Proceedings of the Standing Senate Committee on
Transport and Communications
Issue No. 19 - Evidence - June 6, 2017
OTTAWA, Tuesday, June 6, 2017
The Standing Senate Committee on Transport and Communications met this day at 9:30 a.m. to continue its study on the regulatory and technical issues related to the deployment of connected and automated vehicles.
Senator Dennis Dawson (Chair) in the chair.
[Translation]
The Chair: Welcome to this meeting of the Standing Senate Committee on Transport and Communications. This morning, the committee is continuing its study on connected and automated vehicles.
[English]
I am pleased to introduce our panel of witnesses from BlackBerry: Sandeep Chennakeshu, President, BlackBerry Technology Solutions, and John Wall, Senior Vice-President and Head of BlackBerry QNX.
[Translation]
Thank you for being here this morning and for meeting with us yesterday at your QNX facilities in Kanata. You can now give your presentation, and afterward, the senators will ask you questions.
[English]
Sandeep Chennakeshu, President, BlackBerry Technology Solutions, BlackBerry: Good morning, Mr. Chair and honourable senators, and thank you for inviting BlackBerry to speak to you today about connected and autonomous vehicles. I am the president of BlackBerry Technology Solutions, a division of BlackBerry. I am joined by John Wall, Senior Vice-President and Head of QNX Software Systems, a subsidiary of BlackBerry, which we refer to as BlackBerry QNX.
As this committee is aware, the automotive industry is undergoing a major transformation where a collection of computers, software, sensors and actuators and connected networks will eventually take over the driving function from humans. BlackBerry is playing a key role in this transformation.
BlackBerry QNX has been a trusted technology supplier to the automotive industry for approximately 20 years. Its software is used by over 40 automakers, is in over 60 million cars and will provide the foundation for autonomous drive systems.
Another BlackBerry subsidiary, Certicom, has been providing cryptographic solutions for automotive telematics for several generations of vehicles. The parent company, BlackBerry, has been in the business of managing mobile endpoints and securing mobile communications of data for almost the same amount of time and, in December of 2016, with Prime Minister Trudeau in attendance, launched the BlackBerry Autonomous Vehicle Innovation Centre at its Ottawa facilities.
This committee has heard from previous witnesses about the many potential benefits and risks that will come with the deployment of autonomous vehicles. One such issue is the safety and security of connected and autonomous vehicles.
The new generation of vehicles will increasingly be dependent on software and connections to external networks to perform critical functions. This will present increased safety risks if vehicle systems are not developed in accordance with best practices and industry standards for safety and security. BlackBerry has developed a framework of disciplines for securing modern cars to reduce the risk of cyber-attacks.
We work closely with auto makers and their suppliers and know that they are taking the issues of safety and security seriously. They are aware of the public's concerns and that failure to take reasonable measures to ensure the safety and security of vehicles will negatively impact the adoption of this technology, not to mention their reputations.
This is not to suggest that the government does not have an important role to play. Governments have a responsibility to ensure that the next generation of vehicles is safely deployed and does not introduce unreasonable risks. For example, best practices, standards and guidelines for the design and development of safe and secure vehicle systems must be promoted. Some of these best practices and standards already exist, while others will be newly created or adapted from those used in other sectors.
To the extent that regulation is necessary, government should endeavour to harmonize regulations across jurisdictions such that a patchwork of divergent laws and standards does not emerge. This will require coordination between multiple departments and levels of government, including foreign governments. The sharing of testing results, ideas and experiences amongst agencies and jurisdictions will also provide an efficient way for governments to keep pace with rapid technological advances and changes. The good news is that the federal government has already begun work in these areas. But more can be done.
In previous committee hearings, there has been discussion of who in government is taking or should take the lead. While divisions of power and departmental mandates must be recognized, it would be helpful to all stakeholders, including the public, if the behind-the-scenes coordination between federal government departments and the provinces and territories were more visible and that stakeholders could have a "one window'' or primary contact point with the government. For example, if industry wishes to engage with government on the issue of cybersecurity and connected and autonomous vehicles, where does it go? Is it Transport Canada? Is it Public Safety Canada? Is it somewhere else?
