tag:theconversation.com,2011:/fr/topics/black-box-9652/articlesBlack box – The Conversation2023-05-22T12:27:06Ztag:theconversation.com,2011:article/2038882023-05-22T12:27:06Z2023-05-22T12:27:06ZWhat is a black box? A computer scientist explains what it means when the inner workings of AIs are hidden<figure><img src="https://images.theconversation.com/files/527116/original/file-20230518-29-egvjik.jpg?ixlib=rb-1.1.0&rect=0%2C20%2C4500%2C2970&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">You can't see inside any opaque box, but the color black adds an air of mystery.</span> <span class="attribution"><a class="source" href="https://www.gettyimages.com/detail/photo/black-box-levitation-on-black-background-3d-royalty-free-image/610655646">chingraph/iStock via Getty Images</a></span></figcaption></figure><p>For some people, the term “black box” brings to mind the recording devices in airplanes that are valuable for postmortem analyses if the unthinkable happens. For others it evokes small, minimally outfitted theaters. But black box is also an important term in the world of artificial intelligence. </p>
<p>AI <a href="https://www.techtarget.com/whatis/definition/black-box-AI">black boxes</a> refer to AI systems with internal workings that are invisible to the user. You can feed them input and get output, but you cannot examine the system’s code or the logic that produced the output. </p>
<p>Machine learning is the dominant subset of artificial intelligence. It underlies generative AI systems like <a href="https://openai.com/blog/chatgpt">ChatGPT</a> and <a href="https://openai.com/product/dall-e-2">DALL-E 2</a>. There are three components to machine learning: an algorithm or a set of algorithms, training data and a model. An algorithm is a set of procedures. In machine learning, an algorithm learns to identify patterns after being trained on a large set of examples – the training data. Once a machine-learning algorithm has been trained, the result is a machine-learning model. The model is what people use. </p>
<p>For example, a machine-learning algorithm could be designed to identify patterns in images, and training data could be images of dogs. The resulting machine-learning model would be a dog spotter. You would feed it an image as input and get as output whether and where in the image a set of pixels represents a dog.</p>
<p>Any of the three components of a machine-learning system can be hidden, or in a black box. As is often the case, the algorithm is publicly known, which makes putting it in a black box less effective. So to protect their intellectual property, AI developers often put the model in a black box. Another approach software developers take is to obscure the data used to train the model – in other words, put the training data in a black box.</p>
<figure>
<iframe width="440" height="260" src="https://www.youtube.com/embed/Q6JbmGQstDM?wmode=transparent&start=0" frameborder="0" allowfullscreen=""></iframe>
<figcaption><span class="caption">Black box algorithms make it very difficult to understand how AIs work, but the situation isn’t quite black and white.</span></figcaption>
</figure>
<p>The opposite of a black box is sometimes referred to as a <a href="https://www.tutorialspoint.com/software_testing_dictionary/glass_box_testing.htm">glass box</a>. An AI glass box is a system whose algorithms, training data and model are all available for anyone to see. But researchers sometimes characterize aspects of even these as black box. </p>
<p>That’s because researchers <a href="https://www.vice.com/en/article/y3pezm/scientists-increasingly-cant-explain-how-ai-works">don’t fully understand</a> how machine-learning algorithms, particularly <a href="https://www.pcmag.com/news/what-is-deep-learning">deep-learning</a> algorithms, operate. The field of <a href="https://theconversation.com/how-explainable-artificial-intelligence-can-help-humans-innovate-151737">explainable AI</a> is working to develop algorithms that, while not necessarily glass box, can be better understood by humans.</p>
<h2>Why AI black boxes matter</h2>
<p>In many cases, there is good reason to be wary of black box machine-learning algorithms and models. Suppose a machine-learning model has made a diagnosis about your health. Would you want the model to be black box or glass box? What about the physician prescribing your course of treatment? Perhaps she would like to know how the model arrived at its decision. </p>
<p>What if a machine-learning model that determines whether you qualify for a business loan from a bank turns you down? Wouldn’t you like to know why? If you did, you could more effectively appeal the decision, or change your situation to increase your chances of getting a loan the next time.</p>
<p>Black boxes also have important implications for software system security. For years, many people in the computing field thought that keeping software in a black box would prevent hackers from examining it and therefore it would be secure. This assumption has largely been proved wrong because hackers can <a href="https://www.merriam-webster.com/dictionary/reverse%20engineer">reverse-engineer</a> software – that is, build a facsimile by closely observing how a piece of software works – and discover vulnerabilities to exploit. </p>
<p>If software is in a glass box, then software testers and well-intentioned hackers can examine it and inform the creators of weaknesses, thereby minimizing cyberattacks.</p><img src="https://counter.theconversation.com/content/203888/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Saurabh Bagchi receives research funding from a large number of sources, federal government, state government, and private enterprises. The full list can be seen from his CV at:
<a href="https://bagchi.github.io/vita.html">https://bagchi.github.io/vita.html</a>
Bagchi is an office bearer of IEEE Computer Society. He is the co-founder and CTO of a cloud computing startup, KeyByte. </span></em></p>Metaphorical black boxes shield the inner workings of AIs, which protect software developers’ intellectual property. They also make it hard to understand how the AIs work – and why things go wrong.Saurabh Bagchi, Professor of Electrical and Computer Engineering, Purdue UniversityLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/1963922023-01-26T19:04:51Z2023-01-26T19:04:51ZPhilosophers have studied ‘counterfactuals’ for decades. Will they help us unlock the mysteries of AI?<figure><img src="https://images.theconversation.com/files/502066/original/file-20221220-25-sbg2ya.jpeg?ixlib=rb-1.1.0&rect=26%2C44%2C5964%2C2757&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">Counterfactuals are claims about what would happen, were something to occur in a different way. For instance, we can ask what the world would be like had the internet never been developed. </span> <span class="attribution"><span class="source">Shutterstock</span></span></figcaption></figure><p>Artificial intelligence is increasingly being rolled out all around the world to help make decisions in our lives, whether it’s <a href="https://www.afr.com/companies/financial-services/banks-warned-using-ai-in-loan-assessments-could-awaken-a-zombie-20210615-p5814i">loan decisions by banks</a>, <a href="https://www.healthcareoutlook.net/top-10-ai-based-medical-diagnostic-tools-to-try-in-2022/">medical diagnoses</a>, or US law enforcement predicting a <a href="https://www.technologyreview.com/2019/01/21/137783/algorithms-criminal-justice-ai/">criminal’s likelihood of re-offending</a>. </p>
<p>Yet many AI systems are black boxes: no one understands how they work. This has led to a demand for “explainable AI”, so we can understand <em>why</em> an AI model yielded a specific output, and what biases may have played a role.</p>
<p>Explainable AI is a growing branch of AI research. But what’s perhaps less well known is the role philosophy plays in its development.</p>
<p>Specifically, one idea called “counterfactual explanation” is often put forth as a solution to the black box problems. But once you understand the philosophy behind it, you can start to understand why it falls short.</p>
<h2>Why explanations matter</h2>
<p>When AI is used to make life-changing decisions, the people impacted deserve an explanation of how that decision was reached. This was recently recognised through the European Union’s <a href="https://gdpr.eu/what-is-gdpr/">General Data Protection Regulation</a>, which supports an individual’s right to explanation.</p>
<p>The need for explanation was also highlighted in the <a href="https://theconversation.com/robodebt-was-a-fiasco-with-a-cost-we-have-yet-to-fully-appreciate-150169">Robodebt case in Australia</a>, where an algorithm was used to predict debt levels for individuals receiving social security. The system made many mistakes, placing people into debt who shouldn’t have been. </p>
<p>It was only once the algorithm was fully explained that the mistake was identified – but by then the damage had been done. The outcome was so damaging it led to a <a href="https://www.abc.net.au/news/2022-12-16/qld-robodebt-scheme-government-royal-commission-victim/101780890">royal commission</a> being established in August 2022.</p>
<p>In the Robodebt case, the algorithm in question was fairly straightforward and could be explained. We should not expect this to always be the case going forward. Current AI models using machine-learning to process data are much more sophisticated.</p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/not-everything-we-call-ai-is-actually-artificial-intelligence-heres-what-you-need-to-know-196732">Not everything we call AI is actually 'artificial intelligence'. Here's what you need to know</a>
</strong>
</em>
</p>
<hr>
<h2>The big, glaring black box</h2>
<p>Suppose a person named Sara applies for a loan. The bank asks her to provide information including her marital status, debt level, income, savings, home address and age. </p>
<p>The bank then feeds this information into an AI system, which returns a credit score. The score is low and is used to disqualify Sara for the loan, but neither Sara nor the bank employees know why the system scored Sara so low. </p>
<p>Unlike with Robodebt, the algorithm being used here may be extremely complicated and not easily explained. There is therefore no straightforward way to know whether it has made a mistake, and Sara has no way to get the information she needs to argue against the decision.</p>
<p>This scenario isn’t entirely hypothetical: loan decisions are likely to be outsourced to algorithms in the US, and there’s a real risk <a href="https://hbr.org/2020/11/ai-can-make-bank-loans-more-fair">they will encode bias</a>. To mitigate risk, we must try to explain how they work.</p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/everyones-having-a-field-day-with-chatgpt-but-nobody-knows-how-it-actually-works-196378">Everyone's having a field day with ChatGPT – but nobody knows how it actually works</a>
</strong>
</em>
</p>
<hr>
<h2>The counterfactual approach</h2>
<p>Broadly speaking, there are <a href="https://news.mit.edu/2022/explained-how-tell-if-artificial-intelligence-working-way-we-want-0722">two types of approaches</a> to explainable AI. One involves cracking open a system and studying its internal components to discern how it works. But this usually isn’t possible due to the sheer complexity of many AI systems.</p>
<p>The other approach is to leave the system unopened, and instead study its inputs and outputs, looking for patterns. The “counterfactual” method falls under this approach. </p>
<p>Counterfactuals are claims about what would happen if things had played out differently. In an AI context, this means considering how the output from an AI system might be different if it receives different inputs. We can then supposedly use this to explain why the system produced the result it did. </p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/502067/original/file-20221220-16-9qsxly.jpeg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/502067/original/file-20221220-16-9qsxly.jpeg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/502067/original/file-20221220-16-9qsxly.jpeg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=399&fit=crop&dpr=1 600w, https://images.theconversation.com/files/502067/original/file-20221220-16-9qsxly.jpeg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=399&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/502067/original/file-20221220-16-9qsxly.jpeg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=399&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/502067/original/file-20221220-16-9qsxly.jpeg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=501&fit=crop&dpr=1 754w, https://images.theconversation.com/files/502067/original/file-20221220-16-9qsxly.jpeg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=501&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/502067/original/file-20221220-16-9qsxly.jpeg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=501&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">One example of a counterfactual would be to ask what the world might be like had the internet never been developed.</span>
<span class="attribution"><span class="source">Shutterstock</span></span>
</figcaption>
</figure>
<p>Suppose the bank feeds its AI system different (manipulated) information about Sara. From this, the bank works out the smallest change Sara would need to get a positive outcome would be to increase her income. </p>
<p>The bank can then apparently use this as an explanation: Sara’s loan was denied because her income was too low. Had her income been higher, she would have been granted a loan.</p>
<p>Such <a href="https://jolt.law.harvard.edu/assets/articlePDFs/v31/Counterfactual-Explanations-without-Opening-the-Black-Box-Sandra-Wachter-et-al.pdf">counterfactual explanations</a> are being <a href="https://www.abc.net.au/news/2022-12-12/robodebt-algorithms-black-box-explainer/101215902">seriously considered</a> as a way of satisfying the demand for explainable AI, including in cases of loan applications and using AI to make <a href="https://www.chemistryworld.com/news/molecular-counterfactuals-method-helps-researchers-explain-ai-predictions/4015381.article">scientific discoveries</a>. </p>
<p>However, as researchers have argued, the counterfactual approach is <a href="https://arxiv.org/abs/2103.04244">inadequate</a>. </p>
<h2>Correlation and explanation</h2>
<p>When we consider changes to the inputs of an AI system and how they translate into outputs, we manage to gather information about correlations. But, as the old adage goes, correlation is not causation. </p>
<p>The reason that’s a problem is because work in philosophy suggests causation is <a href="https://www.jstor.org/stable/2214338#metadata_info_tab_contents">tightly connected to explanation</a>. To explain why an event occurred, we need to know what caused it.</p>
<p>On this basis, it may be a mistake for the bank to tell Sara her loan was denied because her income was too low. All it can really say with confidence is that income and credit score are correlated – and Sara is still left without an explanation for her poor result. </p>
<p>What’s needed is a way to turn information about counterfactuals and correlations into explanatory information.</p>
<h2>The future of explainable AI</h2>
<p>With time we can expect AI to be used more for hiring decisions, visa applications, promotions and state and federal funding decisions, among other things.</p>
<p>A lack of explanation for these decisions threatens to substantially increase the injustice people will experience. After all, without explanations we can’t correct mistakes made when using AI. Fortunately, philosophy can help. </p>
<p>Explanation has been a central <a href="https://plato.stanford.edu/entries/scientific-explanation/">topic of philosophical study</a> over the last century. Philosophers have designed a range of methods for extracting explanatory information from a sea of correlations, and have developed sophisticated theories about how explanation works.</p>
<p>A great deal of this work has focused on the relationship between counterfactuals and explanation. I’ve developed <a href="https://quod.lib.umich.edu/p/phimp/3521354.0017.003/--how-mathematics-can-make-a-difference?view=image">work on this</a> myself. By drawing on philosophical insights, we may be able to develop better approaches to explainable AI. </p>
<p>At present, however, there’s not enough overlap between philosophy and computer science on this topic. If we want to tackle injustice head-on, we’ll need a more integrated approach that combines work in these fields. </p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/when-self-driving-cars-crash-whos-responsible-courts-and-insurers-need-to-know-whats-inside-the-black-box-180334">When self-driving cars crash, who's responsible? Courts and insurers need to know what's inside the 'black box'</a>
</strong>
</em>
</p>
<hr>
<img src="https://counter.theconversation.com/content/196392/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Sam Baron receives funding from the Australian Research Council.</span></em></p>AI models are increasingly being used to make important decision about people’s lives – just take Robodebt. Yet the complexity of these systems means we hardly understand them.Sam Baron, Associate Professor, Philosophy of Science, Australian Catholic UniversityLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/1791432022-03-24T15:06:36Z2022-03-24T15:06:36ZHow to investigate when a robot causes an accident – and why it’s important that we do<figure><img src="https://images.theconversation.com/files/454117/original/file-20220324-27-11mk6pf.jpg?ixlib=rb-1.1.0&rect=13%2C0%2C8661%2C5787&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">
