Each year EMF conducts a thorough and detailed survey of embedded developers. Using the EMF Dashboard – a web based tool that permits vendors and developers to correlate information between any responses to any question, EMF presents selected insights from its analysis of the 2010 data. The Dashboard enables vendors to look at the responses of their customers (and potential customers) as well as to their competitor’s customers. This provides invaluable insights for strategic and sales planning. Developers can see what their fellow developers are considering in their selection criteria

Product opportunity windows are fleeting and time-to-market issues dominate design considerations. The two factors that invariably make the difference between success and failure are first, the knowledge that comes with an accurate insight into the internal and external forces which drive product markets and second, an insight into the concerns, desires and thought processes of those customers who make the decision to purchase a particular product or deal with a particular vendor.

The following data, taken from the 2010 EMF Embedded Developer Survey, addresses many issues of which embedded vendors and developers should be aware. The data presented here represents embedded industry averages across many application verticals, many vendors, many OSes, etc. Subscribers to the 2010 EMF Market Intelligence Program have access to their custom Executive Dashboard with which they can create unlimited cross tabs to further examine the data presented here.

 Criteria Most and Least Important to Developers in Selecting an OS

Developers were asked to indicate which criteria were most important to their decision in selecting an OS. The top (most important) responses and the bottom (least important) responses are presented in the following tables.

There was an interesting reordering of priorities in 2010 with cost remaining the principal factor. Realtime performance fell (no surprise here) and safety certifiable and virtualization remained as a small part of the collective consciousness of embedded developers.

(Sung to Santa Claus is coming to town)

Yes indeed – and it’s about time.

The FDA’s Center for Devices and Radiological Health (CDRH) reported that in 2006, 21% of all medical device recalls were for software defects – it is also estimated that one-in-three software-based products is recalled. They haven’t updated this data since, but one can assume that it might have gotten worse.

Medical device developers and company CEOs should be aware that this is an unacceptable situation and that it is going to rapidly change. Those that don’t get it straight won’t be around when the dust settles.

Is there enough aspirin to relieve the headaches that our government is giving medical device manufacturers? Moreover, is it deserved? The answer – depending on the specific question – is yes and no.

The Obama administration under pressure from Congress has combined the medical device industry with the drug industry as one, and has proposed taxing both industries to pay for universal health insurance (Obama Care). Hopefully thoughtful senators and representatives will see that these taxes will be passed down to patients and other users and will also impact the elderly who most require such devices and can ill afford them – then again it is more realistic to depend on the Easter bunny. But logic is no relief for the headache – particularly when it comes to Washington politicians. It’s no fun being a medical device executive today.

On July 31, 2008 a Senate Bill cosponsored by Senators Edward Kennedy (D – MA) and Chuck Grassley (R– IA) was filed that  would require senior officers or directors of drug and medical device companies to certify under penalty of perjury that all information submitted for a product’s approval is accurate and in compliance with federal regulations.

The Drug and Medical Device Accountability Act Bill expired at the end of the two year Senate session on December 31, 2008, but was refiled in the Senate (2009) with the hope of amending the current legislation by the end of October 2009. This is an important piece of legislation, and medical device executives should get their house in order to accommodate the provisions.

The Bill provided that product applications later found to have contained false or misleading information would be subject to stiff fines (up to $5,000,000), assessed both to companies and their senior officers, who, in addition, could face jail sentences of up to 20 years. These are serious issues. Currently the CDRH has a forensic group that looks at device software only after a device has been recalled.

This is a bad time, and a very costly time (regardless of the Act) for a medical device manufacturer – particularly if software development hadn’t been given the detailed oversight of using best practices. The new Obama tax on medical devices – used to pay for Obama Care – is allready a blow to the industry and to smaller medical device manufacturers.

The “Drug and Medical Device Accountability Act” will change the medical devices industry similarly to how the Sarbanes-Oxley bill impacted corporate accountability. Laws being what they are, we should expect overkill from its enactment. This is why medical device company’s senior management should take time to rethink their strategic approach to the delivery of their products.

EMF has available a report presenting alternative paths for developers to produce quality software for medical devices, minimize product recalls, and affordably provide comprehensive audit trails for CDRH inspectors (Critical Issues Confronting Medical Device Manufacturers). Keeping the company alive and your CEO out of jail are bonuses.

Fanaticism consists in redoubling your efforts when you have forgotten your aim – George Santiago

Tough economic times create strange bedfellows. Does one need to purchase a technology, with its attendant costs and complications, when a lease or partner relationship would suffice? Do complimentary technologies and markets provide a return greater than the sum of the parts – or is the result characterized as “subtraction by addition?”