Second, while government is very receptive to hearing from industry, consultations are on an ad hoc basis. There does not appear to be an established mechanism or platform for ongoing industry and government consultation. Such a platform would help industry better understand the regulators' priorities, as well as to learn from industry how it is addressing these issues. Again, launching such a platform will require better coordination within government departments and amongst different levels of government.
To summarize, on issues such as safety and cybersecurity, industry and government both want to get it right. To do so will require a collaborative approach. We at BlackBerry are ready to assist.
This concludes my opening remarks. Mr. Wall and I am pleased to take any questions you have.
The Chair: Thank you very much for your presentation.
Senator Eggleton: Thank you very much for being here today. I am sorry I missed the tour yesterday of your facility in Kanata. I hope I am not repeating questions that my colleagues may have asked at that time.
I want to focus on privacy. The Privacy Commissioner was here and he noted the importance of privacy by design. That is, privacy needs to be considered in the design and architecture of these products right from the start. To what extent does BlackBerry QNX integrate privacy considerations into the design of its automotive products?
Mr. Chennakeshu: Essentially, managing privacy and privacy data is done by the automaker. They control it because they put together the whole system. We put together pieces where we provide components that can help you protect data. However, the ultimate policies and the configuration of how to protect the data that goes back and can be used is with the automaker.
Senator Eggleton: Well, that concerns me. It is shuffling around the responsibility here.
Let me ask you about the separation of privacy and security issues. We have had people make representations here that suggest that safety-critical and non-safety-critical systems should be separated. What is your view on that? What are you doing with respect to that?
Mr. Chennakeshu: This question has two parts, senator. One is the privacy of data; the other is security.
For security, we have promulgated seven pillars that we think are necessary in order to make the car really secure. It is not just one piece; it is a series of things that you need to do to make it secure. Isolation is one aspect, but it's not the only aspect. I can elaborate, if you want, about the seven pillars.
Senator Eggleton: Yes. I don't know if we have much time to do that now, though.
Mr. Chennakeshu: I can give you a high level view.
First, if you look at the auto ecosystem, it is a large number of suppliers. As a result, a vulnerability can be introduced anywhere in the supply chain. Step one is how you secure the supply chain, because it is very porous. That is the first pillar. We have techniques to secure the supply chain.
The second is: How do you use only hardened components? That is, components that are proven hardware and software applications? That is the second pillar.
The third pillar is isolation. How do you isolate safety-critical and non-safety-critical computers on isolated networks so that the communication between them is managed? There are two aspects. One is isolation on a network level; the second is isolation of messages that are passed between these computers.
The third aspect is what we call in-field health checks. Software evolves or morphs over time. People download applications. That can introduce a vulnerability. How do you scan code or software in the field to make sure a vulnerability has not been introduced? That is pillar number four.
Once you have identified these vulnerabilities, how do you make sure that you can patch them? This is pillar number five.
There is also an element where we would like to introduce the ability to train people in ensuring that they have the right culture so that they don't build products incorrectly. We have introduced and we have cultural training for our engineers on how to build safety products.
The last thing is that when you have a vulnerability, you need to have a crisis network to exchange vulnerability among all stakeholders so you can take immediate action. We call that crisis connect.
These are the seven pillars that we recommend as a framework in order to build such holistic cybersecurity solutions.
Senator Eggleton: It would be good if you could give us the longer description of all of that in writing.
Mr. Chennakeshu: Yes, absolutely.
Senator Eggleton: You said in your comments that you want one window in terms of primary contact with the government and that the government needs to work out who will be the lead player in that respect.We have also received the suggestion here at committee about a new agency being established specifically along these lines. Do you have any thoughts about that process?
Mr. Chennakeshu: No, I don't particularly. I will defer to John because he has been dealing with the government.
John Wall, Senior Vice-President and Head of BlackBerry QNX, BlackBerry: On this topic, from a cybersecurity perspective, there definitely needs to be more of a point of contact. I have been working with a number of government agencies on this topic, both provincially and federally. I believe that it is an important enough topic that, whether it is a different agency or a very specific point of contact, it would be beneficial.
The Chair: We have a long list of questioners so I will ask everyone for a bit of discipline.
Senator Bovey: I will be disciplined. Thank you for yesterday. It was an interesting tour.
