Big Techday 5

We will see the creative processes to create these 3D animations: “Inspirations”, “Nature by Numbers”, “Isfahan” and “Snakes”, works where we find a connection between arts, maths, architecture, geometry, nature and of course: M. C. Escher. These creative processes are not exactly the same in all cases, but we can establish some common guideline

• Looking for ideas and inspiration.

• Searching visual references and documentation.

• Developing a screenplay and layouts.

• Research and Development (a very important part). When you make a 3D animation you will find technical and complex problems, for sure. And it’s a good idea to anticipate them and search for solutions before crashing with a “productive wall”.

• We will make a quick tour to the production process itself: modeling, texturing, lighting, animation, rendering and postproduction.

The talk will be absolutely based on images and video screenshots, starting with the shorts projection.

Every Internet start-up soon has to ask the question - where should we host our web application? Scalability is an important consideration for selecting a provider: if the new application has the desired success - or even exceeds all expectations - the young company's growth must not be inhibited by hardware bottlenecks. At the same time, financial constraints prevent large, up-front hardware investments. Providers of cloud-based, platform-as-a-service solutions offer a practical answer: the cloud provider supplies a complete runtime environment for web applications, administrating all resources while guaranteeing scalability. Only the computing time and accumulated storage space actually used are billed. This allows the application to grow with its users, keeping the operational costs in proportion to that growth. The start-ups development team can concentrate on application functionality instead of web server administration. In this talk we will focus on the concepts behind platform-as-a-service. We will examine the architecture of the Google App Engine and establish an understanding of the implications this architecture has on the development process. Additionally, we will explore several data storage interfaces available in the Google Cloud and how they provide scalability.

Steven Pinker is Harvard College Professor and Johnstone Family Professor of Psychology at Harvard University. He is the best-selling author of numerous books, including, most recently, "The Better Angels of Our Nature: Why Violence Has Declined", about which he will talk at the BTD5, and "The Stuff of Thought: Language as a Window into Human Nature". His previous works include "The Blank Slate" and "How The Mind Works", which were both Pulitzer Prize finalists. Dr. Pinker has been listed on  TIME magazine's "100 Most Influential People in The World," as well as on Foreign Policy’s list of "The World's Top 100 Public Intellectuals". He has received awards and honors from the National Academy of Sciences, the Cognitive Neuroscience Society, and the American Psychological Association.

Some insights into the THOR project will be presented, which involves the modelling and semi-autonomous control of an 18-ton mobile bucket excavator from Volvo Construction Equipment.

The continuously rising energy hunger of modern societies dictates the exploration of new energy sources. Nuclear fusion is one important future option to address this task. Similar to the sun, the fusion of hydrogen isotopes is intended to be used for energy production. If it were possible to open up this new primary energy source, a safe and environmentally sound energy source would have been created for a million years of more. Such nuclear fusion reactions can be created in very hot plasmas with temperatures of about 100 million degrees and plasma pressures of several bars. Contrary to the sun, the plasma is locked in by magnetic fields in order to isolate it from the cold walls of the container. Wendelstein 7-X is currently under construction in Greifswald and is scheduled to become operational in 2014. The novel concept promises the economic operation of later fusion power stations. Plasma confinement will last half an hour, enabled by superconducting magnetic coils, resulting in a significant improvement of the experiment compared to its predecessors. New challenges have to be faced for data acquisition and control, too. The plasma diagnostics are expected to create streaming datarates of 30 GBytes/s, which have to be permanently archived. Partially, they have to be analysed in real-time for control purposes. The typical doubling of the data rates within one to two years requires horizontally scalable data center structures. Additionally, the development of an intelligent and adaptive plasma control will be a subject of research, in order to stabilise the plasma against changes in the border conditions, for example magneto-hydrodynamic properties. Data analysis and modelling will be based on Bayesian methods to automate complex data analyses and to develop safety signals based on inexact sensory information.