</span> <span class="attribution"><a class="source" href="https://www.shutterstock.com/image-photo/close-robots-hand-holding-magnifying-glass-720412816">Andrey_Popov/Shutterstock</a></span></figcaption></figure><p>Robots are featuring more and more in our daily lives. They can be incredibly useful (bionic limbs, robotic lawnmowers, or robots which <a href="https://www.independent.co.uk/tech/coronavirus-quarantine-robot-china-little-peanut-food-delivery-a9308166.html">deliver meals</a> to people in quarantine), or merely entertaining (robotic dogs, dancing toys, and acrobatic drones). Imagination is perhaps the only limit to what robots will be able to do in the future.</p>
<p>What happens, though, when robots don’t do what we want them to – or do it in a way that causes harm? For example, what happens if a bionic arm is involved <a href="https://www.theguardian.com/world/2010/oct/21/bionic-arm-driver-crashes-austria">in a driving accident</a>?</p>
<p>Robot accidents are becoming a concern for two reasons. First, the increase in the number of robots will naturally see a rise in the number of accidents they’re involved in. Second, we’re getting better at building more complex robots. When a robot is more complex, it’s more difficult to understand why something went wrong.</p>
<p>Most robots run on various forms of <a href="https://www.wired.co.uk/article/machine-learning-ai-explained">artificial intelligence</a> (AI). AIs are capable of making human-like decisions (though they may make objectively good or bad ones). These decisions can be any number of things, from identifying an object to interpreting speech.</p>
<p>AIs are trained to make these decisions for the robot based on information from vast datasets. The AIs are then tested for accuracy (how well they do what we want them to) before they’re set the task.</p>
<p>AIs can be designed in different ways. As an example, consider the robot vacuum. It could be designed so that whenever it bumps off a surface it redirects in a random direction. Conversely, it could be designed to map out its surroundings to find obstacles, cover all surface areas, and return to its charging base. While the first vacuum is taking in input from its sensors, the second is tracking that input into an internal mapping system. In both cases, the AI is taking in information and making a decision around it. </p>
<p>The more complex things a robot is capable of, the more types of information it has to interpret. It also may be assessing multiple sources of one type of data, such as, in the case of aural data, a live voice, a radio, and the wind.</p>
<p>As robots become more complex and are able to act on a variety of information, it becomes even more important to determine which information the robot acted on, particularly when harm is caused. </p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/were-teaching-robots-to-evolve-autonomously-so-they-can-adapt-to-life-alone-on-distant-planets-153159">We’re teaching robots to evolve autonomously – so they can adapt to life alone on distant planets </a>
</strong>
</em>
</p>
<hr>
<h2>Accidents happen</h2>
<p>As with any product, things can and do go wrong with robots. Sometimes this is an internal issue, such as the robot not recognising a voice command. Sometimes it’s external – the robot’s sensor was damaged. And sometimes it can be both, such as the robot not being designed to work on carpets and “tripping”. <a href="https://link.springer.com/chapter/10.1007/978-3-030-66494-7_6">Robot accident investigations</a> must look at all potential causes.</p>
<p>While it may be inconvenient if the robot is damaged when something goes wrong, we are far more concerned when the robot causes harm to, or fails to mitigate harm to, a person. For example, if a bionic arm fails to grasp a hot beverage, knocking it onto the owner; or if a care robot fails to register a distress call when the frail user has fallen.</p>
<p>Why is robot accident investigation different to that of human accidents? Notably, robots don’t have motives. We want to know why a robot made the decision it did based on the particular set of inputs that it had. </p>
<p>In the example of the bionic arm, was it a miscommunication between the user and the hand? Did the robot confuse multiple signals? Lock unexpectedly? In the example of the person falling over, could the robot not “hear” the call for help over a loud fan? Or did it have trouble interpreting the user’s speech?</p>
<figure class="align-center ">
<img alt="A person writing with a bionic arm." src="https://images.theconversation.com/files/454119/original/file-20220324-23-1ja40yk.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/454119/original/file-20220324-23-1ja40yk.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=400&fit=crop&dpr=1 600w, https://images.theconversation.com/files/454119/original/file-20220324-23-1ja40yk.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=400&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/454119/original/file-20220324-23-1ja40yk.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=400&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/454119/original/file-20220324-23-1ja40yk.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=503&fit=crop&dpr=1 754w, https://images.theconversation.com/files/454119/original/file-20220324-23-1ja40yk.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=503&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/454119/original/file-20220324-23-1ja40yk.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=503&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px">
<figcaption>
<span class="caption">When a robot malfunctions, we need to understand why.</span>
<span class="attribution"><a class="source" href="https://www.shutterstock.com/image-photo/disabled-girl-learning-write-prosthetic-arm-1302106870">UfaBizPhoto/Shutterstock</a></span>
</figcaption>
</figure>
<h2>The black box</h2>
<p>Robot accident investigation has a key benefit over human accident investigation: there’s potential for a built-in witness. Commercial aeroplanes have a similar witness: <a href="https://www.proquest.com/openview/29f5ebb4337bb7dc1d2dbdb002f9e57f/1.pdf?pq-origsite=gscholar&cbl=182">the black box</a>, built to withstand plane crashes and provide information as to why the crash happened. This information is incredibly valuable not only in understanding incidents, but in preventing them from happening again.</p>
<p>As part of <a href="https://www.robotips.co.uk/home">RoboTIPS</a>, a project which focuses on responsible innovation for social robots (robots that interact with people), we have created what we call the <a href="https://doi.org/10.1007/978-3-319-64107-2_21">ethical black box</a>: an internal record of the robot’s inputs and corresponding actions. The ethical black box is designed for each type of robot it inhabits and is built to record all information that the robot acts on. This can be voice, visual, or even <a href="https://www.sciencedaily.com/releases/2017/02/170206084904.htm">brainwave activity</a>.</p>
<p>We are testing the ethical black box on a variety of robots in both laboratory and simulated accident conditions. The aim is that the ethical black box will become standard in robots of all makes and applications.</p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/medical-robots-their-facial-expressions-will-help-humans-trust-them-154868">Medical robots: their facial expressions will help humans trust them</a>
</strong>
</em>
</p>
<hr>
<p>While data recorded by the ethical black box still needs to be interpreted in the case of an accident, having this data in the first instance is crucial in allowing us to investigate. </p>
<p>The investigation process offers the chance to ensure that the same errors don’t happen twice. The ethical black box is a way not only to build better robots, but to innovate responsibly in an exciting and dynamic field.</p><img src="https://counter.theconversation.com/content/179143/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Keri Grieman receives funding from EPSRC and The Alan Turing Institute.</span></em></p>Researchers have developed a black box for robots: an internal record of the robot’s inputs and actions. This will help investigate accidents and hopefully prevent future incidents.Keri Grieman, Research Associate, Department of Computer Science, University of OxfordLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/1517372021-01-13T13:18:32Z2021-01-13T13:18:32ZHow explainable artificial intelligence can help humans innovate<figure><img src="https://images.theconversation.com/files/378380/original/file-20210112-23-1yp7zx5.jpg?ixlib=rb-1.1.0&rect=13%2C25%2C885%2C572&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">Understanding how artificial intelligence algorithms solve problems like the Rubik's Cube makes AI more useful.</span> <span class="attribution"><a class="source" href="https://commons.wikimedia.org/wiki/File:Rubikscube-ecksteine.jpg">Roland Frisch via Wikimedia Commons</a>, <a class="license" href="http://creativecommons.org/licenses/by-sa/4.0/">CC BY-SA</a></span></figcaption></figure><p>The field of artificial intelligence (AI) has created computers that can <a href="https://www.nytimes.com/2019/06/20/business/self-driving-cars-cadillac-super-cruise.html">drive cars</a>, <a href="https://arxiv.org/pdf/2006.15820.pdf">synthesize chemical compounds</a>, <a href="https://doi.org/10.1038/d41586-020-03348-4">fold proteins</a> and <a href="https://doi.org/10.1038/ncomms5308">detect high-energy particles</a> at a superhuman level.</p>
<p>However, these AI algorithms cannot explain the thought processes behind their decisions. A computer that masters protein folding and also tells researchers more about the rules of biology is much more useful than a computer that folds proteins without explanation.</p>
<p>Therefore, <a href="https://scholar.google.com/citations?user=R3ru5X8AAAAJ&hl=en&oi=ao">AI researchers like me</a> are now turning our efforts toward developing AI algorithms that can explain themselves in a manner that humans can understand. If we can do this, I believe that AI will be able to uncover and teach people new facts about the world that have not yet been discovered, leading to new innovations.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/378382/original/file-20210112-15-rz08zh.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="A humanoid robot reading a book." src="https://images.theconversation.com/files/378382/original/file-20210112-15-rz08zh.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/378382/original/file-20210112-15-rz08zh.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=400&fit=crop&dpr=1 600w, https://images.theconversation.com/files/378382/original/file-20210112-15-rz08zh.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=400&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/378382/original/file-20210112-15-rz08zh.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=400&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/378382/original/file-20210112-15-rz08zh.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=503&fit=crop&dpr=1 754w, https://images.theconversation.com/files/378382/original/file-20210112-15-rz08zh.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=503&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/378382/original/file-20210112-15-rz08zh.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=503&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">When machines are left to learn and solve problems through their own experience, this is called reinforcement learning.</span>
<span class="attribution"><a class="source" href="https://www.gettyimages.com/detail/photo/futuristic-cyborg-learning-about-humans-royalty-free-image/1147104972?adppopup=true">Gremlin/E+ via Getty Images</a></span>
</figcaption>
</figure>
<h2>Learning from experience</h2>
<p>One field of AI, <a href="https://en.wikipedia.org/wiki/Reinforcement_learning">called reinforcement learning</a>, studies how computers can learn from their own experiences. In reinforcement learning, an AI explores the world, receiving positive or negative feedback based on its actions.</p>
<p>This approach has led to algorithms that have independently learned to <a href="https://doi.org/10.1126/science.aar6404">play chess at a superhuman level</a> and prove <a href="https://papers.nips.cc/paper/2018/hash/55acf8539596d25624059980986aaa78-Abstract.html">mathematical theorems</a> without any human guidance. In my work as <a href="https://scholar.google.com/citations?user=R3ru5X8AAAAJ&hl=en&oi=ao">an AI researcher</a>, I use reinforcement learning to create AI algorithms that learn how to <a href="https://doi.org/10.1038/s42256-019-0070-z">solve puzzles such as the Rubik’s Cube</a>. </p>
<p>Through reinforcement learning, AIs are independently learning to solve problems that even humans struggle to figure out. This has got me and many other researchers thinking less about what AI can learn and more about what humans can learn from AI. A computer that can solve the Rubik’s Cube should be able to teach people how to solve it, too.</p>
<h2>Peering into the black box</h2>
<p>Unfortunately, the minds of superhuman AIs are currently out of reach to us humans. AIs make terrible teachers and are what we in the computer science world call “<a href="https://doi.org/10.1038/538020a">black boxes</a>.”</p>
<figure class="align-left zoomable">
<a href="https://images.theconversation.com/files/378384/original/file-20210112-13-rkrav9.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="An open black case" src="https://images.theconversation.com/files/378384/original/file-20210112-13-rkrav9.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=237&fit=clip" srcset="https://images.theconversation.com/files/378384/original/file-20210112-13-rkrav9.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=400&fit=crop&dpr=1 600w, https://images.theconversation.com/files/378384/original/file-20210112-13-rkrav9.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=400&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/378384/original/file-20210112-13-rkrav9.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=400&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/378384/original/file-20210112-13-rkrav9.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=503&fit=crop&dpr=1 754w, https://images.theconversation.com/files/378384/original/file-20210112-13-rkrav9.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=503&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/378384/original/file-20210112-13-rkrav9.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=503&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">Researchers have been trying for decades to understand how AIs solve problems.</span>
<span class="attribution"><a class="source" href="https://www.gettyimages.com/detail/photo/open-black-box-royalty-free-image/146799217?adppopup=true">rockz/iStock via Getty Images Plus</a></span>
</figcaption>
</figure>
<p>A black-box AI simply spits out solutions without giving reasons for its solutions. Computer scientists have been trying for <a href="https://papers.nips.cc/paper/1995/file/45f31d16b1058d586fc3be7207b58053-Paper.pdf">decades to open this black box</a>, and recent research has shown that many AI algorithms actually do think in ways that are similar to humans. For example, a computer trained to recognize animals will learn about different types of eyes and ears and will put this information together <a href="https://doi.org/10.1007/978-3-319-10590-1_53">to correctly identify the animal</a>. </p>
<p>The effort to open up the black box is called <a href="https://proceedings.neurips.cc/paper/2020/file/2c29d89cc56cdb191c60db2f0bae796b-Paper.pdf">explainable AI</a>. <a href="https://cse.sc.edu/%7Eforesta/">My research group</a> at the AI Institute at the University of South Carolina is interested in developing explainable AI. To accomplish this, we work heavily with the Rubik’s Cube. </p>
<p>The Rubik’s Cube is basically a <a href="https://en.wikipedia.org/wiki/Pathfinding">pathfinding problem</a>: Find a path from point A – a scrambled Rubik’s Cube – to point B – a solved Rubik’s Cube. Other pathfinding problems include navigation, theorem proving and chemical synthesis. </p>
<p>My lab has set up a website where anyone can see how our <a href="http://deepcube.igb.uci.edu/">AI algorithm solves the Rubik’s Cube</a>; however, a person would be hard-pressed to learn how to solve the cube from this website. This is because the computer cannot tell you the logic behind its solutions.</p>
<p>Solutions to the Rubik’s Cube can be broken down into a few generalized steps – the first step, for example, could be to form a cross while the second step could be to put the corner pieces in place. While the Rubik’s Cube itself has over 10 to the 19th power possible combinations, a generalized step-by-step guide is very easy to remember and is applicable in many different scenarios. </p>