 EMF believes that embedded consolidation through acquisition will be the norm over the next few years as roll backs in DoD discretionary funding impact the larger purchasers.

 Let’s look to four recent acquisitions with an eye on compatibility, growth potential and whether there is a measurable outcome. Is this a trend, a lifeline or a passing strategic initiative – you decide.

 These include:

• IBM Rational buys Telelogic
• Intel buys Wind River Systems
• Cavium acquires MontaVista
• Artisan acquires Aonix

IBM Rational buys Telelogic

Recently IBM Rational added Telelogic to their comprehensive product lines and markets. We reported it as the Great Train Robbery of 2007. For around $800 million, IBM became the most capable and effective organization in the embedded industry, bringing with it the ability to link IT and embedded technologies under the rubric they call “smart devices”.

• Telelogic (along with the I-Logix acquisition) products have particular strength in software and systems development (in particular Rhapsody), requirements management (DOORS), and enterprise architecture. Their reputation in automotive, telecom/datacom and aerospace and defense systems development is outstanding
• IBM Rational brings IT development strengths to the partnership in addition to Rational Development Suite, DoDAF and MoDAF architecture capabilities as well as an impressive and unique communication infrastructure for cooperative development and management (JAZZ)
• This combination of capabilities gave IBM the ability to integrate many embedded tools within their Model Driven Development (MDD) thereby expanding the marketplace for other embedded vendors.
• IBM Rational acquired some of the best systems engineering talent in the industry – and a senior VP that is revolutionizing the embedded industry by organizing around markets rather than products.

How has this worked out for IBM? In 2009, which was admittedly an economically troubling year, Telelogic revenue contributions nearly tripled. EMF calls this acquisition multiplication by addition.

Intel buys Wind River Systems

Did either of them think through their respective channel strategies? Will Freescale users convert to Intel – or will there be a sales pushback?

So what’s in it for Intel and why did Wind River bail out so easily? Let’s look at the facts:

• Intel doesn’t need the additional $300 million on their balance sheet
• Wind River has been losing market share for some time now and their “subscription model” (which they have abandoned) cost them dearly in the mil/aero marketplace and allowed Green Hills and LynuxWorks to significantly underbid them
• Rob Davidson, Wind River’s VP of A&D, has publicly stated (at an Intel sponsored event no less) “let me make it clear that Wind River is a Freescale company” since 85% of Wind River’s military sales are Freescale-based
• Intel, notwithstanding their successes, is facing considerable competition from ARM and FPGA vendors – particularly Xilinx
• The Mil/Aero market and the primes that spend the most on embedded technologies and sub-systems are decidedly moving to a “software centric” purchasing model

Truth be told – both Intel and Wind River are missing the emergence of an embedded market that will soon overshadow the rest of the marketplace. It involves several components – none of which are within the new Intel’s capabilities.

• Systems development – with C and C++ developers working simultaneously on the same development
• Unique problems associated with systems-within-systems developments and long term support of such
• The new era of “Smart Devices” – the merging of IT database capabilities securely communicating with intelligent embedded devices

So, how will this work out? From Intel’s point they have really nothing to lose. They bought Wind River at a somewhat generous revenue multiple, but will get that back in a few years, even if Wind River continues to lose market share. Wind River’s legacy revenues in telecom and mil/aero will continue to fill Intel’s coffers.

There is a chance that Intel will be able to cut into Freescale markets and pick up new chip revenues. However they have a terrible record when it comes to successful acquisitions. Wind River’s global distribution agreement with Kontron for VxWorks and Wind River Linux is an example of why Intel will make back their investment.

EMF rates this “neutral”. Whereas mission critical applications are a fraction of broader embedded sales, the lack of a competent systems capability will be a drag on future expansion opportunities.

Cavium acquires MontaVista

Here we have a chip company acquiring an established Linux market leader. MontaVista is a major driving force in the embedded Linux marketplace.

Fifty percent of the current Linux marketplace is for in-house, or what EMF calls Roll Your Own (RYO) Linux – as distinct from commercial Linux offerings. EMF research has shown, year-over-year, that commercial Linux products produce a superior ROI to RYO Linux. The commercial Linux market opportunity would be enhanced if commercial Linux products were able to rescue RYO developments that are failing or reach a stand-still position. MontaVista is the first commercial Linux provider to provide such a solution.

By having Cavium’s resources behind them, MontaVista is in a commanding position to maintain their advantage over Blue Cat and Wind River Linux initiatives.