On the 21st of October 2011, Thomas Senkel from the e-volo team succeeded in building the Volocopter VC1 after more than one year of development, the first manned flight with a solely electrical operated and vertically starting aircraft worldwide. The invention of the Volocopter and the first flight with the VC1 have been celebrated as pioneer work in aviation in numberless TV- and press releases all over the world.
In contrast to any other aircraft, the handling of the Volocopter is very easy. The automatic position control and the directional control (fly by wire via joystick) are carried out with several independent and one another monitoring on-board computers, regulating each electric motor in its agitator speed separately. Besides the convincingly easy architecture without complex mechanics (no pitch-control), the redundancy of the drives is another considerable advantage. Thus, even in case of a breakdown of several motors, the Volocopter can be landed safely.
In this talk there first will be given a short historic summary of the development of the first helicopter. The history of origins of the Volocopter VC1 is presented with many images. An insight in the underlying physics shall show, why the VC1 was configured and dimensioned as it is. At least, a future prospect in the planned further development of the Volocopter will be given.

Profiling is fun - and yields results.

Programming is hard. Programming correct C++ is particularly hard. Indeed, it is uncommon to see a screenful containing only well defined and conforming code. Why do professional programmers write code like this? Because most programmers do not have a deep understanding of the language they are using. While they sometimes know that certain things are undefined or unspecified, they often do not know why it is so. In this presentation we will study small code snippets in C++, and use them to discuss the fundamental building blocks, limitations and underlying design philosophies of this wonderful but dangerous programming language.

Possibly the greatest strength of a statically typed language is that it can prevent large classes of bugs from ever compiling. But even with a strongly typed language such as Haskell, APIs can be designed that do not properly take advantage of the type system, and therefore reduce the compiler from a powerful robustness tool to a nuisance. Based on experience in creating Yesod's type-safe interfaces for URLs, data persistence, and templating, Michael will discuss various techniques to make solid, reliable APIs, using such Haskell features as associated types, phantoms, and GADTs.

Kotlin is a new statically typed programming language developed by JetBrains, targeting JVM and JavaScript. Kotlin on the JVM is designed to be fully Java compatible, and at the same time safer, more concise than Java and not requiring a PhD to fully understand. Kotlin is developed with IDE support from the very beginning and is currently available as a preview release. During the presentation, we will give an overview of the language, look at some of its distinguishing features in more detail, and demonstrate the IDE support capabilities.

Dr. Eckart Pech is the Managing Director Service Technology/Chief Information Officer at Telefónica Germany. Prior to his appointment he served as Vice President Planning & Transformation and Technology Development at Telefónica Germany. Prior to his current position, Dr. Pech served as president and CEO of Detecon, Inc. in U.S.A., a subsidiary of Deutsche Telekom AG. He was responsible for the entire North and South American region with offices in Reston, San Francisco and Mexico City.  During this time, Dr. Pech was also a partner in the executive committee of Detecon International. Before joining Detecon, Dr. Pech was a Partner at Diebold, a consulting company and subsidiary of Daimler AG, with responsibility for the areas of mobile telecommunications and internet technologies. Dr. Pech has also held positions with Hewlett Packard and has served as a Business Consultant in China, where he advised foreign companies on direct investments in Asia. Dr. Pech has studied Business Administration and Chinese at the University of Bayreuth and the Shanghai International Studies University. He completed his dissertation at the University of Armed Forces in Munich, focusing on Product Development concepts for Electronic Commerce. Dr. Pech is a Young Leader Alumnus of the American Council on Germany and serves as a member of the foundation advisory board for the Gerald Asamoah Foundation, an organization for children with heart disease.

In our omnipresent information society, it is essential to process the rapidly growing, unstructured corporate data, so that users can access their relevant data as quickly as possible.
There are several open source software projects which can centralize heterogeneous corporate data that are distributed over a variety of IT systems. One important software platform is Apache Solr, which builds upon the Lucene search engine and simplifies development of Lucene-based applications.
This talk demonstrates how quick searches can be realized with the help of Solr. We will integrate Solr with Apache Nutch, which allows us to index, structure and display several million documents from the web usind commodity hardware. Today, these technologies are used successfully by thousands of companies, such as AOL, IBM, LinkedIn and MySpace.

Within the software development field, "big data" is used as a term to denote data volumes that grow so strongly and/or have to be analysed so fast, that novel approaches are required to deal with them. Using specific examples from the telecommunications field, we show how current technologies like Apache Hadoop or Amazon EC2 can be used to generate new insights - or even new business models - from existing data. We talk about some of the challenges faced during these projects, including their solution.