The Chair: Don't take it personally.
Senator Bovey: I am picking up on what Senator Eggleton said. You said that the behind-the-scenes coordination should be more visible. You also talked about that being on a cross-country and international basis. It would be helpful, from your perspective, to get a sense as to how you would recommend that being put in place.
With what we have heard, I have been concerned about who is leading all of this. There seem to be a lot of leaders and followers, but who is tying it together? That is one point.
The other point is that you said you felt there should be a sharing of testing results. I am aware you work with many automakers and they all have their plans that they want to keep secret until they unveil them, but with the connected work you are doing, I would be interested in knowing what should that baseline be for sharing testing results and how should that happen.
Mr. Chennakeshu: That is a very good question, senator.
We think that at the base level, we get vulnerabilities all the time. We call them "common vulnerabilities and exposures.'' The acronym is CVE. We see a lot of these.
Each software developed by an automaker or their suppliers is slightly different, so the impact of a vulnerability could be slightly different on each part. Even if it is similar, it could be a digital instrument cluster in one vehicle versus another make, and they could be impacted slightly differently. At the base level, I think companies need to share what are the vulnerabilities, what did they see and how did it impact. With that knowledge, we can develop a lot of techniques to harden the entire system.
What we have to understand in a system of connected cars and infrastructure — because these autonomous cars will talk to each other and to infrastructure — is that if one of them gets infected, it can become viral. As a result, it is very important to share vulnerabilities amongst the group in order to take proactive action.
Senator Bovey: If you were writing regulations from the government's perspective, would that be one of the regulations you would put into the system?
Mr. Chennakeshu: I think I would have a guideline to say we need some sort of crisis connect network and people should subscribe to it. Today, we have things like Auto-ISAC where they share information, but not everyone is a member and there is no immediate response. It takes time between these committees. We would like it to be instantaneous. Even if you have a vulnerability, how do you tell? Today with mobile phones, when you have a vulnerability, it takes a month, 30 to 45 days, to get a patch. If you have a car out there with a vulnerability, would you want to wait 45 days? No. As a result, it would be better to share this information instantaneously.
Senator Saint-Germain: I see that you don't have your earphones. I will ask the question in French but, in the meantime, I want to apologize for not being able to attend yesterday. However, some ambassadors among the senators told me that it was outstanding, and I thank you for that.
[Translation]
I'm mainly concerned about the accessibility of cellphone service in the different regions of Canada. This service is critical to the ability to drive an autonomous vehicle. We've heard mixed views on the subject. We're told that short- range communications should play an important role in vehicle-to-vehicle and vehicle-to-infrastructure communications. However, some experts say that it would be a temporary measure. Dedicated short-range communications may be replaced by 5G cellular technology.
I'm concerned about both the infrastructure's accessibility and the appropriateness of financial investments, in particular taxpayer-funded investments. What do you think of the co-existence of these two technologies, and of the game plan, in financial terms, for eventually making cellphone service accessible and cost efficient?
[English]
Mr. Chennakeshu: Again, that is a very good question. Industry is debating what is the optimal wireless technology to use. I think there won't be just one technology; there will be a number of technologies.
I think that, for a wide area of coverage, you absolutely need cellular technology. Then you need to get a centimetre- level accuracy on location. That can be provided. You can't do it with normal GPS; you have to do it with augmented GPS. That would be another wireless technology that is needed. That will probably be provided by satellite. Then there is short-range technology that will be needed, for example, talking from one car to another. That could be DSRC.
I think there'll be a combination of technologies. That is why we believe that autonomous cars, at level 4, level 5, which is fully autonomous driving, are going to take some time. It will not come overnight. We think it will take time before you get full coverage across countries to make it really practical. That investment needs to be made if such vehicles have to really be put on the road.
I think it should start with cities and then gradually expand to the other areas, because it is much easier to cover local areas than very wide areas.
Mr. Wall: To add to that, this is an open debate. A lot of the Europeans are waiting for 5G. In the U.S., they are going with DSRC. In speaking to a lot of the carriers like AT&T and whatnot, they believe the future's going to be a hybrid. There will be short-distance wireless. There'll be 5G. The hope is that 5G will provide more ubiquitous coverage. As Sandeep mentioned, these are some of the challenges that we believe are going to push autonomous drive, in a very large-scale way, many years down the road.
