Programming systems that obey hard real-time constraints is difficult. So is programming multiple CPUs that interact to solve a single problem.

On rare occasions it is possible to mix two difficult problems to create one easy problem and multi-CPU real-time is, on the face of it, just such an occasion. Give each deadline its own CPU and it will never be missed. This intuition is, unfortunately, incorrect, which does not, however, prevent it being tried in many real-time systems.

For three decades, fourth year students have been exploring this problem in CS452, using multiple tasks (virtual CPUs) running on a single CPU. It is now time to consider whether modern developments in CPU architecture make it possible to use multiple CPUs in CS452 given the practical constraint of a twelve week semester.

This talk will describe the nature of computation typical of real-time systems, architectural solutions currently employed in the course, and possible architectures for multi-CPU systems.

Veteran Mozilla engineer Ehsan Akhgari presents a talk on the internals of web browsers. The material ranges from the fundamentals of content rendering to the latest innovations in browser design.

Web browsers have evolved. From their humble beginnings as simple HTML rendering engines they have grown and evolved into rich application platforms. This talk will start with the fundamentals: how a browser creates an on-screen representation of the resources downloaded from the network. (Boring, right? But we have to start somewhere.) From there we'll get into the really exciting stuff: the latest innovations in Web browsers and how those innovations enable — even encourage — developers to build more complex applications than ever before. You'll see real-world examples of people building technologies on top of these "simple rendering engines" that seemed impossible a short time ago.

GPGPU (general purpose graphics processing unit) computing is an expanding area of interest, with applications in physics, chemistry, applied math, finance, and other fields. nVidia has created an architecture named CUDA to allow programmers to use graphics cards without having to write PTX assembly or understand OpenGL. CUDA is designed to allow for high-performance parallel computation controlled from the CPU while granting the user fine control over the behaviour and performance of the device.

In this talk, I'll discuss the basics of nVidia's CUDA architecture (with most emphasis on the CUDA C extensions), the GPGPU programming environment, optimizing code written for the graphics card, algorithms with noteworthy performance on GPU, libraries and tools available to the GPGPU programmer, and some applications to condensed matter physics. No physics background required!

Theoretical neuroscience is a new discipline focused on constructing mathematical models of brain function. It has made significant headway in understanding aspects of the neural code. However, past work has largely focused on small numbers of neurons, and so the underlying representations are often simple. In this talk I demonstrate how the ideas underlying these simple forms of representation can underwrite a representational hierarchy that scales to support sophisticated, structure-sensitive representations.

Theoretical neuroscience is a new discipline focused on constructing mathematical models of brain function. It has made significant headway in understanding aspects of the neural code. However, past work has largely focused on small numbers of neurons, and so the underlying representations are often simple. In this talk I demonstrate how the ideas underlying these simple forms of representation can underwrite a representational hierarchy that scales to support sophisticated, structure-sensitive representations. I will present a general architecture, the semantic pointer architecture (SPA), which is built on this hierarchy and allows the manipulation, processing, and learning of structured representations in neurally realistic models. I demonstrate the architecture on Progressive Raven's Matrices (RPM), a test of general fluid intelligence.

BareMetal is a new 64-bit OS for x86-64 based computers. The OS is written entirely in Assembly, while applications can be written in Assembly or C/C++. High Performance Computing is the main target application.

In this talk, I will go over the concepts used in video encoding (such as motion estimation/compensation, inter- and intra- frame prediction, quantization and entropy encoding), and then demonstrate these concepts and algorithms in use in the MPEG-2 and the H.264 video codecs. In addition, some clever optimization tricks using SIMD/vectorization will be covered, assuming sufficient time to cover these topics.

With the recent introduction of digital TV and the widespread success of video sharing websites such as youtube, it is clear that the task of lossily compressing video with good quality has become important. Similarly, the complex algorithms involved require high amounts of optimization in order to run fast, another important requirement for any video codec that aims to be widely used/adopted.