<p>Approaching a problem by breaking it down into steps is often the default manner in which people explain things to one another. The Rubik’s Cube naturally fits into this step-by-step framework, which gives us the opportunity to open the black box of our algorithm more easily. Creating AI algorithms that have this ability could allow people to collaborate with AI and break down a wide variety of complex problems into easy-to-understand steps.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/375160/original/file-20201215-15-j1knsg.png?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="An image showing the thought process of a Rubik's Cube-solving AI algorithm" src="https://images.theconversation.com/files/375160/original/file-20201215-15-j1knsg.png?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/375160/original/file-20201215-15-j1knsg.png?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=408&fit=crop&dpr=1 600w, https://images.theconversation.com/files/375160/original/file-20201215-15-j1knsg.png?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=408&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/375160/original/file-20201215-15-j1knsg.png?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=408&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/375160/original/file-20201215-15-j1knsg.png?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=513&fit=crop&dpr=1 754w, https://images.theconversation.com/files/375160/original/file-20201215-15-j1knsg.png?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=513&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/375160/original/file-20201215-15-j1knsg.png?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=513&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">A step-by-step refinement approach can make it easier for humans to understand why AIs do the things they do.</span>
<span class="attribution"><span class="source">Forest Agostinelli</span>, <a class="license" href="http://creativecommons.org/licenses/by-nd/4.0/">CC BY-ND</a></span>
</figcaption>
</figure>
<h2>Collaboration leads to innovation</h2>
<p>Our process starts with using one’s own intuition to define a step-by-step plan thought to potentially solve a complex problem. The algorithm then looks at each individual step and gives feedback about which steps are possible, which are impossible and ways the plan could be improved. The human then refines the initial plan using the advice from the AI, and the process repeats until the problem is solved. The hope is that the person and the AI will eventually converge to a kind of mutual understanding.</p>
<p>[<em>Deep knowledge, daily.</em> <a href="https://theconversation.com/us/newsletters/the-daily-3?utm_source=TCUS&utm_medium=inline-link&utm_campaign=newsletter-text&utm_content=deepknowledge">Sign up for The Conversation’s newsletter</a>.]</p>
<p>Currently, our algorithm is able to consider a human plan for solving the Rubik’s Cube, suggest improvements to the plan, recognize plans that do not work and find alternatives that do. In doing so, it gives feedback that leads to a step-by-step plan for solving the Rubik’s Cube that a person can understand. Our team’s next step is to build an intuitive interface that will allow our algorithm to teach people how to solve the Rubik’s Cube. Our hope is to generalize this approach to a wide range of pathfinding problems. </p>
<p>People are intuitive in a way unmatched by any AI, but machines are far better in their computational power and algorithmic rigor. This back and forth between man and machine utilizes the strengths from both. I believe this type of collaboration will shed light on previously unsolved problems in everything from chemistry to mathematics, leading to new solutions, intuitions and innovations that may have, otherwise, been out of reach.</p><img src="https://counter.theconversation.com/content/151737/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Forest Agostinelli does not work for, consult, own shares in or receive funding from any company or organization that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>AI algorithms can solve hard problems and learn incredible tasks, but they can’t explain how they do these things. If researchers can build explainable AI, it could lead to a flood of new knowledge.Forest Agostinelli, Assistant Professor of Computer Science, University of South CarolinaLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/1297092020-01-10T15:05:06Z2020-01-10T15:05:06ZIran plane crash: here is what happens during an air disaster investigation<p>The investigation into the January 8 plane crash in Iran that killed all 176 people on board is likely to attract more attention than most thanks to the ongoing political conflict in the region. Canadian prime minister Justin Trudeau claimed intelligence evidence suggested Ukrainian flight PS752 was <a href="https://www.bbc.co.uk/news/world-middle-east-51055219">hit by a missile</a>, as did unnamed US officials <a href="https://www.nytimes.com/2020/01/09/world/middleeast/iran-plane-crash-ukraine.html">speaking to journalists</a>. Iran initially denied the plane was hit by its air defences but has <a href="https://www.bbc.co.uk/news/world-middle-east-51073621">since claimed it</a> unintentionally shot down the craft.</p>
<p>The country also <a href="https://www.bbc.co.uk/news/world-middle-east-51042326">made a point of stating</a> that, as is standard practice for air crash investigations, it would not hand over the plane’s black box flight recorder to aircraft manufacturer Boeing or the US authorities. But the US has since <a href="https://www.reuters.com/article/us-iran-crash-investigation/u-s-to-join-probe-of-jet-disaster-no-missile-link-iran-official-idUSKBN1Z82SE">accepted an invitation</a> from Iran to take part in the investigation, and Iran’s Civil Aviation Authority head has reportedly said it <a href="https://edition.cnn.com/middleeast/live-news/iran-plane-crash-investigation/h_4a1d75c8b10e369297e098ace9ff0317">may need help</a> decoding the black box data. </p>
<p>Aircraft investigations are covered by a <a href="https://www.emsa.europa.eu/retro/Docs/marine_casualties/annex_13.pdf">set of international standards</a> and recommended practices produced by the UN’s International Civil Aviation Organisation. The aim to understand what has happened and why, but more importantly, to also produce recommendations to prevent future accidents. Here’s how we should expect the investigation to proceed.</p>
<h2>Who takes part?</h2>
<p>The standards, known in the aviation industry as “Annex 13”, determine that Iran, as the country where the incident occurred, will lead the investigation. It will nominate an investigator in charge, someone who normally works for an independent aviation or transport accident investigation agency and so should be able to conduct an impartial inquiry.</p>
<p><a href="https://www.icao.int/about-icao/FAQ/Pages/icao-frequently-asked-questions-faq-10.aspx">Other countries entitled to take part</a> include those where the plane was designed and made (the US) and where it was registered and operated from (Ukraine). Countries with a significant number of citizens involved in the incident can also participate, which in this case will be Canada, Sweden and Afghanistan. These countries will all send “accredited representatives” from their respective aviation or transport safety investigation agencies.</p>
<p>The investigator in charge can also draw on technical specialists. It would be normal, for example, for investigators from Boeing to be part of the team but always working with the investigators’ national agency, to ensure transparency and fair play.</p>
<p>The investigator in charge controls and decides what counts as evidence. If there is evidence of a criminal act at any point, then the safety investigation may hand over the inquiry to law enforcement. This was the case with the Pan Am 103 <a href="https://www.gov.uk/aaib-reports/2-1990-boeing-747-121-n739pa-21-december-1988">bombing over Lockerbie</a> in 1988 where it became clear that it was not an aircraft accident, but rather an act of sabotage. In that case, a safety investigation continued in parallel to the criminal investigation but focused on aircraft and passenger survivability.</p>
<h2>Black boxes</h2>
<p><a href="https://www.abc.net.au/news/2014-03-26/black-box-flight-recorders/5343456">Black boxes</a> are a key source of evidence in most aircraft incidents. An aircraft, such as the one involved in the Iranian crash, would carry a cockpit voice recorder covering up to two hours of audio from the flight deck. It would also contain a digital flight data recorder that would record hundreds of parameters over 25 hours of operation. This could include engine settings, the positions of flight controls, warnings and so on.</p>
<p>The recorders are designed to withstand impact, high temperature and immersion in water, but are not indestructible. They also depend on receiving power from the aircraft and data from a range of sources across the aircraft. The first generations of recorders used wire and magnetic tape that could collect a limited amount of data and were quite vulnerable to damage. But modern recorders use solid-state memory chips in a crash-survivable memory unit.</p>
<p>An undamaged recorder can be read out relatively simply using equipment from the manufacturer of the recorders, but in the case of a damaged recorder, investigators need to be very careful to ensure they do not corrupt the data. They also need to be careful to protect their chain of evidence from any accusations of tampering.</p>
<p>For this reason, they would never be sent to the aircraft manufacturer (in this case Boeing) for read-back. This is a task that would be done by a state safety investigation agency. Yet while most states have such an agency, only some have data recorder read-back facilities and even fewer have experience in dealing with badly damaged recorders.</p>
<p>If the country leading the investigation does not have the technical capability in-house then they are likely to ask another country, either one involved in the investigation or an outside country. Or they might ask a “technical adviser”, an organisation or person with specific expertise, who could specialise in, for example, the recovery of data from damaged memory chips.</p>
<p>If a recorder is to be examined in a different country, it will be taken there by one of the investigation team. They will agree on a method of work, record the opening of the recorder’s protective casing, and be present throughout the examination. If data can be recovered easily then it may be possible to examine the recorded data relatively swiftly.</p>
<p>In the case of the cockpit voice recording, <a href="https://www.bea.aero/fileadmin/user_upload/Guidance_on_CVR_recording_Inspection.pdf">there are strict protocols</a> on who can listen. This is partly to protect the pilots, as the recording is solely for safety purposes and should not be used as evidence against them. This is a principle established because many pilots feared the recorder would be a “spy in the cockpit”. Another reason for limiting who listens to recordings is that they may contain very traumatic sounds associated with the last moments of those on board.</p>
<p>Following the loss of flight PS752, it was correct for the Iranian authorities to say that they would not hand over the recorder to Boeing. If Iran has the technical capability within its safety investigation agency, then they would be following protocol to analyse them. If not, then they are entitled to approach, or accept the assistance of another capable safety investigation agency or technical adviser. In this case, the US and Canada both have a high level of expertise in the examination of flight data recorders. </p>
<p>What is of critical importance is that the recordings are not damaged by the read-back process. A chain of evidence must be preserved to ensure the integrity of the investigation and that the recordings are not released without the express permission of the investigator in charge.</p>
<p><em>This article has been updated to reflect Iran’s admission of unintentionally shooting down the plane.</em></p><img src="https://counter.theconversation.com/content/129709/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Graham Braithwaite does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>Crash investigators never give black boxes to aircraft makers but involve them in the process.Graham Braithwaite, Director of Transport Systems, Professor of Safety & Accident Investigation, Cranfield UniversityLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/1206912019-07-30T12:16:10Z2019-07-30T12:16:10ZInsurance black boxes and the surveillance state – how free are you, really?<figure><img src="https://images.theconversation.com/files/286017/original/file-20190729-43149-ybfn1t.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">
</span> <span class="attribution"><a class="source" href="https://www.shutterstock.com/image-photo/long-exposure-photo-traffic-on-move-60939574?src=gyfcdCUqd3tTVSNrDdi1jA-1-38&studio=1">AdamEdwards/Shutterstock.com</a></span></figcaption></figure><p>Over the last few years there’s been a noticeable rise in the number of drivers opting to fit a “black box” to their cars in order to obtain cheaper insurance. According to some recent reports, these black boxes could save drivers as much as <a href="https://uk.finance.yahoo.com/news/young-motorists-drive-well-two-230100453.html?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAAHMou75J_2hl4FABJ9ic8elE1s93N0W3kEI5wmurUljhEkdD11n0Ye3VN31G4sf-Z3L8pJ6kAY7G6XWkR89EwRp1ReToMJ4KQ8oL7nSgVC8UGDjMBeyz48PsFyNipREoOQEXEIl9tbwYBpSWcm8owWAJS96KAAb3HiWvw-c5PnE4">£300</a> a year.</p>
<p>The idea is that the boxes send location data to nearby satellites, allowing insurance companies to monitor how people are driving, offering discounts and even refunds to those deemed to be driving more safely. As a result, black box drivers tend to drive quite cautiously, avoiding fast acceleration and never exceeding the speed limit, wherever they happen to be.</p>
<p>While this type of driving can be rather annoying to more experienced drivers, the whole concept of the “black box” exposes some fundamental truths about the way rules work.</p>
<p>For example, you may notice the trend of some black box drivers adding bumper stickers to their cars, informing fellow road users that they’re not driving slowly on purpose, but rather that they’re only doing it for cheaper insurance.</p>
<p><div data-react-class="Tweet" data-react-props="{"tweetId":"1071520349769162754"}"></div></p>
<p>These bumper stickers put into words a process that we all tacitly accept on a day-to-day basis, but never quite admit. That is, that there is a certain amount of “flexibility” in the application of law – whether it be on Britain’s roads or elsewhere. This “flexibility” is based on the fact that it is quite simply impossible for the authorities to enforce all rules on all people at all times.</p>
<p>In this case, the black box driver freely admits that they’d break the rules if they weren’t driving under the watchful gaze of their electronic overseer; they’re only obeying the rules because they’re being watched.</p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/technology-dominates-our-lives-thats-why-we-should-teach-human-rights-law-to-software-engineers-102530">Technology dominates our lives – that's why we should teach human rights law to software engineers</a>
</strong>
</em>
</p>
<hr>
<p>But what’s most fascinating about this example is that logically speaking, the black box doesn’t even need to contain any electrical gadgetry at all.</p>
<p>Certainly, the driver wouldn’t know any different, as they’d still drive with the same acute awareness of the rules of the road. Of course, this would mean the insurance companies wouldn’t receive any telemetric data, but then, what are they using the data for, if not to compel “safe” driving?</p>
<h2>Rules, codes and conventions</h2>
<p>The reason I find this all so interesting is because it ties in very much with my <a href="https://mjryder.net/category/research/">own research</a> relating to human and robotic behaviour, and the power structures that shape our everyday lives. Think CCTV warning signs for example, or target pictures painted on men’s urinals.</p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/a-urinal-in-a-scottish-pub-reveals-why-toilets-matter-in-international-politics-118312">A urinal in a Scottish pub reveals why toilets matter in international politics</a>
</strong>
</em>
</p>
<hr>
<p>But beyond these more overt forms of control, there are also many hidden social structures that encode our behaviour and prompt us to behave in a certain way. The cinema is a good example. While there will always be those who flout the rules to a minor degree – such as people who check their phone while film is on or noisily chat to their friends – instances of major disruption are few and far between, as most people adhere to the unspoken rules of cinema etiquette.</p>
<figure class="align-center ">