Senator Saint-Germain: Do you have an opinion about the sharing of the funding of both the development and the implementation of these technologies? Who should pay? Who should get the benefits? Is it the user?
Mr. Wall: The taxpayer.
Senator Saint-Germain: The taxpayer?
Mr. Wall: Well, it is government infrastructure.
Mr. Chennakeshu: Typically, from what we have seen, I think it should be very similar to how we have done it over the years. I come from the telecom industry. I've spent a very long time in Ericsson. I used to be their chief technology officer for the mobile phone group, and I was involved in developing seven to nine standards around the world. There are large telecom companies that have to make the investment to build the equipment, and the chip companies need to build the right type of chips. The handset or the co-equipment companies need to make the right investment in R&D to be present because they also make profits from this.
Then the government's involvement is allocation of spectrum. That is where, basically, they need to make sure that the right spectrum is allocated so that we can get that coverage so that it's not piecemeal and fragmented to make it too expensive.
Senator Mercer: I would like to thank the people who did the presentation yesterday. I did make the trip to Kanata yesterday. It was excellent, and I learned a lot. But you have raised a lot of questions that are still lingering.
It is now 2030. We have now had a lot of cars on the road that are equipped in this way. I want to talk about two different vehicles from two different manufacturers. They are different models. They have different software in them from different companies, but they have the same problem. They go to the same mechanic, who is not the dealer that they bought their vehicle from. How does he or she get their car fixed?
We have a situation that has developed in other industries, for example, in the agricultural sector, where there is a large manufacturer of agricultural equipment. In the sale of equipment, they have written into the sales agreement that you must have your service done at the dealership of that manufacturer. Well, that is not available to everybody everywhere their equipment is. Farmers, being the most innovative people in the world, have found a way around it, but it may end up screwing up their warranty for their vehicles.
How is that going to work when you have software? This is a software-driven industry that we are talking about now. It is not about the size of the engine; it is about how good your software is more than anything. How is this going to work for the consumer? I have a problem; Senator Dawson has a problem with his vehicle. We bought it from different manufacturers. How does that help?
Mr. Chennakeshu: It is an excellent question. We have actually thought through this quite carefully because there are actually two problems to solve here. I will address the problem of how to fix it, and then I will go back to a more fundamental problem.
You have two manufacturers, and they could have a similar function but different software. If something goes wrong, how do you fix it? It is already starting now; we are going to have diagnostic elements in the software that report what is happening to the piece of software continually. We monitor that on dashboards, and the minute we find any form of malfunction in a predictive analytics manner, we would download new software into the car over the air and correct it. Going to a mechanic will actually be very old-fashioned. It will be completely done over the air. Already, we are finding that many manufacturers are working with either us or others to completely cure the software problem by just downloading it. This is called over-the-air software updates.
But there is a more fundamental issue. Let's take an autonomous car. You have an autonomous car that has to sense its environment. Then it has to understand the environment in context, which is what we do with our brain, and then we take an action. For an autonomous car, if two manufacturers have two different policy engines and one decides that there is a baby and that it has to turn right and the other determines that it has to turn left, there is a problem. Identifying certain policy criteria of how to take care of contextual issues and solve them is going to be a much more tricky problem.
Senator Mercer: That goes right to this point. Yesterday, when we were out for a visit, we did talk about OTA, over- the-air software delivery and updates. That works if you have the proper spectrum to deliver it, if there is the ability for you to deliver that. Currently, we don't have the ability to deliver that coast to coast to coast. This is the issue for the consumer. I will not buy a car that I don't know that I can have repaired. In this case, in the new versions, if we are to have the over-the-air updates, I don't know that you can do that. You can't reach me because I live in a rural part of Nova Scotia. I do have reasonably good service because I am not that far from the biggest city, but many people in this country live in rural and remote areas. How do they get that kind of service? Do they have to drive to the city to get the update?