EMF believes that this acquisition is an excellent one for the following reasons:

• Acquisition gives Cavium multiple revenue account opportunities. One by winning the processor design, two by selling SW and services during the design phase, and three by generating revenue in software and services even if they lose the design
• Cavium’s 140-2 security certification makes MontaVista the only embedded OS vendor that has this important security capability
• The investment of a public company like Cavium in MontaVista adds stability to the embedded Linux market.
• Cavium has a history of delivering solutions that work with other company’s processors.
• MontaVista should be able to invest in more multi-core and virtualization technologies. While this benefits Cavium, it also benefits all of their semi partners with multi-core processors. Another potential win-win.

Embedded Darwin Awards

One of my favorite readings is from the book The Darwin Awards which reports on those whose carelessness and ignorance contributed to their early demise thereby removing them from the genetic pool. Quoting from the 2008 edition:

“The Darwin Awards celebrates those who have dived headfirst into the shallow end of the gene pool. From offering a beer to a bear to self-testing a Taser to jumping off a drawbridge on a bike, The Darwin Awards Next Evolution honors these macabre and entertaining feats of hapless misjudgment”

The following example might have earned at least an “Honorable Mention” if the book carried embedded technologies as a subtopic. 

Artisan acquires Aonix

This to EMF is one of the stranger acquisitions of recent memory. Take one struggling company attempting to climb out of the shadow of IBM Rational’s MDD capabilities acquiring a company that has been in free-fall for the past four years in order to gain a marginal Real-time Java capability. Some may remember that 5 years ago RT Java was forecast to displace C++ for military applications in order to permit software to inter-operate across different hardware systems. Fast forward to today and C++ is more entrenched than is Java. So buying Aonix to gain access to a Java capability (that is competing directly with IBM) that is readily available for licensing is like searching for a wife at a brothel – too expensive (it’s cheaper to rent) and history suggests that it won’t provide a long term benefit.

Someone should have asked Artisan if they knew the difference between Santa Claus and the need for RT Java (the answer is that there really is a Santa Claus) before they made the acquisition.

To add to their pain is the re-branding of the combined companies to be called Atego (this makes sense to change the Aonix name – but the Artisan name as well?). This is the 3^rd acquisition in the past year and a half for the management bought-out Artisan enterprise.

So why did this happen – and to whose benefit is the establishment of Atego?

EMF might accept the argument that a vastly superior management team could make a difference by integrating the companies – yet the Aonix component remains under the control of the management team that ignored market realities and caused Aonix’ decline.

The Artisan folks are not stupid people – misguided perhaps. EMF suspects that the investment bankers at risk for each of the enterprises came together to attempt to minimize risk.

If so, perhaps these bankers might have followed Warren Buffett’s admonition “When management with a reputation for brilliance tackles a business with a reputation for poor fundamental economics, it’s the reputation of the business that remains intact.”

We are not sure that Buffett would characterize Artisan’s management as brilliant.

File under “assured mutual destruction.”

When will the next shoe fall? 

The Aerospace and Defense Industries of the United States are poised to undergo one of the most significant changes since the end of the Cold War, perhaps the most significant since World War II.  These impacts will reach down into the value chain, in some cases devastating naive second and third tier vendors as major primes shift financial challenges to their vendors.

These third tier vendors constitute the embedded marketplace.

One might expect that in a market contraction, companies would seek to outsource more seeking lower costs.  We think this will happen for some defense firms, but we believe that many will resist it.  The largest contractors have been formed by merger and acquisition activity.   In many cases, large fixed costs have been merged on corporate balance sheets and are being depreciated.  Outsourcing might require significant write downs for post-merger balance sheets.

EMF expects to see a necessary consolidation of vendor offerings as falling revenues will be stabilized by acquisitions – acquisitions that might be found at bargain prices.

EMF considers the usual suspects in its published research.

On November 10, 2009, Cavium, a publicly traded provider of highly integrated semiconductor processors that enable intelligent networking, communications, storage and security applications, signed a definitive agreement to acquire MontaVista. Cavium stated that MontaVista would continue to operate separately and their customers would not be restricted to using only Cavium processors.

Immediately, questions arose: 

• Who is Cavium and why did they make this acquisition?
• Was Jim Ready clipping Burger King Coupons?
• Was MontaVista motivated by Intel’s acquisition of Wind River?
• Does this mean that commercial Linux is facing financial do-do?
• What is really going on behind the scenes and is this a good or poor marriage? 

Security is one of the most neglected aspects of the embedded world and whereas the embedded focus has been on MILS and EAL security (which is a systems level orientation), the National Security Agency’s (NSA) mandated requirement for communications security (FIPS 140-2) is largely ignored. 