The CSC is happy to be hosting Jeff Potter, author of "Cooking for Geeks" for a presentation on the finer arts of food science. Jeff's book has been featured on NPR, BBC and his presentations have wowed audiences of hackers & foodies alike. We're happy to have Jeff joining us for a hands on demonstration.

But you don't have to take our word for it... here's what Jeff has to say:

Hi! I'm Jeff Potter, author of Cooking for Geeks (O'Reilly Media, 2010), and I'm doing a "D.I.Y. Book Tour" to talk about my just-released book. I'll talk about the food science behind what makes things yummy, giving you a quick primer on how to go into the kitchen and have a fun time turning out a good meal. Depending upon the space, I’ll also bring along some equipment or food that we can experiment with, and give you a chance to play with stuff and pester me with questions.

We develop a programming model built on the idea that the basic computational elements are autonomous machines interconnected by shared cells through which they communicate. Each machine continuously examines the cells it is interested in, and adds information to some based on deductions it can make from information from the others. This model makes it easy to smoothly combine expression-oriented and constraint-based programming; it also easily accommodates implicit incremental distributed search in ordinary programs. This work builds on the original research of Guy Lewis Steele Jr. and was developed more recently with the help of Chris Hanson.

I have stolen my title from the title of a paper given by Marvin Minsky in the 1960s, because it most effectively expresses what I will try to convey in this talk.

We have been programming universal computers for about 50 years. Programming provides us with new tools to express ourselves. We now have intellectual tools to describe "how to" as well as "what is". This is a profound transformation: it is a revolution in the way we think and in the way we express what we think.

For example, one often hears a student or teacher complain that the student knows the "theory" of some subject but cannot effectively solve problems. We should not be surprised: the student has no formal way to learn technique. We expect the student to learn to solve problems by an inefficient process: the student watches the teacher solve a few problems, hoping to abstract the general procedures from the teacher's behavior on particular examples. The student is never given any instructions on how to abstract from examples, nor is the student given any language for expressing what has been learned. It is hard to learn what one cannot express. But now we can express it!

Expressing methodology in a computer language forces it to be unambiguous and computationally effective. The task of formulating a method as a computer-executable program and debugging that program is a powerful exercise in the learning process. The programmer expresses his/her poorly understood or sloppily formulated idea in a precise way, so that it becomes clear what is poorly understood or sloppily formulated. Also, once formalized procedurally, a mathematical idea becomes a tool that can be used directly to compute results.

I will defend this viewpoint with examples and demonstrations from electrical engineering and from classical mechanics.

I'll survey the evolution of our computer science curriculum over the past thirty-five years to try to convey the reasons (not always entirely rational) behind our current mix of courses and their division into core and optional. After some remarks about constraints and opportunities in the near future, I'll open the floor to discussion, and hope to hear some candid comments about the state of CS at UW and how it might be improved.

About the speaker:

Prabhakar Ragde is a Professor in the School of Computer Science at UW. He was Associate Chair for Curricula during the period that saw the creation of the Bioinformatics and Software Engineering programs, the creation of the BCS degree, and the strengthening of the BMath/CS degree.

Bill Gates in his article "A Robot in every home" in the Scientific American describes how the current robotics industry resembles the 1970's of the Personal Computer Industry. In fact it is not just Microsoft which has already taken a step forward by starting the Microsoft Robotics studio, but robotics researchers around the world believe that robotics and automation systems are going to be ubiquitous in the next 10-20 years (similar to Mark Weiser's analogy of Personal Computers 20 years ago). Natural User Interfaces (NUIs) are going to revolutionize the way we interact with computers, cellular phones, household appliances, automated systems in our daily lives. Just like the GUI made personal computing a reality, I believe natural user interfaces will do the same for robotics.

During the presentation I will be presenting my ongoing software project on natural user interfaces as well as sharing my goals for the future, one of which is to provide an NUI SDK and the other to provide a common Robotics OS for every hardware vendor that will enable people to make applications without worrying about underlying functionality. If time permits I would like to present a demo of my software prototype.