Mr. Chennakeshu: Actually, there are a couple of ways to solve that, senator. It is not necessary that it is cellular, where you might not have coverage in rural areas. That is completely correct. However, we can also deliver the software through Wi-Fi. So you can develop it through the Internet, onto Wi-Fi, and into your garage. That's another way to do it. The third way — and we are already talking to satellite providers about this — is to see if satellite providers can provide it. So there are three ways to do it. Then, of course, there is the traditional way. A mechanic can drive up with a box and, within short range, update it.
Senator Mercer: I'll raise one more quick question. If you're going to do that, you've now introduced a security issue for yourself. If you are sending it by satellite or by some other method, your competitor or someone else can be intercepting it.
Mr. Chennakeshu: No, I don't think that's possible. Our OTA system is super secure. The way we do it today is that we take the software package, and every block of software is signed. Then the manifest containing the whole package is also signed, and that is sent over to the car and it's only unpacked by the car. The car has to have special keys and certificates before it can even open this. But then even if the car opens it, it can't do anything. We send a completely separate control command, which is also encrypted and which is 128 times more secure than a banking transaction, and that is used to execute the command. The probability of someone intercepting both, and decoding, is highly improbable.
Senator Mercer: Don't you love it when they always have the answers?
Mr. Wall: To the example we spoke about yesterday, farm equipment, what Sandeep talked about in terms of securing the supply chain would make it impossible for somebody to repair the vehicle without authorized parts.
Senator Galvez: Thank you very much for a very interesting presentation and a very interesting visit yesterday. Part of me is thrilled and fascinated with the technology and all the futuristic images of automated cars on the highways. However, when I wear my hat of senator, I want to think more as a citizen, taxpayer and user.
Right now we have a very inefficient transport system. Cars are in the backs of our garages 90 per cent of the time. When they circulate, there is three-quarters empty space there. Then there is the congestion. Everybody drives their car at the same time, during rush hour. We use fossil fuel, which is not good for the air or for anybody.
How will this technology solve these social problems? So far, I don't like the answers we have been hearing. They are telling me that there will be just one person sitting there, so the congestion will continue. It's going to be very expensive, because we are talking about 5G satellite interconnection in the car, and we will need infrastructure. But if there continues to be a single passenger, and it will be my mother or another mother — that is what we heard, that it's the mother of somebody who will be in the back seat — this person won't have the money to pay for all of this.
I know you cannot answer completely, but how do you see this? What is your view?
Mr. Chennakeshu: I can only agree with you that initially such technology is expensive; however, in time, it always comes down. When I was building 3G, I actually spent US $1 billion before I brought out my first chip. Everybody thought, of course, I was nuts. The first chip I did cost $60, but within five years it was down to $15 and today it's down to $5.
What happens is that, with time, innovation actually brings down that cost curve. So I do believe that, with time, this will become much more efficient because there will be a lot more users and applications, which will offset some of the cost of cars. This technology won't just be used for cars, because we are using the same type of sensors for other things. So, in time, that cost will come down.
The biggest benefit I see with autonomous vehicles is an improvement in productivity. The average American is spending one to one and a half hours on the road. There is a tremendous amount of congestion, which means cars idling, and idling means pollution and consumption of fossil fuel.
There are accidents. A lot of accidents are due to distraction, people being tired. The cost of human life is important.
When you do the entire equation, you actually find there are tremendous savings, and I believe that these costs will come down with utilization.
A lot of people who don't know how to drive today or who can't drive, such as senior citizens, would have the ability now to use transportation. I think that, with time, this problem will be solved, like we have solved many other problems. But it will take a concerted effort and research.
Senator Galvez: On another aspect: You are not a public company?
Mr. Wall: We are.
Senator Galvez: So you get funding from outside?
Mr. Chennakeshu: No. We are a public company. We fund our own R&D today. Of course, we would be considering other investments, but today we are completely self-funded and we run it based on the earnings we make.
Senator Galvez: The reason I'm asking that is because of the intellectual property. What you are developing in terms of security is so important to keep. By sharing or diffusing it, do we bring down the level of security?
Mr. Chennakeshu: There are two very good questions embedded in your question, senator. The first one is intellectual property. It's important for us to maintain our intellectual property, because that is the only deterrent to keep the costs down.