Cavium owns and markets the NITROX XL CN16XX-NFBE Adapter family; the world’s fastest FIPS 140-2 level 2 and 3 certified Hardware Security Module (HSM) with PCI-Express Gen2 connectivity in a low profile form factor and integrated Network Interface Card (NIC) functionality. 

Cavium’s more than 100 customers include Cisco Systems, Inc., F5 Networks, Inc., Sumitomo, SonicWALL, Inc., Citrix Systems, Inc., Blue Coat Systems, Inc., Juniper Networks, Aruba Networks, Inc., Nokia Siemens Network, IBM, Yamaha Corporation and Nokia Corporation. This is a very good match for MontaVista as well as for the embedded Linux marketplace.

EMF research has shown, year-over-year, that all commercial Linux products out perform “roll-your-own (RYO)” Linux developments – and MontaVista’s commercial Linux offerings showed a higher ROI than other commercial offerings. Their latest MVL-6 offering would permit RYO developers to move to MontaVista in the middle of their design without having to restart the development. EMF data also shows that embedded Linux use has become pervasive across many verticals and is preferred in many cases to commercial RTOSes.

EMF has been publishing and warning embedded developers, OEMs and systems integrators about the need for NSA required FIPS 140-2 security for the past 5 years – and we have been largely ignored. Cavium’s customers are keenly aware of the NSA’s requirement. Hence, we believe that this acquisition will further expand the use of MontaVista Linux by adding a security component that is unique in the embedded industry. 

The world of computer technology is rapidly changing and Tier 1 vendors that depend on military revenues are facing an 18% to 40% reduction in DoD discretionary funding. EMF believes that there will be a ripple down effect that will have dire consequences to Tier III (embedded) vendors. 

Hence, this acquisition makes sense to us as it brings financial stability to MontaVista and to the broader Linux community that it serves. 

In summary, we believe that this acquisition is an excellent one for the following reasons: 

1. Acquisition gives Cavium multiple revenue account opportunities. One by winning the processor design, two by selling SW and services during the design phase, and three by generating revenue in software and services even if they lose the design

2. Cavium’s 140-2 security certification make MontaVista the only embedded OS vendor that has this important security capability

3. The investment of a public company like Cavium in MontaVista adds stability to the embedded Linux market. We believe that it positions MontaVista in a more favorable light by dispelling rumors (true or false) regarding their financial stability (an important consideration in up selling to military and prime contractors whose projects run for decades)

4. Cavium has a history of delivering solutions that work with other company’s processors.

5. MontaVista should be able to invest in more multi-core and virtualization technologies. While this benefits Cavium, it also benefits all of their semi partners with multi-core processors. Another potential win-win.

IBM gets it! Why don’t others? With $22 billion in annual software sales they certainly qualify as the 2000 pound gorilla – but they don’t act that way. They didn’t get to that level by being arrogant (like another large gorilla?) – quite to the contrary.

• They strongly support the analyst community
• They have no secrets – they show us their roadmap, tell us what they have in the works, confess their concerns, listen and encourage other views
• Unlike some embedded vendors, they see analysts/editors as a respected strength to their business
• They subscribe to the best market intelligence and they study it and use it

I was recently at their Software Analyst Connect meeting as one of 100 invited analysts and the only guy who covers the broad embedded market (they do have a strong embedded component in Rational and their larger software business certainly supports embedded needs). Steve Mills, IBM’s number 2 guy (he reports directly to Sam P) was once again in attendance and available. Think of it – the genius who runs IBM’s $22 billion software division not only makes himself available, but brought more than a dozen of his VPs, IBM Fellows, Distinguished Engineers, etc. Casually dressed, friendly and likely to be spotted at the Sushi Bar, Steve is more than willing to engage in discussion or debate. Although he confesses to stealing from his people the best slides for his presentations, he is eager to make sure that we meet with whatever expert we need to engage.

This is a smaller IBM venue than what I annually attend in Orlando (over 2500 developers). The meetings go off like clockwork and events are managed professionally. They hire the best and support them. Diane Flis’ group is superb – the support staff that makes sure that I can meet who I need to speak with (one-on-one if I choose) and supports me all year is the best in the business. Linda Voyles, Silvia Galgano, Carol Gibbins, Teressa Jimenez, and Claudia McQuade were not only available, but frequently sought me out to make sure that I had the information I needed and that I was aware of presentations in which I might be interested. 

Diane’s dedication to supporting us is very important to the IBM philosophy. Some embedded vendors are very good at working with analysts – conversely and unfortunately some embedded CEOs don’t look upon us as a virtue – but as an obstacle. 