<img alt="" src="https://images.theconversation.com/files/286018/original/file-20190729-43122-uetbjz.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/286018/original/file-20190729-43122-uetbjz.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=400&fit=crop&dpr=1 600w, https://images.theconversation.com/files/286018/original/file-20190729-43122-uetbjz.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=400&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/286018/original/file-20190729-43122-uetbjz.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=400&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/286018/original/file-20190729-43122-uetbjz.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=503&fit=crop&dpr=1 754w, https://images.theconversation.com/files/286018/original/file-20190729-43122-uetbjz.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=503&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/286018/original/file-20190729-43122-uetbjz.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=503&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px">
<figcaption>
<span class="caption">Playing their parts.</span>
<span class="attribution"><a class="source" href="https://www.shutterstock.com/image-photo/friends-bride-groom-on-wedding-ceremony-373586557?src=f8WQXfAKj5leY-44I3npYA-1-3&studio=1">IVASHstudio/Shutterstock.com</a></span>
</figcaption>
</figure>
<p>In a similar vein, there are no formal rules about how to behave at a wedding ceremony, a funeral, or a job interview. While there may not be a written code as such, we all tend to have an idea of what constitutes appropriate behaviour. In this way, we self-manage our own conformity, and in so doing also share the same hidden cultural codes with those around us. We do this to avoid censure and bad feeling, as we are constantly aware of the gaze of others.</p>
<h2>The invisible black box</h2>
<p>To bring this back to the car insurance example, what’s fascinating is that it doesn’t matter whether or not the black box really is watching our every move. Much rather, what’s important is that we think we might be being watched and we modify our behaviour to suit.</p>
<p>This concept links somewhat with Jeremy Bentham’s famous <a href="https://en.wikipedia.org/wiki/Panopticon">panopticon</a> concept from the 18th century. In his writings, Bentham describes an “ideal” form of prison where inmates live under the constant threat of surveillance. While in reality, each individual prisoner is barely watched at all, there remains the chance that they might be being watched at any point.</p>
<figure class="align-center ">
<img alt="" src="https://images.theconversation.com/files/286019/original/file-20190729-43122-19kd0d4.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/286019/original/file-20190729-43122-19kd0d4.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=661&fit=crop&dpr=1 600w, https://images.theconversation.com/files/286019/original/file-20190729-43122-19kd0d4.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=661&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/286019/original/file-20190729-43122-19kd0d4.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=661&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/286019/original/file-20190729-43122-19kd0d4.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=831&fit=crop&dpr=1 754w, https://images.theconversation.com/files/286019/original/file-20190729-43122-19kd0d4.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=831&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/286019/original/file-20190729-43122-19kd0d4.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=831&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px">
<figcaption>
<span class="caption">Plan of Jeremy Bentham’s panopticon prison, drawn by Willey Reveley in 1791.</span>
<span class="attribution"><a class="source" href="https://en.wikipedia.org/wiki/Panopticon#/media/File:Panopticon.jpg">Wikimedia Commons</a></span>
</figcaption>
</figure>
<p>While Bentham’s panopticon has since fallen out of favour, the concept continues to this day, and ties in very closely with our understanding of surveillance culture and biopolitics – that is, the way the state takes life as its central objective, and frames our lives as constantly under threat. You may for example, notice that in many shopping outlets, the warning signs about CCTV coverage are far more prominent than the cameras themselves. This is because the security team can’t watch all people at all times. But the possibility of surveillance is used as a means to <a href="https://theconversation.com/how-gaming-in-the-classroom-prepares-children-for-life-in-a-surveillance-state-77287">encourage good behaviour</a>.</p>
<hr>
<p>
<em>
<strong>
Read more:
<a href="https://theconversation.com/how-gaming-in-the-classroom-prepares-children-for-life-in-a-surveillance-state-77287">How gaming in the classroom prepares children for life in a surveillance state</a>
</strong>
</em>
</p>
<hr>
<p>This same concept also applies to the insurance black box. While insurance companies undoubtedly do monitor the habits of drivers, the purpose of the box isn’t so much to monitor but to enforce good behaviour. For this reason, it doesn’t really matter whether or not there are any electronic gizmos inside the box at all. The important thing is not that we are watched, but rather, that we obey.</p>
<h2>Whose rules are they anyway?</h2>
<p>This whole concept of the black box and what it says about our surveillance culture poses some interesting dilemmas. We accept conformity and regulation as a way to keep us safe from harm. But at the same time, we also want “freedom”, and to feel as though we’re still in control.</p>
<p>This is why transgression is such an important part of modern life. If we don’t believe we have the free will to break rules, then we are forced to confront the tension at the heart of our everyday lives. On the one hand, we want “freedom”, but we also want safety, and to live our lives free from harm. </p>
<p>We can’t have it both ways, so breaking the rules, if only just a little bit, gives us some access to the (illusory) “freedom” that we give up as citizens of the surveillance state. Something to think about next time you drive 80mph down the motorway.</p><img src="https://counter.theconversation.com/content/120691/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Mike Ryder does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>Those money-saving black boxes reveal a lot about the rules that govern our lives.Mike Ryder, Associate Lecturer, Literature and Philosophy, Lancaster UniversityLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/520542015-12-15T13:33:22Z2015-12-15T13:33:22ZNo aircraft is too small to kill – fitting flight recorders could prevent further deaths<p>When a police helicopter plunged into the Clutha pub in Glasgow city centre two years ago, killing ten, an accident investigation was launched to find out what happened and why – a process that is essential to preventing future accidents and loss of life. But smaller aircraft are not required to carry flight data recorders – the black box so vital to air crash investigations. So while the <a href="http://aerossurance.com/helicopters/clutha-police-heli-accident/">recently published report</a> into the crash revealed some of the reasons behind, there were details – potentially vital details – that couldn’t be known because they weren’t recorded. It’s about time this changed.</p>
<p>The Air Accident Investigation Branch (AAIB) report recorded that the Strathclyde Police <a href="http://www.flyingmag.com/aircraft/helicopters/eurocopter-ec-135">Eurocopter EC135</a> with a crew of three took off from Glasgow Heliport at 8:44pm in order to conduct missions between Glasgow and Edinburgh. The captain was an experienced pilot aged 51, who had flown RAF helicopters in Afghanistan, Iraq, Bosnia and Northern Ireland before retiring from military service and flying helicopters for the police.</p>
<p>At about 10:06pm the helicopter arrived over Bothwell in Glasgow with about 122kg of fuel remaining. Local rules require that a fuel “urgency” (the level below an “emergency” or “Mayday” call) be declared if levels fall below 100kg, and landing should be quick. Because the fuel pump switches that transfer fuel from a reserve to the main fuel tanks were turned off, a fuel warning went off in the cockpit – a warning the pilot cancelled five times. </p>
<p>At about 10:22pm there was 73kg of fuel left in the reserve tank, the helicopter’s two engines failed 45 seconds apart and the pilot was forced to attempt an emergency landing. This was handled poorly for reasons including the helicopter’s low altitude, the darkness of the night, and the failure of the radio-altimeter, a device that in this model of helicopter relied upon engine power to tell the crew the exact distance between the helicopter and the ground below. The helicopter crashed at high speed into the roof of the Clutha Bar, killing the crew and seven others in the pub. </p>
<h2>Unanswered questions</h2>
<figure class="align-right zoomable">
<a href="https://images.theconversation.com/files/105607/original/image-20151213-29732-1kutrel.JPG?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/105607/original/image-20151213-29732-1kutrel.JPG?ixlib=rb-1.1.0&q=45&auto=format&w=237&fit=clip" srcset="https://images.theconversation.com/files/105607/original/image-20151213-29732-1kutrel.JPG?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=563&fit=crop&dpr=1 600w, https://images.theconversation.com/files/105607/original/image-20151213-29732-1kutrel.JPG?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=563&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/105607/original/image-20151213-29732-1kutrel.JPG?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=563&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/105607/original/image-20151213-29732-1kutrel.JPG?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=708&fit=crop&dpr=1 754w, https://images.theconversation.com/files/105607/original/image-20151213-29732-1kutrel.JPG?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=708&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/105607/original/image-20151213-29732-1kutrel.JPG?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=708&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">If all aircraft had these, it’s possible more lives could be saved.</span>
<span class="attribution"><a class="source" href="https://commons.wikimedia.org/wiki/File:Black_Box_Cockpit_Voice_Recorder,_Model_AV557D,_Sunderstrand_Data_Control,_Inc.,_c._1990s_-_National_Electronics_Museum_-_DSC00090.JPG">Daderot</a></span>
</figcaption>
</figure>
<p>The AAIB established the technical facts of the accident by examining memory chips from various aircraft electronics that survived. This took a very long time, and some uncertainties remain.</p>
<p>Why would a hugely experienced professional helicopter pilot leave the fuel transfer pumps turned off, cancel five low fuel warnings and continue to fly his helicopter below safety minimum fuel levels? Did his experienced police observers challenge him about the warnings he kept cancelling? Critically, we know nothing about what was happening in the cockpit: when and why the pilot cancelled the audible warnings, what was discussed, and whether any action was taken. We don’t know the answers to these questions, nor can we ever know.</p>
<p>There have been at least 12 police aircraft crashes in Britain since 1985, eight of them fatal, <a href="http://news.bbc.co.uk/1/hi/uk/139680.stm">one air ambulance helicopter crash</a>, and several other small helicopter crashes, such as <a href="http://www.bbc.co.uk/news/uk-england-21040410">one in central London</a>. None carried the standard <a href="http://www.skybrary.aero/index.php/Cockpit_Voice_Recorder_%28CVR%29">cockpit voice recorder</a> to record cockpit discussions or the <a href="http://www.skybrary.aero/index.php/Flight_Data_Recorder_%28FDR%29">flight data recorder</a> that records data about the aircraft’s systems and which are required for larger aircraft. Both could have assisted the investigation.</p>
<p>The AAIB have repeatedly recommended that recorders be fitted, and <a href="http://aerossurance.com/helicopters/clutha-police-heli-accident/">did so again with this report</a>. A flight recorder is <a href="http://latitudetech.com/latitudes-ionode-fdm-system-faa-certified-for-airbus-ec135-airframes/">an available option</a> for the EC135 model helicopter, but as the helicopter has a maximum take-off weight of 2,910kg it falls below the legal limit that requires crash-proof flight recording equipment to be fitted, which is 3,175kg.</p>
<h2>Learning from mistakes</h2>
<p>The loss of <a href="http://www.bea.aero/docspa/2009/f-cp090601e2.en/pdf/f-cp090601e2.en.pdf">Air France flight 447</a>, an Airbus A330, in the South Atlantic in June 2009 was, initially, an inexplicable crash. But once the cockpit voice recorder had been retrieved from ocean floor it revealed that the crew <a href="http://www.popularmechanics.com/flight/a3115/what-really-happened-aboard-air-france-447-6611877/">misunderstood the conditions they were in</a> and consequently mishandled the jet, leading to the crash and the deaths of 228 people.</p>
<p>Flight recorders are a tried and tested technology fitted to many aircraft. While systems will be necessarily simpler than that on an airliner, the only conceivable reason that smaller aircraft do not carry them is to save money. Yet their absence hampers investigation of this and other crashes, and prevents us from understanding what occurred – something that could cost more lives in the future.</p>
<p>As with the Strathclyde Police helicopter, the same is true of most smaller military aircraft too. But following a tragic <a href="https://www.gov.uk/government/publications/service-inquiry-accident-involving-lynx-mk-9-zf540-on-26-april-2014--2">Lynx helicopter training crash in 2014</a>, the Ministry of Defence changed its position and now requires flight recorders to be fitted on all military aircraft, irrespective of size. The government must force other operators to follow suit.</p>
<p>Aircraft accidents happen, but accidents and risks can be minimised by identifying the causes of accidents and ensuring that the lessons learned are carried forward in changes to equipment, training, procedures, or the law. We need legislation or regulation to close these loopholes and ensure proper flight recorders are mandatory on all commercial and public service aircraft, as the AAIB has repeatedly called for, and the government and aircraft operators continue to ignore.</p><img src="https://counter.theconversation.com/content/52054/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Guy Gratton is an aerospace academic who regularly appears in the media discussing air accidents. He occasionally receives payment for doing so.</span></em></p>Without the details that flight recorders provide, crash investigators cannot fully understand the causes of catastrophe.Guy Gratton, Visiting Senior Research Fellow, Brunel University LondonLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/362012015-01-13T19:30:07Z2015-01-13T19:30:07ZAnswers needed from the flight recorders of Air Asia flight QZ8501<p>Now the flight data recorders from Air Asia flight QZ8501 have been found investigators have a chance to find out what really brought down the aircraft with 162 passengers and crew on board.</p>
<p>Divers have recovered the flight data recorder which records around 2,000 parameters from the aircraft systems, such as airspeed, altitude, heading, flight control positions, engine data and many more. They are also <a href="http://www.abc.net.au/news/2015-01-13/divers-retrieve-airasia-cockpit-voice-recorder/6014360">reported</a> to have located and recovered the cockpit voice recorder which records the last two hours of voice communications and, via a cockpit area microphone, other noises in the cockpit. </p>
<p>The discovery and recovery of these <a href="https://theconversation.com/the-black-box-an-australian-invention-that-nearly-didnt-happen-25435">black boxes</a> marks the first major breakthrough in the search effort since the flight QZ8501 <a href="http://qz8501.airasia.com/28-dec-2014/index.html">disappeared on December 28, 2014</a>, en route from Surabaya in Indonesia to Singapore.</p>
<h2>Questions with no answers</h2>
<p>Since the initial disappearance there has been much speculation as to what happened to the flight. The last known communication between air traffic controllers and the pilots was their <a href="http://www.news.com.au/travel/travel-updates/airasia-flight-qz8501-how-other-planes-dodged-the-storm/story-fnizu68q-1227169713074">request to climb</a> to avoid some bad weather which was denied by air traffic controllers. Soon after the aircraft disappeared.</p>
<p>Air Asia <a href="http://qz8501.airasia.com/12-jan-2015/index.html">said earlier this week</a> that 48 bodies have so far been recovered from the Java Sea with 34 identified. Wreckage from the Airbus A320-200 is also still being recovered.</p>
<p>Investigators will be keenly interested in what the flight data recorder and cockpit voice recorder will reveal. Their recovery was vital.</p>
<p>Until now the available information coming from the investigation supported many diverse scenarios which could have led to the loss of the aircraft, passengers and crew. Some theories were more plausible than others but none were supported by conclusive facts.</p>
<h2>The data analysis begins</h2>
<p>If the recorders are undamaged, it should only take a few days or even less for the data to be downloaded from the recorders and analysed.</p>
<p>Sophisticated computer modelling programs can use the recorded data to provide real time computer generated video images of what took place. This will show in detail all the progress of the final stages of the flight, readings of all the relevant flight instrumentation, flight control movements, all synthesised with what was said on the flight deck. Investigators will get a very clear picture of what took place. </p>