For example, today what is happening with cellphones is that when people ask me to pay royalties because I'm using their technology, if I don't have intellectual property myself, I just have to pay. But if I have intellectual property, then its détente, so nobody pays. That's very important in keeping the costs down. It's important for me to keep my intellectual property file patents and to be in a position of strength. That's important for a Canadian company.
The second question you had was about diffusing our knowledge. For example, when John provides his software to companies, we don't provide it in source code format. We only provide it as object code, so they have no way of reverse engineering that code.
We also have special techniques where we can expose some of our security, but we only expose the public portion, not the private portion. So you can never reverse engineer. It's like a trapdoor — easy to go one way but difficult to go the other way.
Mr. Wall: This is important when you hear about Linux and Android. These will never be in safety-certified and secure systems because there are so many back doors and so many ways of getting the source code; it's fragmented. So we are very careful with our intellectual property that, in order to maintain the highest levels of security, we do not share our source code, with no exceptions.
[Translation]
Senator Dagenais: Mr. Chennakeshu, I found your presentation fascinating. Thank you. It took me back to the 1950s and 1960s, when my father and I drove with a roadmap on hand. Nowadays, we have cars equipped with GPS devices. You're saying that, in the 2030s, we'll have access to autonomous cars. In 2030, I'll be 80 years old, and I hope that I'll have an autonomous car. I'll program the car to drop me off at the grocery store. It's so perfect. Even if I lose my driver's licence for health reasons, I'll have my autonomous car. Please, keep researching! I encourage you to do so. If I can help you in any way, let me know.
Sorry, Mr. Chair. I got a little emotional. I feel better now that I've expressed myself.
I'll ask my question now. You must have competitors in the automotive technology development sector, and I imagine that you know them. What benefits do you offer car manufacturers? I also imagine that Chinese and Korean companies aren't far behind. How will car manufacturers benefit from using your application?
[English]
Mr. Chennakeshu: We have, fundamentally, two major advances. One is we probably have the best operating system, which is safety certified. It's not just safety certified. It also gives you very high reliability and security. We have built this over 20 years and have experience of working with car manufacturers, and there is a reason they trust us. That's a very strong foundation. There may be two or three companies in that space, and, clearly, we seem to be gaining on that competition.
The second very important aspect is security. We are, fundamentally, a company that has security built into our DNA. We are very holistic. We don't look at security from one single aspect. We talked about the seven pillars of cybersecurity. I would dare say there is not a single company on the planet that has all seven elements. We think we can bring together something that is fundamentally differentiating and that nobody else can have.
That also means that in order to maintain our lead, we need to keep investing. This does not mean a massive amount of investment. This can mean, over five years, $30 million to $50 million or $70 million. It doesn't mean more than that. But we absolutely have to.
We have to be very good at encouraging our universities to produce students who can come and join us, because unless those students are excited and continue to be educated to maintain this leadership, we will lose the leadership.
We can never get complacent. I have seen this happen over years. I'm getting to be old, too. We have to invest in the universities to produce the talent to hand the baton over to in this relay race to the next generation to continue what we are doing.
[Translation]
Senator Dagenais: Apple, which could be your competitor, is extremely rich. Where do you stand in relation to Apple?
[English]
Mr. Chennakeshu: Apple is a consumer company, and they are good at what they do. They have very deep pockets, but they haven't really worked in the type of things we do. Can they compete? Absolutely. We have a head start, and we have to keep that head start.
Mr. Wall: One point of clarification: You mentioned Koreans, Japanese and other companies. We are not competing with any car company. We are actually providing the foundational software to all the car companies. It could be Google with Waymo; it could be Tesla; it could be the new EVs in California; it could be the traditional OEMs.
Our value add is providing the foundational real time operating system and all the security assets of BlackBerry.
Apple, in particular, have not sold their software in the way that we sell our software. They produce products: cellphones and whatnot. Like Sandeep said, they are very capable. It will be interesting to see where they go.
Senator Griffin: I really like your recommendations. One is better coordination among government departments, both federal and provincial, in terms of discussion of behind-the-scenes coordination. I like your one-window approach for ongoing industry and government consultation. I think those are great recommendations. Of course, the federal government is already involved in the innovation supercluster funding.So all good things, but keeping in mind that government has both the economic as well as the policy regulation instruments at its hands, are there any other things you would like to see done in order to advance your industry?