Steve Mills outlined the IBM software roadmap – and it has tremendous implications for the embedded world. His markets (at least those that want to survive) are moving away from the inefficiencies and associated cost of their businesses and industries to software solutions that not only have long term financial benefits, but have total payback of their investments within a year.

What does Steve Mills and his team know that you may not?

First, IBM focuses on solutions – not isolated tool or software sales. Their sales teams are focused on long term ROI as well as short term payback. Annual EMF surveys of embedded developers demonstrate that change for the sake of change is no longer a virtue. EMF data shows that the willingness of customers to change vendors is based on long term ROI and we have shown that the willingness to change processes and tool sets can be predicted from the gap between what developers say they use and what they perceive to be a best practice. 

IBM recognizes that businesses today are confronted by unparalleled rates of change that create tremendous challenges. Companies need to differentiate products, react to on-going market shifts, efficiently streamline support of deployed products and exploit globalization. 

What does IBM see that helps to size the potential embedded market? 

• Our world is becoming instrumented: there are 30 billion RFID tags in use and over a billion cell phones
• Our world is becoming interconnected: 2 billion people on the web and a trillion connected objects
• All things are becoming intelligent. There are 15 petabytes of new information generated daily

These are building blocks for what IBM calls a smarter planet. As our cell phones adopt 4G capabilities, think of the plethora of embedded devices that can connect to higher level databases through IT systems and Cloud-based computing. News specific to one’s needs can be automatically captured and delivered in realtime to hand-based or home-based systems. The same holds true for entertainment, banking, external monitoring of utilities, weather updates, automotive planning and assistance, police/emergency access, personal pharmaceutical updates and drug interaction alerts – just to name a few applications that are within our reach and are based on interactivity of embedded devices matched with enterprise data sources. 

Add to this the fact that we have extensive infrastructures built into energy delivery (National Grid), telecom/datacom facilities, traffic control systems, warehouse and supply systems, etc.

What characterizes these infrastructures is that it would cost trillions of dollars to replace them – this won’t happen given the cost of each and the disruption that would be created. They are further characterized as being systems and systems-within-systems and therefore address a different level of embedded requirements than RTOSes, software tool sets, embedded databases, and communication protocols. Such systems technologies lend themselves to Model Driven Development (MDD) usually based upon the Unified Modeling Language (UML), and currently used in not only advanced systems developments, but more now in complex embedded product developments. 

These infrastructures do lend themselves to innovative software developments, using established technologies that make these systems more intelligent and easier to deploy and maintain. The efficiency that such software development brings to these older technologies is that businesses see very short term payback of their investments by reducing the cost of support teams and having longer time periods before required maintenance, 

This is the future of embedded computing and a peek at the growth potential of our industry. Interestingly, IBM doesn’t own it alone. Their IT, enterprise and MDD capabilities are designed to make use of other embedded products. 

Just a final note: the conference was held at the beautiful Hilton Hotel in Stamford Connecticut which had wonderful accommodations – BUT can’t figure out how to deploy wireless access. Their rooms only have Ethernet connections (only half work) and their open spaces seem to provide the quality usually found with two tin cans tied together. 

I suggest that Hilton try to get IBM to buy them. Steve, are you interested? They have a really nice Sushi Bar.

Co-authored by: Dolores Krasner, VP Market Intelligence, EMF

Remember the old song “Signs”? The lyrics went “signs, signs, everywhere signs, messing up the scenery blowing my mind - don’t do this do that, can’t you read the signs”?

With all of the FUD, claims and counterclaims of misrepresentation between embedded vendors, what is a developer, manager or executive to believe, and how is one to make sense of whether one product or another is best suited for one’s use? No wonder potential users are leery of advertised and promoted claims.

Is it possible that those making the most noise and creating the most FUD are those messing up the scenery for the rest of us? Moreover, are these disruptions taking us away from the real signs that are defined by developers and managers that detail their likes, dislikes, and issues of greatest importance? Finally, what are the market trends that are characterized by revenue growth, best practices and ROI calculations?

I chose the above graphic to illustrate my frustrations (and I suspect the frustrations of others) with the misleading hype that has unfortunately become part of our embedded market culture. What I loved about the graphic was the ridiculous message that hid the information of most importance to the reader – the bridge was out!

So what should the embedded market signs tell us – based on year-over-year EMF Developer Surveys, vendor reported shipments and EMF privleged information - about the road ahead and how to avoid the bridges that are out?

 Sign #1: Danger Ahead

There will be a ripple down effect from primes to Tier II suppliers to Lower Tier vendors. The larger embedded vendors will get hit but not as bad as smaller vendors that don’t plan ahead and create value. Consolidations may result with financially more stable companies picking and choosing technologies at fire sale prices. It wasn’t too long ago that Motorola Computer Group/Force Computers were sold off at 28 cents on the dollar.