<p>Analysis of the wreckage as it’s recovered from the sea floor may also help shed light on what happened. The distribution of the wreckage may give an indication of whether there was any inflight break up, separation of any flight critical components or whether the aircraft entered the water intact. Analysis of any structural failures and other damage will also help provide evidence of what happened.</p>
<p>Of course, the investigation will then need to focus on looking for evidence of why whatever happened, happened. But this may not be as easy as it sounds. </p>
<p>Let’s assume that the aircraft entered icing conditions – as <a href="http://www.smh.com.au/world/ice-may-have-caused-airasia-engine-damage-says-indonesias-weather-agency-20150104-12hry9.html">has been speculated</a> – and ice formation adversely affected the air pressure detection capability of the pitot heads, used to measure air speed.</p>
<p>If that were the case, the ice will have melted long before the investigators have an opportunity to inspect the systems recovered from the wreckage. So, investigators will have to carefully analyse the flight recorder data looking for deviations in those critical systems which are fed with signals from the pitot heads which may be consistent with what could be expected if icing of the pitot heads had occurred. Icing may also become more of a focus if no other malfunctions or contradictory evidence is found.</p>
<p>Of course this is just one of many possible causation scenarios which the already available information supports. Forensic analysis of all evidence available to the investigators may reveal something different entirely.</p>
<h2>Questions of passenger flight safety?</h2>
<p>The crash of Air Asia flight QZ8501 comes after a difficult year for the passenger airline industry.</p>
<p>The search is still on going for the missing Malaysia <a href="https://theconversation.com/au/topics/flight-mh370">flight MH370</a>. That aircraft departed Kuala Lumpur in Malyasia on March 8, 2014, and was en route to Beijing,but the search effort is now concentrated in deep water <a href="http://www.atsb.gov.au/mh370.aspx">off the coast of Western Australia</a>.</p>
<p>How the Malaysia Airlines’ Boeing 777 with 239 passengers and crew on board may have gone down there is still a mystery.</p>
<p>Then on July 17, 2014, there was the downing of Malaysia Airlines <a href="https://theconversation.com/au/topics/flight-mh17">flight MH17</a> which had not long before left Amsterdam airport with 283 passengers and 15 crew on board, heading for Kualar Lumpur airport in Malaysia. Evidence from the wreckage suggests the Boeing 777 was brought down by a <a href="https://theconversation.com/damage-to-the-cockpit-gives-a-clue-to-loss-of-flight-mh17-31513">missile</a> as it flew over troubled areas in the Ukraine.</p>
<p>So the crash of another aircraft is only going to raise concerns over passenger airline safety. But how unusual is the flight QZ8501 case?</p>
<h2>The safest part of flying</h2>
<p>Ordinarily, airliner accidents during the cruise stages of flights are the most infrequent, as the the <a href="http://www.boeing.com/news/techissues/pdf/statsum.pdf">Boeing Statistical Summary of Commercial Jet Airplane Accidents 1959 to 2013</a>, published in August 2014 shows.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/68820/original/image-20150113-28434-1ldc455.png?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/68820/original/image-20150113-28434-1ldc455.png?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/68820/original/image-20150113-28434-1ldc455.png?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=300&fit=crop&dpr=1 600w, https://images.theconversation.com/files/68820/original/image-20150113-28434-1ldc455.png?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=300&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/68820/original/image-20150113-28434-1ldc455.png?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=300&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/68820/original/image-20150113-28434-1ldc455.png?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=377&fit=crop&dpr=1 754w, https://images.theconversation.com/files/68820/original/image-20150113-28434-1ldc455.png?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=377&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/68820/original/image-20150113-28434-1ldc455.png?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=377&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">Note that only 10% fatalities occur during the main cruise period of flights from 2004 through to 2013.</span>
<span class="attribution"><a class="source" href="http://www.boeing.com/news/techissues/pdf/statsum.pdf">Boeing's Statistical Summary of Commercial Jet Airplane Accidents, Worldwide Operations 1959 -- 2013.</a></span>
</figcaption>
</figure>
<p>So the recent spate of mid flight crashes, if you take MH17 and MH370 into consideration (even though they may both yet be confirmed to be due to acts of terrorism, they still represent failures of the overall aviation safety arrangements), is really atypical of the experience of the past ten years where only one in ten accidents have occurred mid flight.</p>
<p>The Boeing report also shows that 57% of the average flight time is in the cruise phase of flight – where only 10% of fatal accidents have occurred – against 16% of time for take-off and climb (22% of accidents), 26% of time on descent and approach (33% of accidents) and significantly 1% of time on landings (25% of accidents) which is clearly the highest risk time.</p>
<p>Despite these high profile accidents, 2014 had the least number of fatal accidents in the previous ten years although the number of actual <a href="http://edition.cnn.com/2014/12/29/travel/aviation-year-in-review/">casualties was high</a>. </p>
<p>Over that same ten year period, the number of flights flown annually by the worlds airlines has steadily grown. </p>
<p>Airlines globally can statistically be shown as providing the safest form of transport of all. That enviable record has been won by ensuring that the lessons are systematically learned from those accidents which do occur.</p>
<p>Accordingly, it is vital that the industry learns everything it can from QZ8501, MH370 and MH17 to inform future improvements in the aviation safety system.</p><img src="https://counter.theconversation.com/content/36201/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Geoffrey Dell does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>Now the flight data recorders from Air Asia flight QZ8501 have been found investigators have a chance to find out what really brought down the aircraft with 162 passengers and crew on board. Divers have…Geoffrey Dell, Associate Professor/Discipline Leader Accident Investigation and Forensics, CQUniversity AustraliaLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/306332014-08-19T05:17:02Z2014-08-19T05:17:02ZHere’s how you find out who shot down MH17<p>More than a month has passed since Malaysia Airlines flight MH17 crashed with the loss of all 298 lives on board. But despite the disturbances at the crash site near the small town of Grabovo, near Donetsk in eastern Ukraine, it is still possible to piece together what happened.</p>
<p>In the immediate aftermath it was reported that the aircraft had been shot down by a surface-to-air missile. Shortly after the crash Igor Girkin, leader of the separatists, <a href="http://www.theepochtimes.com/n3/805200-igor-girkin-commander-of-donetsk-peoples-army-igor-strelkov-says-they-shot-down-malaysia-airlines-mh17-photos/">took credit for the incident</a>, claiming his troops had shot down what was assumed to be a Ukranian military transport. After learning it was a civilian airliner he later denied any involvement, claiming his forces had no weapons capable of shooting down an aircraft flying at 33,000ft, as MH17 was.</p>
<p>In the weeks since there have been <a href="http://www.theguardian.com/world/2014/jul/23/mh17-ukraine-separatists-buk-missile-system">claims</a> and <a href="http://www.bbc.co.uk/news/world-europe-28458324">counterclaims</a> about which side had access to one of the Buk M1 missile launchers thought to be responsible for downing the plane. It’s possible to get nearer the truth of what happened to MH17 by combining evidence from four elements in play the moment the airliner came down: the air traffic control secondary radar systems that monitor the airspace over eastern Ukraine, the flight data recorder (“black box”) carried by the Boeing 777 aircraft, the specification of the Buk M1 missile system, and the wreckage of MH17.</p>
<h2>The view from air traffic control</h2>
<p>The main source of information about aircraft for air traffic controllers is <a href="http://news.bbc.co.uk/dna/place-lancashire/plain/A588684">secondary surveillance radar</a> (SSR). This is a link between an interrogator unit on the ground and a transponder on board an aircraft. The interrogator will connect to any aircraft within its range at least once every four seconds. The aircraft responds with flight information, such as the aircraft’s identification, position coordinates, course, and height. </p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/56736/original/jv85fthw-1408378251.GIF?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/56736/original/jv85fthw-1408378251.GIF?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/56736/original/jv85fthw-1408378251.GIF?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=325&fit=crop&dpr=1 600w, https://images.theconversation.com/files/56736/original/jv85fthw-1408378251.GIF?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=325&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/56736/original/jv85fthw-1408378251.GIF?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=325&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/56736/original/jv85fthw-1408378251.GIF?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=408&fit=crop&dpr=1 754w, https://images.theconversation.com/files/56736/original/jv85fthw-1408378251.GIF?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=408&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/56736/original/jv85fthw-1408378251.GIF?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=408&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">The still-contested area in which MH17 came down.</span>
<span class="attribution"><span class="license">Author provided</span></span>
</figcaption>
</figure>
<p>This information is processed, and correlated with the results of any <a href="http://news.bbc.co.uk/dna/place-lancashire/plain/A588684">primary surveillance radar</a> – the objects detected by beaming out radio waves and recording what is reflected back by solid objects. </p>
<p>This is displayed to the en-route controller, in this case the staff monitoring <a href="http://uksatse.ua/index.php?act=Part&CODE=319&lang=en">Dnipropetrovsk flight information region</a> eastbound, in Ukrainian government-controlled territory. This system will be able to locate a flight’s position to within 500 metres, given (in this case) MH17’s cruising speed of around 600 knots. All this coordinate tracking data will be recorded and stored for investigation purposes. It is highly unlikely that it could have been tampered with.</p>
<h2>Data within the black box</h2>
<p>A <a href="https://theconversation.com/explainer-what-will-mh17s-black-boxes-reveal-29555">flight data recorder</a> (FDR), commonly referred to as the black box, records all instructions sent to any electronic systems on the aircraft. It records flight characteristics such as altitude, speed, commands sent to the engines or control surfaces of the plane such as the rudder, ailerons and flaps, and streams of data from many sensors and on-board computers, including GPS coordinates accurate to within 10 metres. </p>
<p>Usually mounted in the aircraft’s tail section where it is more likely to survive a crash, the recorder is updated several times a second throughout the flight. This information is crucial for accident investigations and to determine safety issues.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/56733/original/x2472nbw-1408377034.JPG?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/56733/original/x2472nbw-1408377034.JPG?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/56733/original/x2472nbw-1408377034.JPG?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=308&fit=crop&dpr=1 600w, https://images.theconversation.com/files/56733/original/x2472nbw-1408377034.JPG?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=308&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/56733/original/x2472nbw-1408377034.JPG?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=308&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/56733/original/x2472nbw-1408377034.JPG?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=388&fit=crop&dpr=1 754w, https://images.theconversation.com/files/56733/original/x2472nbw-1408377034.JPG?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=388&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/56733/original/x2472nbw-1408377034.JPG?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=388&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">The flight data recorder, a crucial part of any investigation.</span>
<span class="attribution"><span class="source">AlliedSignal Aerospace</span></span>
</figcaption>
</figure>
<p>It would be virtually impossible to tamper with the FDR data as it is encoded, and follows a given sequence using set patterns that would reveal any attempt to alter it. Using the black box data it will be possible to pinpoint the coordinates of MH17 when it suffered the catastrophic event to within ten metres, and to demonstrate that this was not caused by any mechanical or electrical failure, or pilot error.</p>
<h2>The missile system</h2>
<p>Intelligence reports have indicated that the missile that brought down MH17 was a Russian <a href="http://www.janes.com/article/40907/missile-profile-9k37-buk">Buk M1 Self Propelled Air Defence System</a>, also known by its NATO reporting name of SA-11 “Gadfly”. Introduced in 1979, its latest updated version is known as the SA-17 “Grizzly”. Both are used by Russia, most of the ex-Warsaw Pact countries, and other nations to which Russia exports weapons. </p>
<figure class="align-left zoomable">
<a href="https://images.theconversation.com/files/56734/original/39yfjv94-1408377200.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/56734/original/39yfjv94-1408377200.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=237&fit=clip" srcset="https://images.theconversation.com/files/56734/original/39yfjv94-1408377200.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=668&fit=crop&dpr=1 600w, https://images.theconversation.com/files/56734/original/39yfjv94-1408377200.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=668&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/56734/original/39yfjv94-1408377200.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=668&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/56734/original/39yfjv94-1408377200.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=839&fit=crop&dpr=1 754w, https://images.theconversation.com/files/56734/original/39yfjv94-1408377200.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=839&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/56734/original/39yfjv94-1408377200.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=839&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">A Buk M1 surface-to-air missile launcher.</span>
<span class="attribution"><a class="source" href="http://commons.wikimedia.org/wiki/File:Buk-M1-2_9A310M1-2.jpg">.:Ajvol:.</a></span>
</figcaption>
</figure>
<p>The <a href="http://www.armyrecognition.com/russia_russian_missile_system_vehicle_uk/9k37_buk-m1_sa-11_gadfly_technical_data_sheet_specifications_information_description_pictures_photos.html">SA-11/SA-17</a> is mounted on tracked vehicles, making it easy to move. Importantly its radar is capable of IFF – that is, identifying an aircraft as friend or foe, or whether it is a commercial – using the same secondary radar transponder as air traffic control. However, the designers also implemented a backup mode that allows missile targeting to operate autonomously, bypassing the IFF safety feature. Used like this, the radar will show all targets in range. In this case, the operator selects one just presses the fire button. Only very basic training is required to operate the system like this. </p>
<p>The missile has a range of 42km (26 miles) and an operational ceiling of 25km (82,000ft). With a speed of 850 metres/sec (1,900mph), the missile could reach MH17 from launch in 11.5 seconds. The warhead is fitted with a proximity fuse which activates 100-300ft (30-90 metres) from its target. This fires a fragmentation charge which results in thousands of pieces of shrapnel accelerating in an spread pattern. The SA-11 is reputed to have a kill rate of 95%, so it is more than capable of downing MH17 even with virtually untrained operators.</p>
<h2>The story told by the wreckage</h2>