Mr. Chennakeshu: I think encouraging our universities and investing in them to produce students, investing in Canadian companies to actually get to the forefront, to develop intellectual property so we are not comprised in terms of our ability to compete, and to invest in Canadian companies to actually get to the forefront would be very important, because the bigger companies with the deeper pockets can make more investments, and we have to keep pace. That would be extremely beneficial for us and other Canadian companies to compete.
Senator Griffin: Thank you, that's great.
I'd like to follow up on Senator Mercer's question. I'm interested in the aftermarket, the local Canadian Tire. What are they going to be able to do with vehicles in the future besides change the tires? Are they going to be able to actually do any repairs on vehicles? Will you be sharing information with them that would be enabling these other aftermarket operators to remain in business?
Mr. Chennakeshu: This comes back to a question of privacy. Senator Eggleton basically raised this.
For example, we build an IoT system, Internet of Things system, which is a platform or cloud. In that system, every bit of data has a little tag, a permission tag, and depending on who you are, you can see the data or not see the data. We can filter based on these tags.
Sharing of information has to be very managed because we need to know who needs to know what, and just enough, because you don't want everyone to have all the information because that can be a potential vulnerability.
We think that we can share information, yes, and based on that information, they can actually improve the quality of their products.
Let me say that you can do clever things with tires. You can have tire pressure monitoring systems that can get more sophisticated, and you could figure out whether the tire pressure is within a certain limit. Is the tire pressure on the four wheels exactly the same or are they imbalanced? All of this data can be fed back. We could do very clever things to take preventive action and say, "You know what? Three of your tires are okay; one is incredibly low. Be careful.''
There are things we can actually do to help. I think the biggest thing with the aftermarket is how do you protect aftermarket components from actually introducing a vulnerability? This is why we have come up with this concept that every chip in every aftermarket device has to have a key and a certificate called a birth certificate. When it is installed, that birth certificate is like your passport or your driver's licence; it has to be issued by some authority. We call it a root authority. In our case, it's the government, but the root authority, you have to determine that that part was qualified and everything in it checked; otherwise, it will be rejected and never be allowed.
Mr. Wall: That means Canadian Tire could still repair your vehicle, but they would have to be getting their parts from a trusted vendor that could provide the parts that have the right authentication to be used in the system. This would damage aftermarket companies that build clone parts. They wouldn't be able to do that because these would have to be aftermarket parts that have the authority from the OEM.
The Chair: I think Canadians have every reason to be proud of BlackBerry in the past, and you are certainly giving us a lot of pride for the future. That being said, you also talked about trust.Using the example of Volkswagon, they betrayed that trust by fundamentally lying to consumers about their product. We all know that banks do not always tell us about the vulnerability of banks because they are protecting their reputation. They don't want to know how many people got their cards swiped last year because it affects the value of the shares of the banks. That trust has to be earned, and to keep it, they use privacy, as in the case of Volkswagen and the banks.
My colleague, who will be asking a question after me, said it's nice that you are protecting your interests but that will put you in a situation of a quasi-monopoly, in a sense, because no one will be able to get a spare part that will be costing less because they're allowed to produce it for less. They will have to buy it from you. Those are questions that could be worrisome for the consumers who are being told they will not be able to find a competitive product.
Mr. Chennakeshu: We are not saying you have to buy anything from us. We will provide component technologies that any manufacturer can use to protect their part. We are only providing mechanisms so that any part can use our technology to protect themselves. There will be complete choice. We do not want to raise the cost of the system but lower it by encouraging free competition. Whoever is out there, whatever way you design it, if you follow these principles and use these technologies, you will be protected.
The Chair: Does that include doing it in a transparent environment?
Mr. Chennakeshu: Yes.
Mr. Wall: It means, for instance, that the OEM will be able to authenticate parts that are put into their car by using our technology.
Senator Mercer: Thank you. I was not going to ask a question on the aftermarket parts, but I agree with Senator Dawson. There is a concern that this is going to drive up the price of aftermarket parts, and you need to continue to address it.