Deep networks can be learned efficiently from unlabeled data. The layers of representation are learned one at a time using a simple learning module, called a "Restricted Boltzmann Machine" that has only one layer of latent variables. The values of the latent variables of one module form the data for training the next module. Although deep networks have been quite successful for tasks such as object recognition, information retrieval, and modeling motion capture data, the simple learning modules do not have multiplicative interactions which are very useful for some types of data.

The talk will show how a third-order energy function can be factorized to yield a simple learning module that retains advantageous properties of a Restricted Boltzmann Machine such as very simple exact inference and a very simple learning rule based on pair-wise statistics. The new module contains multiplicative interactions that are useful for a variety of unsupervised learning tasks. Researchers at the University of Toronto have been using this type of module to extract oriented energy from image patches and dense flow fields from image sequences. The new module can also be used to allow motions of a particular style to be achieved by blending autoregressive models of motion capture data.

Paul Lutus describes his early Apple II software development days, conducted from the far end of a 1200-foot power cord, in a tiny Oregon cabin. Paul describes how he wrote a best-seller (Apple Writer) in assembly language, while dealing with power outages, lightning storms and the occasional curious bear.

Paul also describes his subsequent four-year solo around-the-world sail in a 31-foot boat. And be ready with your inquiries -- Paul will answer your questions.

Paul Lutus has a wide background in science and technology. He designed spacecraft components for the NASA Space Shuttle and created a mathematical model of the solar system used during the Viking Mars lander program. Then, at the beginning of the personal computer revolution, Lutus switched career paths and took up computer science. His best-known program is "Apple Writer," an internationally successful word processing program for the early Apple computers.

For many problems, randomized algorithms are either the fastest algorithm or the simplest algorithm; sometimes they even provide the only known algorithm. Randomized algorithms have become so prevalent that deterministic algorithms could be viewed as a curious special case. In this talk I will describe some startling examples of randomized algorithms for solving some optimization problems on graphs.

The interactive proof system model of computation is a cornerstone of complexity theory, and its quantum computational variant has been studied in quantum complexity theory for the past decade. In this talk I will discuss an exact characterization of the power of quantum interactive proof systems that I recently proved in collaboration with Rahul Jain, Zhengfeng Ji, and Sarvagya Upadhyay. The characterization states that the collection of computational problems having quantum interactive proof systems consists precisely of those problems solvable with an ordinary classical computer using a polynomial amount of memory (or QIP = PSPACE in complexity-theoretic terminology). This characterization implies the striking fact that quantum computing does not provide any increase in computational power over classical computing in the context of interactive proof systems.

I will not assume that the audience for this talk has any familiarity with either quantum computing or complexity theory; and to be true to the spirit of the interactive proof system model, I hope to make this talk as interactive as possible -- I will be happy to explain anything related to the talk that I can that people are interested in learning about.

Cairo is an open source, cross platform, vector graphics library with the ability to output to many kinds of surfaces, including PDF, SVG and PNG surfaces, as well as X-Window, Win32 and Quartz 2D backends.

Unlike the raster graphics used with programmes and libraries such as The Gimp and ImageMagick, vector graphics are not defined by grids of pixels, but rather by a collection of drawing operations. These operations detail how to draw lines, fill shapes, and even set text to create the desired image. This has the advantages of being infinitely scalable, smaller in file size, and simpler to express within a computer programme.

This talk will be an introduction to the concepts and metaphors used by vector graphics libraries in general and Cairo in particular.

Concurrency is hard. Well maybe not hard, but it sure is annoying to get right. Even the simplest of synchronization tasks are hard to implement correctly when using synchronization primitives such as locks and semaphores.

Raymond Laflamme is the director of the Institute for Quantum Computing at the University of Waterloo and holds the Canada Research Chair in Quantum Information. He will give a brief introduction to quantum computing and why it matters, followed by a talk on programming quantum computers. This is followed by tours of IQC Labs.