Sign #2: Remind your Customers of Your Value 

It’s a good idea to not only remind your best customers how you are performing for them, making a contribution to their goals and saving them money. It’s best to show them (if you have the data) how developers and managers throughout the industry (or within a specific vertical) are performing with your products and support. 

Sign #3: Promote Yourself

Certain behaviors which are highly correlated with high performing product developments may also help suppliers target their preferred customers, and help promote being a high performance supplier.

EMF data was separated into two cadres: A ‘Leader” cadre tended to complete product development on, or ahead of schedule, and the delivered product met pre-design expectations.  A “Laggard” cadre tended to finish late and delivered products that missed pre-design expectations

Comparing these high and low performing organizations, we found that high performers:

• Were more likely to use modeling for design, system architecture, algorithms and other components than laggards, who were less likely to incorporate modeling into the design process.
• Were less likely to use CMMI.  Laggards were roughly twice as likely to use CMMI as leaders
• Were more likely to use object oriented development.  Leaders were roughly twice as likely to describe their development processes as “object oriented” than laggards.
• Were likely to use peer reviews and design walkthrough in their product design processes, and were more likely to describe their design processes as model driven.  Laggards were more likely to say they had no product development processes.
• Were likely to have some budget for product developer tools, but were less likely to have large budgets.  Laggards were roughly twice as likely to have personal budgets of more than $25,000 for developer tools, than leaders.
• Tend to use licensed (not free) Linux when Linux is used. They were, for example roughly twice as likely to use MontaVista, or LynuxWorks Blue Cat than laggards. This is consistent with prior EMF findings that “free” tools usually cost more in the long run.

 Sign #4

Simulation-Modeling:

EMF’s report, 2009 Embedded Tools and RTOSes illustrates the growth of embedded market segments. RTOS/IDE has grown (through 2008) at a 11.2% CAGR while software tools grew at an 8.8% rate and the total embedded market grew at an 11% rate. Simulation-modeling grew at a substantially higher CAGR.

 Simulation Modeling (Model Driven Development – MDD – is the major component of this growth) is growing at a 17.5% CAGR based on 3 factors: 

• It provides an enhanced (and documented) ROI
• Systems and systems-within-systems represent the largest embedded growth market and require MDD
• Product line engineering is predicated on MDD

Have you noticed that we didn’t include any signs for EAL or MILS certification?

The Cheshire cat could have been talking to some embedded vendors. If you don’t  understand or track the broader marketplace and what your customers and potential customers are doing and experiencing,  then how can you possibly develop the best strategic plan?

Historically, new and more forceful markets that redefine economic demand replace markets that create economic downturns. Today we are at a transition point in our economic recovery that will redefine markets, and we are currently witnessing an irrevocable upheaval in the marketplace for software design and development tools, components and services. There will be winners and losers. How then does an embedded vendor mitigate against uncertainty and find direction? We believe that market intelligence is the antidote to market uncertainty.

Let me suggest from the get-go that the true value of market intelligence is to assist companies in recognizing emerging (and disruptive) markets as they begin to crystallize. I hope that the following will make my case and affect your thinking. Whether you are a developer, a manager or a company executive, the survival and growth of your company is a priority. It is sometimes difficult for those of us trained in science to relate to non-deterministic market dynamics – yet, like it or not, we live in a market driven world – not a technology driven one.
 
Competitive markets are defined and exploited by matching product features to customer requirements. What is seldom recognized is that when competitive products exceed what the market demands, customers can no longer base their choice on which is the higher performing product.
 
This is when disruptive technologies take root, only to emerge later when they completely change the product landscape. Remember Hayes modems? Remember DEC/Compaq/HP? Remember Motorola Computer Group? It happens to small companies all the time – we just don’t read about their demise. They all read a lot of self-serving market research (which validated their view of the industry – not the realities) yet they are all gone now.

I am shocked more than amused when vendors wishing for me to write favorably about what their product can do fail to answer my question of “what do your customers think they actually need?” I am usually told that “once they see how cool our product is they will change their minds” or “we ran this past a couple of our customers”. Many, many moons ago in a universe far away (that’s 1969 for you young-ins) I built several medical companies and I ran clinical research and marketing. Every time a physician told me what he/she wanted I asked for a PO. We got really good feedback that way.

I was recently at a breakfast which included very senior industry representatives and industry analysts. I was given the opportunity to set the initial discussion and I spoke to the challenges that major reductions in DoD discretionary funding would create within the mil/aero marketplace. I was politely ignored while others gushed about the exciting future for their industry –offering glowing growth estimates as high as 20%.
 