I will go back to my original question on the other concern that we've had expressed to us by members of the public and members of the industry. It's an employment question about where the employment opportunities are as we move forward and as we retool the industry. We are moving away a bit from the mechanic that works on my car to someone who is much more technically knowledgeable. I think our universities are probably doing okay, but are our community colleges ready to deal with how they train mechanics in the future? Are they ready to talk about dealing with other people who are going to be servicing the industry, and are we keeping pace?
I tell a story about community colleges because I think they have the best ability to respond quickly to changes in the market, and I will go off topic for one second. When the government announced that Irving shipyards in Halifax was going to get the large 30-year contract to help rebuild the Canadian navy, the next day the community college in Nova Scotia was meeting and talking about how to change what they're doing to meet the needs that the shipyard will have. They can do that. Universities take much longer to do that.
What are the opportunities there? I assume there are some in R&D, but is the industry ready to train teachers and administrators of universities how to respond to the new needs that this new technology will drive?
Mr. Chennakeshu: I agree, senator. I think that there needs to be a major retooling because if you look at what has happened, a few years ago a car had maybe tens of millions of lines of software. Today, a reasonably premium car has between 100 and 200 million lines of software, and that is more than the Boeing 787, the F22 and Microsoft windows put together. It's massive, so software is a control point and differentiator. Getting to understand software will be important.
The second thing that will be important is machine learning. What happens is they have what they call deep neural networks, and these neural networks try to mimic the brain. They are simple processors that they feed tonnes and tonnes of routine data, and they train it. They keep adjusting some weights on these little machines until it ultimately learns, and you can teach it to paint a Picasso. You can teach it to do anything. Instantaneous language translation will come about. Massive compute farms are being built to do this.
What we have to do is actually start retraining everyone — the younger generation — to get very proficient in this so that the whole economy drives this, exactly as you said, and unless we start reacting now, we won't be ready in 10 years.
Senator Bovey: On the same line, I was at a Universities Canada conference this morning before this meeting where we were talking about intellectual property and digital issues. There were some very exciting presentations. And I loved one of their lines that, yes, these new technologies are destructive but they are providing new contexts, and I'm interested in the new context side and moving forward.
But coming back to curriculum, you mentioned universities, and we know that education is provincial. We know that our school curriculums for little tykes are developed by provinces. My concern is that this is no longer provincial. We have to find a way, and I'd be interested in your thoughts, at the university and community college levels, going right down into elementary school. My grandchildren are learning times tables the way I did. That will not tool them for these new contexts; it will tool them for destruction. From your perspective, are you able to interface with every level of education and let them know what your needs and criteria are going to be so the curriculum can be developed and taught?
Mr. Chennakeshu: Actually, sadly, no, because I would love to do something like that, but I think about just actually staying afloat in our business trying to turn things around. I think that we have to have people from industry present problems and get that distilled down into concrete little projects. The only way you really learn is to do projects. Every high school has a robotics team. What if they started working on this? These students are extremely clever. From when I went to college and what I see my son doing today at 19 is dramatically different. They are 10 years more advanced than I was, and the next generation has to be 10 years more advanced, so we have to allow that to percolate down. I don't think that we as an industry do enough of that.
Mr. Wall: I think that at the high school level, definitely there is not enough being done, and especially enough to encourage girls to get into this industry. I see it. I have three daughters, and they've all finished high school with no interest in this area at all and are just starting to have interest in university.
At the university level, we have observed from a computer science perspective that a lot of emphasis has been on web technology because the cellphone and the Internet have been so predominant. The reality of the type of software we develop is deeply embedded software, and the world is headed toward more deeply embedded software in your fridge, toaster and everything you do in life. The universities don't seem to be focused on embedded software, and industry has to help guide some of these.
Senator Bovey: Yesterday's tour was really great. I'm sure you don't have the time or the opportunity, but I would hope that all sorts of schools of all levels are able to come into that room of yours because it will inspire imagination, and I think that's what we need.
[Translation]
The Chair: I want to thank the BlackBerry representatives for participating today.
[English]
Honourable senators, for our meeting tomorrow night, we will hear from the representatives of the Canadian Vehicle Manufacturers' Association and from General Motors. We have some good questions that have been raised.
(The committee adjourned.)