This talk will describe a non-traditional functional approach to the classical problems of lexing (breaking a stream of characters into "words" or tokens) and parsing (identifying tree structure in a stream of tokens based on a grammar, e.g. for a programming language that needs to be compiled or interpreted). The functional approach can clarify and organize a number of algorithms that tend to be opaque in their conventional imperative presentation. No prior background in functional programming, lexing, or parsing is assumed.

The slides for this talk can be found here as a pdf.

The combination of computer graphics, geometry, and rapid prototyping technology has created a wide range of exciting opportunities for using the computer as a medium for creative expression. In this talk, I will describe the most popular technologies for computer-aided manufacturing, discuss applications of these devices in art and design, and survey the work of contemporary artists working in the area (with a focus on mathematical art). The talk will be primarily non-technical, but I will mention some of the mathematical and computational techniques that come into play.

The slides for this talk can be found here as a pdf.

TODO

TODO

Instant messaging (IM) is an increasingly popular mode of communication on the Internet. Although it is used for personal and private conversations, it is not at all a private medium. Not only are all of the messages unencrypted and unauthenticated, but they are all routedthrough a central server, forming a convenient interception point for an attacker. Users would benefit from being able to have truly private conversations over IM, combining the features of encryption, authentication, deniability, and forward secrecy, while working within their existing IM infrastructure.

In this talk, I will discuss "Off-the-Record Messaging" (OTR), a widely used software tool for secure and private instant messaging. I will outline the properties of Useful Security and Privacy Technologies that motivated OTR's design, compare it to other IM security mechanisms, and talk about its ongoing development directions.

Globally, issues about information privacy in the marketplace have emerged in tandem with the dramatic and escalating increase in information stored in electronic formats. Data mining, for example, can be extremely valuable for businesses, but in the absence of adequate safeguards, it can jeopradize informational privacy. Dr. Ann Cavoukian talks about how to use technology to enhance privacy. Some of the technologies discussed included instant messaging, RFID tags and Elliptical Curve Cryptography (ECC). Then Dr. Cavoukian explained the “7 Privacy – Embedded Laws” followed by a discussion on a biometrics solution to encryption.

A good programming language is far more than a simple collection of features. My ideal is to provide a set of facilities that smoothly work together to support design and programming styles of a generality beyond my imagination. Here, I briefly outline rules of thumb (guidelines, principles) that are being applied in the design of C++0x. Then, I present the state of the standards process (we are aiming for C++09) and give examples of a few of the proposals such as concepts, generalized initialization, being considered in the ISO C++ standards committee. Since there are far more proposals than could be presented in an hour, I'll take questions.

Dr. Bjarne Stroustrup is the original designer and implementer of the C++ Programming Language.

Copyright developed in the age of the printing press, and was designed to fit with the system of centralized copying imposed by the printing press. But the copyright system does not fit well with computer networks, and only draconian punishments can enforce it.

The global corporations that profit from copyright are lobbying for draconian punishments, and to increase their copyright powers, while suppressing public access to technology. But if we seriously hope to serve the only legitimate purpose of copyright -- to promote progress, for the benefit of the public -- then we must make changes in the other direction.

This talk by Richard M. Stallman is broken into two parts: the main talk and the question and answer sessions following the talk. Both are available in only Ogg/Theora format in keeping with Stallman's wishes. They are available under the Creative Commons NoDerivs 1.0 license.

What is the typical monitor resolution of a GIMP user? How many monitors do they have? What size images do they work on? How many layers are in their images? The answers to these questions are generally unknown: no means currently exist for open source applications to collect usage data. In this talk, Professor Michael Terry will present ingimp, a version of GIMP that has been instrumented to automatically collect usage data from real-world users. Prof. Terry will discuss ingimp's design, the type of data we collect, how we make the data available on the web, and initial results that begin to answer the motivating questions. ingimp can be found at http://www.ingimp.org.