As I was leaving a kind gentleman approached me and said “Jerry, you won’t sell any of your research if you continue to be so negative”. I was very taken back by the remark which he intended as a helpful hint. The role of market intelligence is a two way street. The supplier and the user have to have the same goals and define the appropriate questions that need to be answered.

This brings me to the point of this discussion. The real question that market intelligence needs to answer is how can vendors of embedded products minimize risk while enhancing their prospects? And how does solid market information support this.

Harvard Business School professor Clayton Christenson in his landmark book “The Innovator’s Dilemma” first described and coined the term “disruptive technology” to describe a new technology that unexpectedly displaces an established technology. Christensen separates new technology into two categories: sustaining and disruptive. Sustaining technology relies on incremental improvements to an already established technology. Disruptive technology lacks refinement, often has performance problems because it is new, appeals to a limited audience, and may not yet have a proven practical application.

Linux was a classic example; when it first appeared it was a lower quality OS that was promoted as being free (the tools weren’t), addressed users that lacked substantial funding and had a total market value south of $5 million annually. What established vendor needing to grow their business by $20 million annually would consider such a technology – and given their cost and accounting structure, could not see a profit if they garnered the entire world market. A similar story is told regarding modems and the demise of world leader Hayes.

So how does a vendor measure the value of their product and how do they monitor the growth of potentially disruptive competitive technologies?

Be one an RTOS/IDE vendor, or one that markets communication middleware, modeling technology, management requirements or static analysis tools, etc. one must consider their current competitive market status.

1)      How does your product compare with the express needs of your current and potential customers? Does it contain competitive features that not many of your customers use?

2)      How does your closest competitor’s products compare with yours? Do both your products offer more features that your customer’s require? Do they only cover some of their needs?

When competitive market forces result in product features that surpass customer requirements, don’t expect them to make choices based on a comparative analysis. Understand that if you don’t have definitive ROI data based on comprehensive market intelligence surveys of appropriate users, then the basis of product choice evolves from functionality to reliability to convenience to price and your competition may very well come from initially underperforming products that are usually less expensive, lower quality, and initially lower gross margin.

By the time that your customers decide that they are more willing to use underperforming technologies rather than to pay a premium for features that they won’t use, you may find yourself far behind in the newer emerging market as these vendors have been upgrading the quality of their offerings while you have been battling your competitors for who has the best features.

Only companies that carefully measure trends in how their mainstream customers (or the industry at large) use their products will be able to catch the points at which the basis for competition will change in the markets they serve. 

Our subscribers use our data files and Dashboard to monitor trending, competitive use, design outcomes and ROI. When done on a year-over-year basis this is very effective. Annually we ask our subscribers to share with us questions of importance to clarify and track their markets and customer values.

We encourage you to use a reputable firm to gather and monitor market intelligence – and we suggest that you look for data handling capabilities that empower you to analyze the data yourself without limitations on the number of crosstabs you can personally perform.

 Given the monies that most firms spend on sales, lead generation, sales support and marketing, the cost of subscribing to a comprehensive market intelligence program is actually short money.

OK. Every meeting I go to I’m asked my opinion regarding Intel’s acquisition of Wind River – so here goes. To be honest, people richer than me are behind and part of this acquisition, so who am I to think that I know better? To be sure, some of the people richer than I am have screwed up on a far grander scale than I have, so why not offer my perspective?

My first response to the question is a flashback to the “Shortest Book” jokes that I used to revel in decades ago. “Ethics for Enron Executives” would certainly be among the “shortest book titles”. My contribution to this arcane joke file was “Famous Jewish Weightlifters” a title that my rabbi didn’t understand.

Recently I thought of adding “Successful Intel Acquisitions and Spin-offs” to the list. Those of you old enough to remember Dialogic and Ziatech (as well as a number of software-based hardware analysis and virtual development platform spin-offs) will get my point.

So what’s in it for Intel and why did Wind River bail out so easily? Let’s look at the facts:

• Intel doesn’t need the additional $300 million on their balance sheet
• Wind River has been losing market share for some time now and their “subscription model” (which they have abandoned) cost them dearly in the mil/aero marketplace and allowed Green Hills and LynuxWorks to significantly underbid them
• Rob Davidson, Wind River’s VP of A&D, has publicly stated (at an Intel sponsored event no less) “let me make it clear that Wind River is a Freescale company” since 85% of Wind River’s military sales are Freescale-based
• Intel, notwithstanding their successes, is facing considerable competition from ARM and FPGA vendors – particularly Xilinx
• The Mil/Aero market and the primes that spend the most on embedded technologies and sub-systems are decidedly moving to a “software centric” purchasing model

Has either of them thought through their respective channel strategies? Remember ages ago when Wind River bought out their arch rival ISI and acquired pSOS? The bloodletting that followed was inevitable given the incompatibility of their respective sales forces. A funny thing has since happened to Wind River, who appeared at that time to monopolize the embedded OS marketplace. Has Intel anything to lose by acquiring Wind River – will they incrementally increase their market share and influence within the embedded world.