A fundemental difficulty with implementing a message-oriented interface to a third party are the unerderlying dynamics, which are often not fully described in the specification. This can lead to an unexpected or surprising message order, which must be countered with immediate changes to the application. The order interface of a large telecommunications operator presents such a message-oriented interface. The risk of interpreting the interface differently is increased by a mandatory compliance test which must be passed within certain time and quality constraints. To mitigate these risks during development of the SOAP interface at M-Net Telekommunikations GmbH, the project team switched their IT landscape from a manual deployment process to a automated delivery pipeline. To achieve this, the team extended the existing continuous integration environment to an automated delivery process. On each code commit, a full build - including unit and service tests, as well as automated static code analysis - is performed. After testing, the JEE application is automatically distributed to four separated servers. In this environment, automated integration tests run subsequently and verify the application against an interface simulator. Finally, the built artifacts can be deployed to production with a single click. Jenkins, Ant/Ivy and Artifactory provided the technological foundation. This talk demonstrates the obstacles the team encountered on its way to continuous delivery, and how they were overcome.

Almost all enterprise software projects are component-based. One would like to steadily test the dependencies of these components through early integration. At the same time, this often hinders the continuous building and promoting of the complete application. To find satisfying solutions that scale without switching to a late, mostly woebegone integration is an important factor for the project’s success. We consider typical sticky points as well as the latest solution strategies. As an example, we take a closer look at a real project with 500 developers and 50 components. We get to know an open source tool that can be used to implement the introduced solution strategies in a classy and powerful way.

Agile methods such as Scrum or Kanban are increasingly becoming mainstream. They are often perceived solely as a set of process mechanics. To gain the full benefit of agile methods, however, a holistic approach must be taken. Scrum and Kanban have their roots in the Japanese philosophy of "Kaizen" - better known in the western world by the more formal Continuous Improvement Process (CIP) of the 1950s. This talk introduces the principles of Kaizen and highlights the aspects relevant to software development. In addition to reflecting on the requirements for a fully functional Kaizen culture, some real-world examples will be presented.

David J. Anderson leads a management consulting firm focused on improving performance of technology companies. He has many years management experience leading teams on agile software development projects. David was a founder of the agile movement through his involvement in the creation of Feature Driven Development. He was also a founder of the APLN, a non-profit dedicated to improving management and leadership in technology companies. Recently David has been focusing his attention on business agility and enterprise scale agile software transitions through a synergy of the CMMI model for organizational maturity with Agile and Lean methods.

Many organizations are child-like. They plan the project as if nothing will go wrong, and when something does go wrong they act shocked and dismayed. Risk Management is not just worrying about your project, and it is not about running away from risk. Risk Management for software projects is all about when you make decisions and when you take action. When done to its greatest advantage, Risk Management starts before a project is even born. (How do projects get started in your organization?) In this talk Tim Lister will present the advantages, and the dangers, of practicing Risk Management in an adult-like fashion. He will offer up a process for you to consider tailoring for your organization, and he will discuss how your organization can grow up.

Scrum with one or two teams is well-proven in many cases and the Scrum framework gives detailed guidelines for this situation. For larger projects it provides less concrete guidance. By now there are concepts, which address the challenges of large-scale projects, such as Scrum of Scrums, feature teams, requirement areas, virtual teams, and dependencies amongst others. Which of these concepts work well and which do not, and how they can be put into practice, is discussed in the talk based on the experience from a project with 15 teams.

The vSphere Perl SDK enables you to create virtual machines entirely programmatically, Kickstart to install a guest operating system, and CFEngine to configure it. For the first time, the whole chain from creating the virtual machine to having a complete, running service can be automated by software. Now, data centers are no longer installed, configured and administrated - they are programmed. The core principles of software development can now be applied to administration. Test-driven development, continuous integration, testing and staging are now available to systems which were previously hand-crafted. All these principles are being implemented successfully at a project with a large, German internet service provider. This talk describes tools, design decisions, as well as what is possible today, and what isn't.

This talk is a practical introduction to node.js, as well as an overview of the applications that are easier to build with node than using other plattforms.

The web is changing so is our need for data, we want more and we want to have it in real time. In this talk Arnout Kazemier will be discussing experiences he has gained while developing real time applications.