The slides from the talk are available here: ingimp_uw_csc_talk_6_27_2007.pdf.

Ralph Stanton reflects on the founding of the University of Waterloo Math Faculty.

Richard Stallman will speak about the goals and philosophy of the Free Software Movement, and the status and history the GNU Operating System, which in combination with the kernel Linux is now used by tens of millions of users world-wide.

Richard Stallman launched the development of the GNU operating system in 1984. GNU is free software: everyone has the freedom to copy it and redistribute it, as well as to make changes either large or small. The GNU/Linux system, basically the GNU operating system with Linux added, is used on tens of millions of computers today.

"The reason I care especially, is that there is a philosophy associated with the GNU project, and this philosophy is actually the reason why there is a system -- and that is that free software is not just convenient and not just reliable.... More important than convenience and reliability is freedom -- the freedom to cooperate. What I'm concerned about is not individual people or companies so much as the kind of way of life that we have. That's why I think it's a distraction to think about fighting Microsoft."

Biography: Stallman has received the ACM Grace Hopper Award, a MacArthur Foundation fellowship, the Electronic Frontier Foundation's Pioneer award, and the Takeda Award for Social/Economic Betterment, as well as several honorary doctorates.

The Question and Answer session (starting shortly after the hour and half mark) posed a number of interesting questions including, "Do you support the Creative Commons license?" and "Can I use ATI and NVIDIA drivers because Mesa isn't nearly as complete?".

A talk for those interested in 3-dimensional graphics but unsure of where to start. Covers the basic math and theory behind projecting 3-dimensional polygons on screen, as well as simple cropping techniques to improve efficiency. Translation and rotation of polygons will also be discussed.

A discussion of software start-ups founded by UW students and what they did that helped them grow and what failed to help. In order to share the most insights and guard the confidences of the individuals involved, none of the companies will be identified.

For decades, mainstream parallel processing has been thought of as inevitable. Up until recent years, however, improvements in manufacturing processes and increases in clock speed have provided software with free Moore's Law-scale performance improvements on traditional single-core CPUs. As per-core CPU speed increases have slowed to a halt, processor vendors are embracing parallelism by multiplying the number of cores on CPUs, following what Graphics Processing Unit (GPU) vendors have been doing for years. The Multi-core revolution promises to provide unparallelled increases in performance, but it comes with a catch: traditional serial programming methods are not at all suited to programming these processors and methods such as multi-threading are cumbersome and rarely scale beyond a few cores. Learn how, with hundreds of cores in desktop computers on the horizon, a local software company is looking to revolutionize the way software is written to deliver on the promise multi-core holds.

The ReactOS operating system has been in development for over eight years and aims to provide users with a fully functional and Windows-compatible distribution under the GPL license. ReactOS comes with its own Windows 2003-based kernel and system utilities and applications, resulting in an environment identical to Windows, both visually and internally.

More than just an alternative to Windows, ReactOS is a powerful platform for academia, allowing students to learn a variety of skills useful to software testing, development and management, as well as providing a rich and clean implementation of Windows NT, with a kernel compatible to published internals book on the subject.

This talk will introduce the ReactOS project, as well as the various software engineering challenges behind it. The building platform and development philosophies and utilities will be shown, and attendees will grasp the vast amount of effort and organization that needs to go into building an operating system or any other similarly large project. The speaker will gladly answer questions related to his background, experience and interests and information on joining the project, as well as any other related information.

Slides from the talk are available here.

Biography

Alex Ionescu is currently studying in Software Engineering at Concordia University in Montreal, Quebec and is a Microsoft Technical Student Ambassador. He is the lead kernel developer of the ReactOS Project and project leader of TinyKRNL. He regularly speaks at Linux and Open Source conferences around the world and will be a lecturer at the 8th International Free Software Forum in Brazil this April, as well as providing hands-on workshops and lectures on Windows NT internals and security to various companies.