Twenty years ago Intel and Microsoft ruled the embedded world (they were referred to as WinTel). Today, the parameters of the marketplace have changed significantly and here we see a hardware company attempting to insure their dominance by acquiring a high-profile software company. Had they merely wanted to increase their product offerings Intel could have bought LynuxWorks at a fraction of the cost of the Wind River acquisition.

Truth be told – both Intel and Wind River are missing the emergence of an embedded market that will soon overshadow the rest of the marketplace. It involves several components – none of which are within the new Intel’s capabilities.

• Systems development – with C and C++ developers working simultaneously on the same development
• Unique problems associated with systems-within-systems developments and long term support of such
• The new era of “Smart Devices” – the merging of IT database capabilities securely communicating with intelligent embedded devices

While Wind River, Green Hills and LynuxWorks are fighting for the limited and misunderstood mil/aero markets that demand DO-178B and MILS level security, operating systems such as ThreadX, Nucleus and Micrium along with MontaVista Linux are providing significant competition – particularly for consumer electronic and medical markets in which the selection of OS and processor to best match the product requirement is displacing the FUD surrounding the “suggested” need for excessively stringent security. ThreadX, Nucleus and MontaVista have hundreds of millions of applications running world wide and Micrium is already DO-178B certified.

EMF ROI analysis shows how these lesser known OSes are outperforming their larger and better known competitors. We don’t see how this helps Intel in the long run.

For the past 5 years, Wind River didn’t consider itself an embedded company but rather an enterprise company. This was a silly way to position themselves and they have paid the price for such folly. Embedded requirements far exceed the 5-nines requirements of enterprise developments.

Finally, over the ensuing three years, the DoD will reduce discretionary funding by between 18% and 40% depending on which reliable source one is using. This will create a huge hole in mil/aero OS markets and will seriously impact Wind River.  Maybe they went for the acquisition to come under Intel’s huge financial umbrella.

Sadly, like an All Pro quarterback whose arm has long ago lost its zip, the Embedded Systems Conference (ESC) returned to Boston the week of September 21, 2009. Once the “must attend event” of the fall season, this important venue continues to struggle with most of the major embedded vendors absent. Other company executives were present to speak with the press – but they didn’t host a booth. I was on a panel with Robert Day; VP marketing for LynuxWorks, who was in prominent attendance – but LynuxWorks didn’t have a booth.

Gone are most of the major chip companies. Only Microsoft, IBM Rational and Green Hills Software continue to host impressive booths. In year’s past Green Hills hosted a demonstration event in which was attended by a dozen or more viewers every hour. This year we saw only a handful of interested developers at each presentation

So what is it that accounts for this lack of participation? The decline began long before the economic crisis that we are experiencing – although the number of layoffs and the unwillingness of embedded OEMs and systems integrators to today invest in new technologies is certainly a contributing factor. In year’s past ESC Boston was a magnet for downscaled engineers to trot out their resumes. Given the severe downturn, many disenfranchised engineers didn’t see the point of pursuing a pointless effort.

It’s the leads stupid:  I believe that the Boston based ESC annual conference is a VERY important venue that is essential to the growth of the embedded industry. Even successful publications such as the ones the RTC Group provides openly hope for the success of the CMP Media sponsored event.

What sadly bring into question the future of the event are the discussions I have had with many vendors that I know well who continued to host a booth. In times of financial stress, discretionary spending becomes paramount. Vendors are focusing their strategies toward garnering qualified leads. We work with many of these companies to help them (through EMF market intelligence research) to identify their strengths and their competitive challenges in being able to not only gain qualified leads but to develop supporting documentation for sales support.

Many of these vendors expressed to me their frustration in the cost of exhibiting at ESC Boston compared with the number of qualified leads they get. Many have told me that with the minimal traffic and the indifference they found with those that stopped by their booth, that they had only a few interested leads – if any!

Given limited resources, they will need to invest their monies in areas in which they can get a significant return on their investment.

Sadly, the Boston event will continue to decline – although it would be difficult to find fault with CMP for this happening. On the bright side, ESC Silicon Valley 2010 in San Jose is only 6 months away.