applications you don’t necessarily associate with infotainment.

They’re “driver assist” applications that automatically help you park your car, or change your ambient light setting, or adjust your rearview mirror.  They’re the kind of non-infotainment applications that consumers perceive as high value.

What’s more, those applications are all run out of the infotainment control unit.  So, the perception between what are “consumer electronics” features in a car and “safety critical” features is blurring.

So what are the ramifications of a Linux tipping point? 

It means a rejuvenated automotive industry.  It means that consortia driving automotive standards (Autosar, JASPAR) really have teeth. It will also mean new safety standards that can keep up with much faster development cycles.

Put on your seatbelts, this will be an interesting ride.

Airbus Gets DSO “Religion”

Most likely, you’ve already read the news about the A380, the Airbus flagship super-jumbo jet.  The plane will now be ready for delivery another year later than originally planned. 

That delay could cost Airbus’ parent company, EADS, $6.1 billion in lost revenue over four years.  And today it cost Airbus CEO, Christian Streiff, his job.

Ouch!

Now I won’t go into the gory details about the root cause of this delay (you can read IET’s story about it; or if you want more on the technical challenges of building one of these babies, read Paul Parkinson’s recent blog).   

But there’s a bigger story here, and it’s not about possible fallout like order cancellations or management changes.

The bigger story is a Device Software Optimization (DSO) story.

One of the key tenets of DSO is standardization.  That means standardized tools, standardized development practices, and standardized training deployed globally across an enterprise.

The Airbus delay resulted from the late implementation of tools, used by engineers who weren’t fully up to speed on how to use those tools. Consequently, moving forward, Airbus has ordered implementation of the same proven, standardized tools at all sites, as well as standardized training for all employees using those tools.

Of course, this delay is painful for Airbus.  But if it means that the company is adopting DSO, it’s the best thing that could ever happen to them.

DSO: It’s About Time

What will be the next advancement in digital video technology? Will it be increased capacity, smaller hardware, or better resolution?

You know what? I really don’t care. In the world of Device Software Optimization (DSO), none of it matters. What really matters is that device software and hardware (like digital video) are fundamentally affecting how we think about time.

Here’s what I mean. Back in the days of still cameras, all of us strived to capture the "perfect moment". It was the moment your son’s bat made contact with a baseball. Or the exact moment your daughter puffed up her cheeks and blew out the candles on her birthday cake.

Timing—and time—was everything.

But the "perfect moment" doesn’t matter anymore. Now it’s all about capturing a "segment" of time.

For example, over the weekend I was at my son’s football game. I used a digital video recorder to capture footage of him running down the field. In the past, I would have used a still camera to chronicle the "perfect moment" when he crossed the goal line. But instead, in a timeless world, I was shooting a "perfect segment" of my son. A segment that is digitally captured and stored, and can be recalled and played at any time.

This new concept of time touches all of our lives in so many ways. Look at what’s happening to the notion of "broadcast". Audio/video capture, display, and storage technologies have converged. Now I can watch millions of "events" whenever I want (even without commercial breaks!) As a result, history as we know it becomes an event shifted in time, rather than a fading memory.

But it doesn’t stop there.

Our global communications infrastructure, for one, allows us to interact with others instantly at any time, in nearly any location. What’s more, GPS technology allows us to know where people are at any moment in time. Combine both these technologies and we change the very way we live our lives.

Time waits for no one? I’m not so sure anymore.