Bill Gates discusses the software and computer industry, and how Microsoft has contributed. Gates also discusses his views on the future of the computing industry. The talk was recorded in 1989 but was only recently digitized.

Topics include:

• The start and history of the microcomputer industry
• Microsoft BASIC and the Altair 880 computer
• The transition from 8-bit to 16-bit computers
• Microsoft's history with IBM
• 640k memory barrier and 16-bit architectures
• 32-bit 386 and 486 architectures
• RISC and multi-processor machines
• EGA graphics and WYSIWYG editors
• Decreasing cost of memory, harddisks and hardware in general
• The importance and future of the mouse
• Object-oriented programming
• MS-DOS and OS/2
• Multi-threaded and multi-application systems
• Synchronization in multi-threaded applications
• Diskette-based software
• UNIX standardization and POSIX
• History of the Macintosh and Microsoft' involvement
• Involvement of Xerox in graphical user interfaces
• Apple vs. Microsoft lawsuit regarding user interfaces
• OS/2 future as a replacement for MS-DOS
• Microsoft Office on Macintosh
• Thin/dumb clients
• Compact discs
• Multimedia applications
• Gates' current role at Microsoft

The following picture was taken after the talk (click for higher-res).

Do spam filters work? Which is the best one? How might filters be improved? Without standards, one must depend on unreliable evidence, such as subjective impressions, testimonials, incomparable and unrepeatable measurements, and vendor claims for the answers to these questions.

You might think that your spam filter works well and couldn't be improved. Are you sure? You may think that the risk of losing important mail outweighs the benefit of using a filter. Could you convince someone who holds the other opinion? If I told you that my filter was 99-percent accurate, would you believe me? Would you know what I meant? Would you be able to translate that 99-percent into the risk of losing an important message?

Gord Cormack talks about the science, logistics, and politics of Spam Filter Evaluation.

Simon Law leads the Quality teams for Ubuntu, a free-software operating system built on Debian GNU/Linux. As such, he leads one of the largest community-based testing efforts for a software product. This does get a bit busy sometimes.

In this talk, we'll be exploring how the Internet is changing how software is developed. Concepts like open source and technologies like message forums are blurring the lines between producer and consumer. And this melting pot of people is causing people to take note, and changing the way they sling code.

The Computer Science Club would like to thank the CS-Commons Committee for co-sponsoring this talk.

Could you write a good image recognizer for a 100 MHz mobile phone processor with 1 MB heap, 320x240 image, on a poorly-optimized Java stack? It needs to locate and read two-dimensional barcodes made up of square modules which might be no more than a few pixels in size. We had to do that in order to establish Semacode, a local start up company that makes a software barcode reader for cell phones. The applications vary from ubiquitous computing to advertising. Simon Woodside (founder) will discuss what it's like to start a business and how the imaging code works.

Eric LaForest delivers a crash-course on modern stack computing, the Forth programming language, and some projects of his own. Stack systems have faster procedure calls and reduced complexity (shorter pipeline, simpler compilation) relative to their conventional counterparts, as well as more consistent performance, which is very important for real-time systems. Many consider stack-based architecture's crowning feature, however, to be the unrivalled price-to-performance ratio.

Note: the slides are hard to make out in the video, so make sure to download the slides as well.

A discussion of how software creators can identify application opportunities that offer the promise of great social and commercial significance. Particular attention will be paid to the challenge of acquiring cross domain knowledge and setting up effective collaboration.

Rico Mariani, (BMath CS/EEE 1988) now an (almost) 18 year Microsoft veteran but then a CSC president comes to talk to us about the evolution of software tools for microcomputers. This talk promises to be a little bit about history and perspective (at least from the Microsoft side of things) as well as the evolution of software engineers, different types of programmers and their needs, and what it's like to try to make the software industry more effective at what it does, and sometimes succeed! Particularly illuminating are his responses to advocates of free/open-source software.