<p>The crash site covers an area of around 20 square kilometres near Grabovo, although larger parts of the aircraft are found in a much smaller area. There is little argument that the plane was shot down, so the investigation will focus on the shrapnel damage. By examining the position and angle of holes in the aircraft’s fuselage it will be possible to tell the approximate direction and angle from which the missile hit the airliner. With this information, investigators will be able to trace the reverse trajectory to locate the missile’s point of launch to within 100 metres. </p>
<figure class="align-left zoomable">
<a href="https://images.theconversation.com/files/56735/original/t4srrzx5-1408377436.PNG?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/56735/original/t4srrzx5-1408377436.PNG?ixlib=rb-1.1.0&q=45&auto=format&w=237&fit=clip" srcset="https://images.theconversation.com/files/56735/original/t4srrzx5-1408377436.PNG?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=685&fit=crop&dpr=1 600w, https://images.theconversation.com/files/56735/original/t4srrzx5-1408377436.PNG?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=685&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/56735/original/t4srrzx5-1408377436.PNG?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=685&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/56735/original/t4srrzx5-1408377436.PNG?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=861&fit=crop&dpr=1 754w, https://images.theconversation.com/files/56735/original/t4srrzx5-1408377436.PNG?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=861&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/56735/original/t4srrzx5-1408377436.PNG?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=861&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">The fuselage will be peppered with holes and fragments, which will aid investigators.</span>
<span class="attribution"><span class="source">Author provided</span></span>
</figcaption>
</figure>
<p>Chemical residues on pieces of shrapnel found among the wreckage will confirm whether the warhead was a SA-11/SA-17 missile. And metallurgy analysis of the shrapnel may even yield important information on the specific production batch – does it match the batch of systems supplied to Ukraine, or is it from another unknown batch, potentially from Russia?</p>
<h2>Assembling the Evidence</h2>
<p>So pull all the evidence together – what do you have?</p>
<ul>
<li>The data from air traffic control’s secondary surveillance radar provides MH17’s course and coordinates to within 500 metres.<br></li>
<li>The flight data recorder can show whether or not a component or systems failure, or pilot error that caused the loss, and can pinpoint the aircraft’s position even more accurately.</li>
<li>The specifications of the SA-11/SA-17 missile launcher demonstrates that it is quite capable of destroying an airliner, and that it can if necessary be operated by an inexperienced team, without the safety of an IFF-capable radar.</li>
<li>Evidence gathered from the wreckage will be able confirm whether the missile was an SA-11/SA-17, and help pinpoint the location of the missile when it was fired, and perhaps even confirm if the system was exported abroad or produced for Russian use.</li>
</ul>
<p>The evidence is there to be found, and once all of it is available, it will provide a compelling case against whoever fired upon and brought down flight MH17.</p><img src="https://counter.theconversation.com/content/30633/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>David Stupples does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>More than a month has passed since Malaysia Airlines flight MH17 crashed with the loss of all 298 lives on board. But despite the disturbances at the crash site near the small town of Grabovo, near Donetsk…David Stupples, Professor of Electrical and Electronic Engineering, City, University of LondonLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/295552014-07-22T13:57:29Z2014-07-22T13:57:29ZExplainer: what will MH17’s black boxes reveal?<p>The news that the black boxes from Malaysia Airlines flight M17 has been <a href="http://www.theguardian.com/world/2014/jul/22/mh17-black-boxes-handed-over-pro-russia-rebels">handed over to Malaysian authorities</a>, will lead to many questions about what information they have to offer.</p>
<p>The two recorders have been in the hands of pro-Russia rebels in the Ukraine for at least some of the four days since the aircraft was shot down over Ukraine. The data inside should give a strong indication about what happened to the aircraft but the question now is whether that data has been tampered with.</p>
<h2>What’s in a black box?</h2>
<p>All commercial aircraft with a take-off weight over 5,700kg are required to carry a flight data recorder (FDR) and a cockpit voice recorder (CVR) – referred to as the black boxes.</p>
<p>The equipment is designed to withstand the forces of any crash impact, subsequent fire damage and water contamination from being submerged at the bottom of an ocean. That means that there is a good chance of recovering the data at the MH17 crash site.</p>
<p>Nowadays, the black boxes on most aircraft are digital, so that all recorded data is stored in a memory, not unlike a modern camera that stores its images in an internal memory or on a memory card. Before take-off, the equipment is reset and typically stores two hours of data which is then progressively overwritten during the flight, so that it will contain two hours of data stored prior to an accident.</p>
<p>The cockpit voice recorder records all communications on the flight deck, including all transmissions with air traffic controllers, discussion between the flight crew, cabin announcements and conversations with any other crew entering the cabin. This information is carefully time-stamped and after an accident can provide valuable insight into the procedures followed and, to some degree, the thinking of the flight crew as they deal with an emergency.</p>
<p>The flight data recorder records the essential flight data parameters. The data is recorded at least 10 times per second and defines the aircraft’s flight path and motion. There are between 50 and 100 parameters recorded including the primary information: position, altitude, airspeed and heading to enable the flight path to be reconstructed. The data also includes other essential aerodynamic and engine parameters, including information derived by the air data computer from sensors and the engine data from the <a href="http://www.aboutflight.com/handbook-of-aeronautical-knowledge/ch-6-aircraft-systems/full-authority-digital-engine-control-fadec">FADEC engine control subsystem</a>.</p>
<p>The pilot primary controls, secondary actuator positions (selectors for flaps, undercarriage and fuel) and avionics settings are also stored, defining the complete state of the aircraft prior to an accident.</p>
<h2>Investigating the scene</h2>
<p>In most countries, following an accident, an air accident investigation team is contacted immediately to take control of the investigation, which is potentially a crime scene. In countries lacking such facilities, they will often request the <a href="http://www.ntsb.gov/">NTSB</a> from the USA or the <a href="http://www.aaib.gov.uk/home/index.cfm">AAIB</a> from the UK to undertake the investigation.</p>
<p>Apart from initial efforts of emergency teams to recover casualties, the site should not be touched until the investigation starts.</p>
<p>Clearly, recovery of the FDR and CVR is critical to any investigation. Normally, the recorders are closely guarded and sent to an approved laboratory where the data is carefully recovered. Once this data is copied the recorders have done their job, but confidence in the integrity of the raw data and the copying process is essential. The data extracted from the recorders is analysed to play back the last few minutes of a flight or monitor critical parameters or crew activity.</p>
<h2>Trusting the data</h2>
<p>A major area of concern is the integrity of the process to recover the data, particularly, as in the case of MH17, the outcome of an investigation may have political ramifications. The reputation of the airline or even the manufacturer may also be at stake. There are several points at which the process of data recovery is vulnerable.</p>
<p>For a start, it’s possible to substitute the black box for another, identical recorder which holds a different set of data. This would be difficult for the rebels to do in such a short period of time but they could carefully swap labels and plates bearing serial numbers to make a substitute box indistinguishable from the original.</p>
<p>More simply though, the the data stored in the memory of a recorder can be erased, overwritten or modified to give an alternative sequence of events. Data in a black box is not encrypted so overwriting its memory would be a straightforward operation.</p>
<p>The flight data recorder on flight MH17 should show, quite conclusively, if the aircraft had a major structural failure, an engine failure, a fire, an internal explosion or was struck by a missile. It may also show reveal the force and direction of any such impact.</p>
<p>However, even though the rebels could in theory tamper with the black boxes, it’s difficult to see what benefits there would be from either hiding the data or tampering with it, other than to add to the overall confusion. The rebels have faced intense scrutiny and were suspected of removing the recorders from the crash site without authorisation so they must know that the data in these boxes will come under extremely detailed analysis. </p>
<p>But while a huge amount of attention has been given to the black boxes in this case, it may, in the end, not come down to the information they yield. The damage done to the airframe of MH17 could tell us exactly who is responsible for this tragedy. The explosive force of the impact, the missile fragments and the explosive material embedded in the airframe will be analysed by forensic scientists who should be able to identify the type of missile used to bring it down and who manufactured it. Tampering with evidence on that scale is a much harder task.</p><img src="https://counter.theconversation.com/content/29555/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>David Allerton has received funding from EPSRC to investigate future aeronautical technology.
I currently serve on Council of the Royal Aeronautical Society.</span></em></p>The news that the black boxes from Malaysia Airlines flight M17 has been handed over to Malaysian authorities, will lead to many questions about what information they have to offer. The two recorders have…David Allerton, Emeritus Professor of Computer Systems Engineering, University of SheffieldLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/294262014-07-18T13:07:31Z2014-07-18T13:07:31ZBlack box scuffle could mask the truth about flight MH17<p>Normally, when a crashed aircraft’s black box is recovered, it can reveal <a href="http://www.popularmechanics.com/technology/aviation/crashes/what-really-happened-aboard-air-france-447-6611877">exactly what happened</a> to cause the disaster but in the case of MH17, the political instability between Ukraine and Russia could hamper operations.</p>
<p>If the black box of MH17 is detained by an organisation in military and political dispute with others, the creditability of the analysis results may be in doubt. This case is different from that of <a href="https://theconversation.com/if-wed-used-the-cloud-we-might-know-where-mh370-is-now-24542">MH370</a>, where the black box has yet to be found, but the idea is the same. The black box is vital, as is the integrity of the evidence it contains. It is how we establish the provenance of the facts used in <a href="https://learn2.open.ac.uk/course/view.php?id=201963">digital forensic analysis</a>.</p>
<p>That means that the black box data needs to be verified. If cloud-computing technology has been used, the verification can be done by cross-validating the flight status with other publicly available flight information. Recently at Microsoft, researchers have used cloud computing technology to predict the wind speeds with better accuracy when crowdsourcing the information with <a href="http://research.microsoft.com/en-us/news/features/windflow-060314.aspx">real-time flight data</a>. </p>
<p>But if cloud-computing technology is not used, the public must rely on whoever found the black box to truthfully relay the information inside. They might not release the raw data: as has happened in the past, the investigator may choose to reveal only a transcript or partial snippets of the sound tracks.</p>
<p>There are lots of forces at play in this evolving situation. It is not yet clear whether MH17’s black box has been found. Some reports have it on its way to <a href="http://www.telegraph.co.uk/news/worldnews/europe/ukraine/10975264/Problems-loom-for-investigators-as-Flight-MH17-black-box-may-be-in-Moscow.html">Russia</a> while others place two black boxes in the hands of <a href="http://www.independent.co.uk/news/world/europe/malaysia-airlines-flight-mh17-rebels-recover-second-black-box-from-crash-site-9614158.html">Ukranian emergency services</a>. The potential for problems is obvious. If both sides of the conflict produce their own analysis, who is to be believed? If only one has a black box, can we be sure of its analysis of the data? </p>
<p>The case of MH17 is exactly the same as MH370 in one respect – they should both have us thinking about how we monitor aircraft.</p>
<p>Real-time rather than offline verification would enable us to gather much the same information as that contained in the black box and allow us to cross-check it across sources more easily. </p>
<p>There would have been other flights nearby which can cross validate some of the data. Take speed for example – wind speed against those neighbouring flights could be used to indicate how fast MH17 was going. And because the real-time speed data is in the public domain – <a href="http://www.fly.faa.gov/flyfaa/usmap.jsp">FAA</a> offers access to them, for example – it would be hard, if not impossible, for whoever has the recorders of the individual flight to tamper with information about all the nearby flights.</p>
<p>Modifying flight data requires a substantial amount of computation. Making one lie is relatively easy but to cover up a lie that could be cross-checked by others, such as those operating neighbouring flights, would be much harder. </p>
<p>If we want the truth about MH17, we may have to look beyond the black box at data held elsewhere. Data is more powerful when it comes from a crowd, especially when the facts are likely to be disputed.</p><img src="https://counter.theconversation.com/content/29426/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Yijun Yu does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>Normally, when a crashed aircraft’s black box is recovered, it can reveal exactly what happened to cause the disaster but in the case of MH17, the political instability between Ukraine and Russia could…Yijun Yu, Senior Lecturer, Department of Computing and Communications, The Open UniversityLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/293962014-07-18T07:19:24Z2014-07-18T07:19:24ZWho handles the investigation on downed flight MH17?<p>The investigation into what happened to Malaysia Airlines Flight MH17 and the location of the black box recorder comes down to a series of complex treaties and international law.</p>
<p>Calls are being made by a number of countries – including Australia – for an international investigation of the crash of Malaysia Airlines MH17 with 298 people on board, including at least 28 Australians. But it should not be overlooked that there is already an international legal regime in place that mandates an investigation.</p>
<p>The obligation under the <a href="http://www.icao.int/publications/pages/doc7300.aspx">Convention on International Civil Aviation</a> (the “Chicago Convention”) is for the Ukraine to conduct the air crash investigation in to the downed Boeing 777-200.</p>
<p>Article 26 of the Chicago Convention vests the Ukraine with the responsibility to conduct such an investigation in providing that:</p>
<blockquote>
<p>[…] the state in which the accident occurs will institute an inquiry into the circumstances of the accident.</p>
</blockquote>
<h2>Who gets the black box?</h2>
<p>Here the event appears to have occurred in sovereign Ukrainian air space and the debris is littered within its territorial borders.</p>
<p>As a consequence of this obligation the <a href="https://theconversation.com/topics/black-box">black box</a> <a href="https://theconversation.com/the-black-box-an-australian-invention-that-nearly-didnt-happen-25435">flight data recorders</a> – reported to be in <a href="http://www.smh.com.au/world/mh17-wheres-the-black-box-from-the-malaysia-airlines-plane-shot-down-over-ukraine-20140718-zub61.html">the hands of the Russians</a> – should be turned over to the Ukrainian authorities.</p>
<p>If the Russians do have the black box and refuse to hand it over they would be in breach of their treaty obligations and there would no doubt be an international outcry. But whoever has the black box, it needs to go to the Ukrainians first for analysis as part of any investigation.</p>
<p>Malaysia also has a right to be a party to the investigations as the flag state of the aircraft. Article 26 of the Chicago Convention also provides that:</p>
<blockquote>
<p>[the] State in which the aircraft is registered shall be given the opportunity to appoint observers to be present at the inquiry.</p>
</blockquote>
<p>In conducting its investigation the Ukraine may call upon other States to assist and it is likely in these circumstances that a number of other States including, possibly, the Netherlands, Australia and other European States may provide support and assistance in the conduct of the investigation.</p>
<h2>MH17 different to MH370</h2>
<p>There is a difference between the situation involving the loss of <a href="https://theconversation.com/topics/flight-mh370">Malaysia Airlines MH370</a> in international airspace over international waters and the situation with MH17.</p>
<p>With MH370 – another Boeing 777-200 – because of where it is believed the plane went down, the state vested with the right to investigate is Malaysia as the State of registry of the aircraft under Annex 13, Chapter 5, 5.2 of the Chicago Convention.</p>
<p>The latest tragic event raises a number of complex legal issues including the position of relatives of the deceased passengers – including at least 28 Australians – under the <a href="http://www.iata.org/policy/Pages/mc99.aspx">Montreal Convention of 1999</a> and the relevance of <a href="http://www.un.org/en/sc/ctc/docs/conventions/Conv3.pdf">Convention for the Suppression of Unlawful acts Against the Safety of Civil Aviation</a> (Montreal, 1971).</p>
<p>The latter’s Article 1 creates an offence for individuals who engage in such acts that unlawfully and intentional destroys an aircraft in service.</p>
<p>There is also the potential issue of liability at State level for State-sponsored insurgency if sufficient evidence is uncovered along those lines.</p>
<h2>It’s happened before</h2>
<p>There are two prior incidents that are loosely analogous to the situation of MH17.</p>
<p>The first was the downing of <a href="http://www.history.com/this-day-in-history/korean-airlines-flight-shot-down-by-soviet-union">Korean Airlines Flight KAL007</a> by a Russian military jet in the Sea of Japan. This led to amendment of the Chicago Convention through the insertion of an <a href="http://www.icao.int/secretariat/legal/Administrative%20Packages/3bis_en.pdf">Article 3 bis</a> which prohibits the use of force against civilian aircraft.</p>
<p>Arguably the closest analogy though is the incident involving the misidentification of and <a href="http://news.bbc.co.uk/onthisday/hi/dates/stories/july/3/newsid_4678000/4678707.stm">shooting down of Iran Air flight 655</a> by the USS Vincennes with the loss of 290 lives during the Iran/Iraq war.</p>
<p>The dispute associated with this event was heading for the International Court of Justice before it was withdrawn with the agreement of both parties.</p>
<p>The Reagan administration subsequently made <em>ex gratia</em> payments of US$300,000 per head to the families of the deceased passengers without ever accepting legal liability for the incident.</p>
<p>Perhaps, if sufficient evidence points in Russia’s direction, it might save face with the international community if it contemplates the making of such payments. </p>
<p>It is perhaps too early in proceedings to contemplate this course of action as we need the investigation to proceed.</p><img src="https://counter.theconversation.com/content/29396/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Vernon Nase does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>The investigation into what happened to Malaysia Airlines Flight MH17 and the location of the black box recorder comes down to a series of complex treaties and international law. Calls are being made by…Vernon Nase, Associate Professor of Law, Curtin UniversityLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/254352014-04-10T20:37:36Z2014-04-10T20:37:36ZThe black box: an Australian invention that nearly didn’t happen<figure><img src="https://images.theconversation.com/files/45944/original/t3ygx6g4-1397021731.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=496&fit=clip" /><figcaption><span class="caption">Inside an early version of the black box flight recorder.</span> <span class="attribution"><span class="source">DSTO</span></span></figcaption></figure><p>The search operation for missing <a href="https://theconversation.com/topics/flight-mh370">flight MH370</a> continues to <a href="http://www.news.com.au/travel/travel-updates/missing-malaysia-airlines-flight-mh370-more-signals-detected/story-fnizu68q-1226878441833">hear signals</a> that could be from the plane’s <a href="https://theconversation.com/what-a-black-box-can-tell-us-about-missing-flight-mh370-25330">black box recorders</a>.</p>
<p>It’s only when those recorders are recovered that investigators will be able to find out what happened to the Malaysia Airlines Boeing 777 and the 239 passengers and crew.</p>
<p>But how did flight recorders come about in the first place? While they were invented in Australia, their adoption by the air industry was far from smooth or straightforward.</p>
<p>Here the story is told by Bill Schofield (co-author of this article) who worked for a time with David Warren, the inventor of the black box.</p>
<h2>Why are our planes crashing?</h2>
<p>The most curious aspect of the development of the black box flight recorder was the widespread resistance to its adoption, as it encapsulated the most fundamental tenet of scientific inquiry – gathering reliable data to draw conclusions.</p>
<p>It arose out of boredom during a meeting in the 1950s in Canberra. Experts from the Aeronautical Research Laboratory (ARL; now part of the Defence Science and Technology Organisation or <a href="http://www.dsto.defence.gov.au/">DSTO</a>) were trying to find why the new British aircraft – the Comets – were crashing.</p>
<p>David Warren was a chemist employed to conduct research on fuels for the new gas turbines that were entering aeronautical service at that time. He’d been asked to calculate what the effect would be if the fuel tanks on the Comets were blowing up.</p>
<figure class="align-right zoomable">
<a href="https://images.theconversation.com/files/46082/original/65hqd265-1397105694.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/46082/original/65hqd265-1397105694.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=237&fit=clip" srcset="https://images.theconversation.com/files/46082/original/65hqd265-1397105694.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=728&fit=crop&dpr=1 600w, https://images.theconversation.com/files/46082/original/65hqd265-1397105694.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=728&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/46082/original/65hqd265-1397105694.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=728&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/46082/original/65hqd265-1397105694.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=914&fit=crop&dpr=1 754w, https://images.theconversation.com/files/46082/original/65hqd265-1397105694.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=914&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/46082/original/65hqd265-1397105694.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=914&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">David Warren.</span>
</figcaption>
</figure>
<p>His answer was that it would not explain the damage of a recovered crashed Comet. While other experts speculated on possible causes during he had an obvious (in hindsight) thought – what they needed was data.</p>
<p>So he went back to his lab and wrote a very short tech memo on the need to record data in aircraft that would assist crash investigation. For the technology of the day, recording all aircraft flight data was impossibility.</p>
<p>Dr Warren thought that the Comet pilots would have known what was wrong with their aircraft, which is why black boxes include a cockpit voice recorder.</p>
<p>(It was eventually determined that the Comets crashed due to a fatigue crack at a square window corner, near the radio direction-finding aerial situated in the roof. The crack would have catastrophically torn, an event the pilot would have not likely sees.)</p>
<h2>Recorders need to survive the crash</h2>
<p>In the early 1950s none of the recording tapes would survive a burning plane crash, but at a trade fair Dr Warren saw the first of the wire voice recorders. The Miniphon wire recorder was the basis of Dr Warren’s first elemental recorder – now in the DSTO library at Port Melbourne (pictured, top).</p>
<p>He found that capturing clear records of cockpit conversation from microphones in the instrument panel and overhead was anything but easy but over time, he developed usable techniques.</p>
<p>Working with instrument colleagues at ARL they found a way of putting flight data as well as voice recording on the wire. This led to the construction of a much improved version in the late 1950s which was very advanced for its time.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/46079/original/sn3v6zgf-1397101028.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/46079/original/sn3v6zgf-1397101028.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/46079/original/sn3v6zgf-1397101028.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=492&fit=crop&dpr=1 600w, https://images.theconversation.com/files/46079/original/sn3v6zgf-1397101028.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=492&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/46079/original/sn3v6zgf-1397101028.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=492&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/46079/original/sn3v6zgf-1397101028.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=619&fit=crop&dpr=1 754w, https://images.theconversation.com/files/46079/original/sn3v6zgf-1397101028.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=619&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/46079/original/sn3v6zgf-1397101028.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=619&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">The preproduction model (left) with the original experimental prototype (right).</span>
<span class="attribution"><span class="source">Bill Schofield</span></span>
</figcaption>
</figure>
<p>Dr Warren showed remarkable tenacity in the black box development; he was a chemist engaged to and under continual pressure to focus on fuels and pass his black box invention over to the instrumentation section.</p>
<p>While others could develop the box it was Dr Warren who tried to get it adopted – and against unbelievable resistance. In an official letter from the Air Force rejecting the suggestion of putting boxes on RAAF planes it was stated that Dr Warren’s voice recorder would yield “more expletives than explanations”.</p>
<p>After the <a href="http://aviation-safety.net/database/record.php?id=19600610-1">fatal crash</a> of a Fokker Friendship approaching Mackay airport in Queensland in 1960, Justice Spicer, chairing the Board of Inquiry, stated that black boxes should be installed in commercial aircraft.</p>
<p>But the Australian Department of Civil Aviation purchased a US system instead of Dr Warren’s.</p>
<p>The US system proved useless in a subsequent air crash investigation. A commercial opinion of the day said the worldwide market would be as little as six boxes per year as they would only be installed on experimental aircraft during proving flights.</p>
<p>The Defence Department declined to patent the device as it saw little commercial justification for the cost of A£2,000.</p>
<h2>A chance meeting</h2>
<p>The whole flight recorder project had languished until 1958 when Laurie Coombes, then director of the ARL, wished to fill a gap in the schedule of a visiting UK official Sir Robert Hardingham, the Secretary of the British Air Registration Board. Coombes introduced him to Dr Warren who talked about the black box, and not fuels research.</p>
<p>Sir Robert thought it was an excellent idea and Dr Warren was soon flying to the UK to present it to the Royal Aeronautical Establishment and a few UK commercial instrument makers.</p>
<p>He came home through the US visiting a number of aeronautical establishments and commercial companies – none of which were at all interested. The first black boxes were initially produced in the UK with acknowledgement of its Australian origins but these acknowledgements soon disappeared.</p>
<h2>A lost invention</h2>
<figure class="align-left zoomable">
<a href="https://images.theconversation.com/files/45970/original/4jv9n8wp-1397030237.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/45970/original/4jv9n8wp-1397030237.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=237&fit=clip" srcset="https://images.theconversation.com/files/45970/original/4jv9n8wp-1397030237.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=593&fit=crop&dpr=1 600w, https://images.theconversation.com/files/45970/original/4jv9n8wp-1397030237.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=593&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/45970/original/4jv9n8wp-1397030237.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=593&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/45970/original/4jv9n8wp-1397030237.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=745&fit=crop&dpr=1 754w, https://images.theconversation.com/files/45970/original/4jv9n8wp-1397030237.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=745&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/45970/original/4jv9n8wp-1397030237.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=745&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">David Warren with a later version of his famous black box flight recorder.</span>
<span class="attribution"><a class="source" href="http://www.dsto.defence.gov.au/news/6237/">DTSO</a></span>
</figcaption>
</figure>
<p>In 1965, cockpit voice recorders were mandated in all commercial aircraft built in the US and the western world followed.</p>
<p>The IP rights of Australia to the invention were, by this time, compromised, but in recognition of the background IP the Department of Defence was paid UK£1,000.</p>
<p>I joined ARL in 1965 when the work on the black box was winding down but Dr Warren often talked to me at length about the history of the black box. </p>
<p>He wasn’t angry that he’d not been taken seriously by the powers-that-be, but more upset that Australia missed out in exploiting an invention which is, today, in hundreds of thousands of aircraft.</p>
<p>After the black box he started work in the early 1960s on fuel cells – the type now being installed in submarines. Again, though the powers-that-be could not at that time see any application for fuel cells so he was moved on again and started analysing the Earth’s need for energy and the likely sources of such energy.</p>
<p>His public lectures were well attended and would be highly relevant today. He was a free spirit with an unusual vision. He died in July 2010, at the age of 85.</p>
<p><br></p>
<p><em>This article was co-authored with Bill Schofield. Bill’s first scientific job was David Warren’s assistant in the mid-1960s.</em></p><img src="https://counter.theconversation.com/content/25435/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>John Faulkner does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>The search operation for missing flight MH370 continues to hear signals that could be from the plane’s black box recorders. It’s only when those recorders are recovered that investigators will be able…John Faulkner, Senior Visiting Fellow, UNSW SydneyLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/253302014-04-09T20:46:43Z2014-04-09T20:46:43ZWhat a black box can tell us about missing flight MH370<p>As the <a href="http://www.news.com.au/world/race-against-time-to-locate-black-box-of-missing-malaysia-airlines-flight-mh370/story-fndir2ev-1226875211181">search continues</a> for wreckage from the missing Malaysia Airlines plane it’s probable that answers surrounding the mystery of <a href="https://theconversation.com/topics/flight-mh370">flight MH370</a> will not be available until the <a href="http://www.atsb.gov.au/publications/2014/black-box-flight-recorders.aspx">recorders are recovered</a>.</p>
<p>There are two main recorders and the main flight data recorder is vital as it has a virtually complete record of all aircraft parameters until a crash, when it loses aircraft power.</p>
<p>It should record the last 25 hours of flight operation with a minimum requirement to measure data across five areas:</p>
<ul>
<li>pressure altitude</li>
<li>indicated airspeed</li>
<li>magnetic heading</li>
<li>normal acceleration</li>
<li>microphone keying (to link it to the second, cockpit recorder).</li>
</ul>
<p>It’s an extremely sturdy piece of equipment and will survive almost any kind of crash. It’s placed at the back of the aircraft as generally aircraft crash nose-first.</p>
<h2>Why is it called a black box?</h2>
<p>The name “black box” survives from the second world war when all advanced electronic devices in aircraft were secret and were either painted black or made from black materials. They’re usually called “crash recorders” in the industry and since the 1960s they’ve been painted bright orange to make them easier to find.</p>
<p>If the recorders end up in the sea, the salt water releases a device – the underwater location beacon (ULB) – that is then capable of sending a sonar signal for up to one month. The frequency of the beacon is 37.5kHz which is identical to that <a href="http://www.theguardian.com/world/2014/apr/06/mh370-search-continues-after-black-box-ping-claim">heard by rescue authorities</a> this week.</p>
<p>This sonar signal can be heard underwater and picked up by systems on all warships and specially fitted ships:</p>
<ul>
<li>the time taken for a sonar signal to hit a target and return gives its distance</li>
<li>change in pitch on return of the signal gives a measure of movement.</li>
</ul>
<p>The challenge, though, is the beacon’s battery power. It must last a minimum of 30 days and although it can last longer, it still means a race against time for authorities before the pinger goes flat. With more modern battery systems, it may be time to revisit and extend the life of this device.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/45946/original/6csgsr4b-1397021989.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/45946/original/6csgsr4b-1397021989.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/45946/original/6csgsr4b-1397021989.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=433&fit=crop&dpr=1 600w, https://images.theconversation.com/files/45946/original/6csgsr4b-1397021989.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=433&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/45946/original/6csgsr4b-1397021989.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=433&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/45946/original/6csgsr4b-1397021989.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=545&fit=crop&dpr=1 754w, https://images.theconversation.com/files/45946/original/6csgsr4b-1397021989.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=545&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/45946/original/6csgsr4b-1397021989.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=545&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">Flight recorder images courtesy of Australian Transport Safety Bureau.</span>
</figcaption>
</figure>
<h2>Conversations in the cockpit</h2>
<p>The other part of the black box is a cockpit voice recorder. This device records all conversation on the flight deck but it overrides the recording every two hours, so only the final two hours can be obtained. Again, following this case this could be extended by updating the equipment.</p>
<p>But there have been industrial problems. Pilots, like most people, guard their privacy and do not readily agree to have all their conversations recorded. Would you like your conversations – in their entirety – recorded on a long flight?</p>
<p>A number of airline systems can erase recordings after aircraft shutdown. The last two hours of flight MH370 may well be useful in giving some indication of what has happened, even if there is silence.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/45947/original/rcff2w2x-1397022237.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/45947/original/rcff2w2x-1397022237.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/45947/original/rcff2w2x-1397022237.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=276&fit=crop&dpr=1 600w, https://images.theconversation.com/files/45947/original/rcff2w2x-1397022237.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=276&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/45947/original/rcff2w2x-1397022237.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=276&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/45947/original/rcff2w2x-1397022237.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=347&fit=crop&dpr=1 754w, https://images.theconversation.com/files/45947/original/rcff2w2x-1397022237.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=347&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/45947/original/rcff2w2x-1397022237.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=347&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">Inside a flight recorder and the memory chip now used for recording information (left) replacing the tape that was once used (right).</span>
<span class="attribution"><span class="source">Australian Transport Safety Bureau</span></span>
</figcaption>
</figure>
<p>There were a number of questions raised by this case including over the transfer of MH370 between Malaysian airspace and that of Vietnam. When the aircraft entered Vietnamese airspace and <a href="http://www.thestar.com.my/News/Nation/2014/03/08/Missing-MAS-flight-Plane-went-missing-over-Vietnamese-airspace-Vietnam-government-says/">it did not call in</a>, this should have been picked up by both the Malaysians and Vietnamese.</p>
<p>Another confusing issue is the <a href="http://www.newscientist.com/article/dn25201-malaysian-plane-sent-out-engine-data-before-vanishing.html">engine monitoring by Rolls Royce</a> as part of the company’s maintenance programme.</p>
<p>The monitoring indicates to the flight deck that the engines are working properly – data that’s also transmitted via satellite to Rolls Royce and used for their maintenance.</p>
<p>But there has been little information about this, which <a href="http://www.independent.co.uk/news/world/asia/missing-flight-mh370-rolls-royce-dragged-into-the-mystery-as-rumours-surface-suggesting-that-data-from-the-planes-engines-showed-it-had-flown-for-four-hours-more-than-thought-9190622.html">leads to speculation</a> as to what happened. The data would, for instance, indicate when the engines were shut down or stopped due to a crash.</p>
<h2>What went wrong?</h2>
<p>So what did happen to the Boeing 777 carrying 239 passengers and crew on board a scheduled regular flight from Kuala Lumpur to Beijing?</p>
<p>Unfortunately in any accident involving humans, human error creeps in somewhere, whether it is in design, maintenance or operation.</p>
<p>We need to know <a href="https://theconversation.com/keep-calm-and-carry-on-dont-rush-the-investigation-of-mh370-24665">what those errors were</a> and then be in a position to improve safety systems to reduce the possibility of re-occurrence.</p>
<p>Suggested <a href="https://theconversation.com/mh370-should-make-us-rethink-how-we-monitor-planes-25302">changes on monitoring aircraft</a> are already being made and one technical improvement would be to transmit the flight recorder data to the ground on a full time basis. This would improve the efficiency of the operation as well as its safety benefits.</p>
<p>But until we can find the recorders, extract the data and determine the cause of the accident it’s not useful to hazard a guess as to what other issues would have helped in this instance.</p><img src="https://counter.theconversation.com/content/25330/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>John Faulkner does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.</span></em></p>As the search continues for wreckage from the missing Malaysia Airlines plane it’s probable that answers surrounding the mystery of flight MH370 will not be available until the recorders are recovered…John Faulkner, Senior Visiting Fellow, UNSW SydneyLicensed as Creative Commons – attribution, no derivatives.tag:theconversation.com,2011:article/250102014-03-31T02:59:17Z2014-03-31T02:59:17ZUnder the deep blue ocean: the search for MH370’s black box<p>As the effort to find Malaysian Airlines <a href="https://theconversation.com/topics/flight-mh370">flight MH370</a> moves inexorably towards the <a href="http://www.theguardian.com/world/2014/mar/31/equipment-to-detect-mh370-black-box-dispatched-after-more-objects-spotted?CMP=ema_632">recovery stage</a>, the challenge of finding the plane’s flight recorder (called the “black box” even though it’s actually <a href="http://www.abc.net.au/news/2014-03-26/black-box-flight-recorders/5343456">bright orange</a>) on the ocean floor will become a game of blind man’s bluff spread over thousands of kilometres.</p>
<p>The inability to penetrate the darkness of the ocean has long been a major handicap for deep-sea exploration. It is why we know more about the surface of the moon than the bottom of the deep ocean.</p>
<p>In a nutshell, the problem is that seawater is opaque to electromagnetic radiation, such as radio waves, microwaves and X-rays. This means that almost none of our normal techniques to “see” over long distances can be used in the ocean.</p>
<p>As you go deeper into the ocean, it gets darker. People who have dived know that colours are quite different once you get to 20m deep. Red starts to look green and, eventually, everything gets greyer.</p>
<p>Once you get down more than 100m, the ocean is essentially pitch black – leaving almost 4km of darkness in most parts of the ocean – because light is absorbed in seawater. </p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/45071/original/tmskm38k-1396224945.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/45071/original/tmskm38k-1396224945.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/45071/original/tmskm38k-1396224945.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=338&fit=crop&dpr=1 600w, https://images.theconversation.com/files/45071/original/tmskm38k-1396224945.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=338&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/45071/original/tmskm38k-1396224945.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=338&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/45071/original/tmskm38k-1396224945.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=424&fit=crop&dpr=1 754w, https://images.theconversation.com/files/45071/original/tmskm38k-1396224945.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=424&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/45071/original/tmskm38k-1396224945.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=424&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">The bright colour of black boxes doesn’t help when submerged in 4,000m of seawater.</span>
<span class="attribution"><a class="source" href="http://www.flickr.com/photos/avgeekjoe/13096509215/sizes/l">AvgeekJoe/Flickr</a>, <a class="license" href="http://creativecommons.org/licenses/by-sa/4.0/">CC BY-SA</a></span>
</figcaption>
</figure>
<p>It’s not only visible light that gets lost. All forms of electromagnetic radiation are sooner or later damped out.</p>
<p>Most of our techniques to remotely study objects (the moon, stars, but also clouds in the atmosphere or ships on the surface of the ocean) rely heavily on electromagnetic radiation. This type of radiation forms the basis of radar, <a href="https://theconversation.com/explainer-what-is-gps-12248">GPS</a> and essentially all telecommunication. </p>
<p>But oceanographers who want to study the deep ocean can’t use most common tools, such as satellites or radio waves.</p>
<h2>Listen to this</h2>
<p>The only signal that can travel through seawater is sound. Sound travels extremely well through seawater; much better than through air. This is why some whales and other marine animals can hear each other over <a href="http://www.sciencedaily.com/releases/2005/02/050223140605.htm">thousands of kilometres</a>.</p>
<p>It’s no surprise, then, that black boxes on aeroplanes have “pingers” that transmit a sound signal. But while sound travels long distances under water, it isn’t a very efficient way of transmitting information: you can’t pack too much information in it and it costs much more energy than electromagnetic radiation.</p>
<p>Another problem is that sound doesn’t travel in a straight line in the ocean and can reflect off sea mountains to create echoes, making it sometimes difficult to locate its origin. The speed and path of sound varies with temperature and salinity and it can bend up or down. In some extreme cases, it can even be reflected in the middle of the ocean. </p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/45078/original/h832c5ft-1396225949.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/45078/original/h832c5ft-1396225949.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/45078/original/h832c5ft-1396225949.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=393&fit=crop&dpr=1 600w, https://images.theconversation.com/files/45078/original/h832c5ft-1396225949.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=393&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/45078/original/h832c5ft-1396225949.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=393&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/45078/original/h832c5ft-1396225949.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=494&fit=crop&dpr=1 754w, https://images.theconversation.com/files/45078/original/h832c5ft-1396225949.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=494&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/45078/original/h832c5ft-1396225949.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=494&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">As sound bounces around under water it can change intensity and create confusing reflections.</span>
<span class="attribution"><span class="source">RA Zingarelli and DB King</span></span>
</figcaption>
</figure>
<p>In layers where the temperature and salinity change rapidly with depth, these layers act as mirrors and can cause some sounds to be completely lost. Naval submarines are extremely good at hiding themselves underneath these mirrors, making them all but invisible to ships on the surface.</p>
<p>Oceanographers have been struggling with this lack of a good way to transmit signals through the ocean for decades. It means that submarines can’t use GPS to get their position, and that data from subsurface instruments need to be either locally stored on logging devices (which then need to be retrieved) or transmitted via cables to land or ship.</p>
<h2>What happens when the pings stop?</h2>
<p>If the flight recorders can’t be found using the pinger, it is very likely the ocean floor will need to be scoured using <a href="http://oceanservice.noaa.gov/facts/sonar.html">sonar</a>. Again, this is technology based on sound, but is not nearly as accurate as, for instance, radar or photography of the land surface.</p>
<figure class="align-center zoomable">
<a href="https://images.theconversation.com/files/45105/original/z9x7zzcx-1396227716.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=1000&fit=clip"><img alt="" src="https://images.theconversation.com/files/45105/original/z9x7zzcx-1396227716.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&fit=clip" srcset="https://images.theconversation.com/files/45105/original/z9x7zzcx-1396227716.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=600&h=360&fit=crop&dpr=1 600w, https://images.theconversation.com/files/45105/original/z9x7zzcx-1396227716.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=600&h=360&fit=crop&dpr=2 1200w, https://images.theconversation.com/files/45105/original/z9x7zzcx-1396227716.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=600&h=360&fit=crop&dpr=3 1800w, https://images.theconversation.com/files/45105/original/z9x7zzcx-1396227716.jpg?ixlib=rb-1.1.0&q=45&auto=format&w=754&h=452&fit=crop&dpr=1 754w, https://images.theconversation.com/files/45105/original/z9x7zzcx-1396227716.jpg?ixlib=rb-1.1.0&q=30&auto=format&w=754&h=452&fit=crop&dpr=2 1508w, https://images.theconversation.com/files/45105/original/z9x7zzcx-1396227716.jpg?ixlib=rb-1.1.0&q=15&auto=format&w=754&h=452&fit=crop&dpr=3 2262w" sizes="(min-width: 1466px) 754px, (max-width: 599px) 100vw, (min-width: 600px) 600px, 237px"></a>
<figcaption>
<span class="caption">A diagram of seafloor sonar ‘pictures’ taken by a ship.</span>
<span class="attribution"><a class="source" href="http://www.flickr.com/photos/usoceangov/4167340394/sizes/l">NOAA's National Ocean Service/Flickr</a>, <a class="license" href="http://creativecommons.org/licenses/by/4.0/">CC BY</a></span>
</figcaption>
</figure>
<p>The real dilemma for the search team is how high above the ocean floor to take the sonar “pictures”:</p>
<ul>
<li>too high up and the plane wreck might be missed because the resolution is too poor</li>
<li>too close to the ocean floor and it will take much more time to map the entire area.</li>
</ul>
<p>In this sense, it doesn’t help that we don’t have any good maps of the ocean floor already. If we would have such maps, we could simply put post-plane crash maps on top of them and see where they differ. These locations could then be scrutinised for possible wreckage, and regions where there are no differences can initially be skipped.</p>
<p>The problems faced by the search time of the MH370 are very similar to those of the <a href="http://en.wikipedia.org/wiki/Air_France_Flight_447">Air France AF447 flight</a> that went down in the Atlantic Ocean in 2009. Back then, the search team was not able to locate the black boxes with a pinger signal. </p>
<p>The search took two years, and it was only through months of scouring the ocean floor with submarines and sonar that the black box was finally found.</p>
<p>Let’s hope MH370’s black box can be located sooner and provide some much-needed insight to the flight’s mystery.</p><img src="https://counter.theconversation.com/content/25010/count.gif" alt="The Conversation" width="1" height="1" />
<p class="fine-print"><em><span>Erik van Sebille receives funding from the Australian Research Council.</span></em></p>As the effort to find Malaysian Airlines flight MH370 moves inexorably towards the recovery stage, the challenge of finding the plane’s flight recorder (called the “black box” even though it’s actually…Erik van Sebille, Research Fellow and Lecturer in Oceanography, UNSW SydneyLicensed as Creative Commons – attribution, no derivatives.