Northeastern ACM Chapter

The Speaker Series Events are held every Monday from 12-1:30 (Activities Period) in 366 WVH and are open to the public (unless stated otherwise). If you have any questions and/or suggestions, email acm@ccs.neu.edu

Bill Burke
Date: Fall, 10.06.04
Topic: Open Source
Abstract: This presentation walks the audience through creating an open source project and the choices one needs to make to foster its success. It also walks through how you can actually build a business with open source products.
Biography: Bill Burke is Chief Architect of JBoss Inc. Bill started contributing to the open source JBoss project back in 2001 and it soon became a full time job. Now, as lead architect of JBoss 4, his main focus is bringing AOP concepts and technology to the JBoss application server. Bill is co-author of O'Reilly's "JBoss 3.2 Workbook", and has numerous other in-print and on-line publications. He graduated magna cum laude with a B.S.C.S. from Northeastern University in Boston in 1994.

Professor Judith A. Perrolle
Date: Fall, 09.29.04
Topic: Why the CAN-SPAM Act failed
Abstract: In January 2004, when the CAN-Spam Act went into effect, Unsolicited Commercial Email, or spam, accounted for an estimated more than half of all email communications. By the time of the Federal Trade Commission Hearings on the Act in March, 2004, spam had increaseds to more than three quarters of all email. The CAN-Spam Act can't cure spam for a variety of reasons:

1. It is based on technologically unworkable premises: that the internet can be regulated in the same way as the U.S. telephone system; that a "Do Not Spam" list could be kept safe from spammers; and that the volume of spam would be reduced if "illegal" spamming activity were ended.
2. It does not address the fundamental conflicts among citizens, advertizers, and intellectual property holders: it supports a burdensome "opt out" model for consumers; it violates existing social norms; it has no effective way to keep advertizers from using deceptive practices to acquire email addresses; it fails to address non-email forms of spam (instant messaging, adware/spyware, browser hijackings, popups, and annoying images); and it ignores the economic costs and benefits of spamming.
3. It is inconsistent with international law and lacks jurisdiction over the locations where a growing amount of spam originates.

By being ineffective, the CAN-Spam Act encourages corporate and citizen vigalanteeism, raises threats of censorship and restrictions of free speech, encourages a disrespect for the law by corporations and individuals, and contributes to U.S. isolation from the world community. It represents an attempt to legislate a solution to a twenty-first century problem using a twentieth-century model of communications.
Biography: Judith Perrolle Associate Professor of Sociology Sociology of Computing, Environment Sociology, Sociology of Science, Knowledge, and Technology PhD, Brown University

Judith Perrolle is engaged in participatory action research on social issues in computer and communications system design supported by Northeastern's Instructional Development Fund. Her previous research has been supported by the National Science Foundation, the Harvard School of Public Health, and the General Electric Company.

Anahuac de Paula Gil
Date: Summer, 08.13.04
Topic: Free and Open Source in Brazil
Abstract: Brazil has determined a culture for Free and Open Source Software (FOSS) that they call "Software Livre!" In this culture, the government, industry and the community band together to do projects using FOSS.
Biography: Anahuac de Paula Gil is a Brazilian citizen and the head of the Linux Easy Server Project.

Barrett Hazeltine
Date: Spring, 03.31.04
Topic: Being an Entrepreneur; In the US and in the Third World
Abstract: Being an entrepreneur requires an idea. It also requires building a team, securing production, establishing distribution, and raising funds. Stories about ventures, both here and abroad, illustrate the joys, tribulations, and models of being an entrepreneur.
Biography: Barrett Hazeltine is Professor Emeritus of Engineering at Brown University. He has received awards for teaching from thirteen senior classes at Brown, and in 1985 the award was named after him. He has taught at the Universities of Malawi, Zambia, and Botswana, and at Africa University in Zimbabwe. Other foreign countries in which he has done teaching or consulting are Bangladesh, Indonesia, The Philippines, Taiwan, and Thailand. In the past he has taught several different electrical engineering courses, and at Brown he currently teaches introductory management, managerial decision-making, and appropriate technology.

James Dennis and Glen Coppersmith
Date: Spring, 03.17.04
Topic: Game playing AI and an application to Tetris
Abstract: Artificial Intelligence is well suited for some things and poorly suited for others. Game playing is one arena in which AI's excel. There are different strategies an AI developer might use depending on the game they're playing and the system resources available to them. Tetris is one place in which AI certainly outperforms humans with little system resource requirements and the possibility for an endless game to occur! James and Glen will explain how this is possible and what their results have been in this endeavor.

Jon Hart
Date: Spring, 02.04.04
Topic: InfoSec Blunders
Abstract: This talk will lump computer security, network security, personal security and other loosely defined security terms into the word "InfoSec". InfoSec has been a concern since the dawn of computing, but only recently as money has poured into the industry and lives (and livelihoods) depended on it has InfoSec got truly interesting. Why? Because the good guys seem to keep finding ways to make mistakes, and the bad guys keep discovering ways to keep the good guys on their toes. This talk will present common InfoSec blunders along with ways to protect yourself. And, time and resource allowing, this talk will cover popular ways of abusing said blunders.

David Blank-Edelman
Date: Spring, 01.14.04
Topic: Through the Lens Geekly, or How Sysadmins Are Portrayed in Pop Culture
Abstract: People outside our profession think they know who we are and what we do for a living.They've formed assumptions about us and how we work even before they've met us, which in turn color their perceptions and shape their interactions with us. To be effective we need to understand this context and its origins. A good portion of it comes from popular culture: movies, television, and other mass media. Movie clips and other source material will entertain you, and more important, give you new insight into just what ways much of the world views our profession.
Biography: David is the Director of Technology at the NU College of Computer Science. He has spent the last 14+ years of his life as a system/network administrator in large multi-platform environments; and is the author of O'Reilly's OtterBook, Perl for System Administration.

Jon 'maddog' Hall
Date: Fall, 12.10.03
Topic: Open Source and the Grid: A tour through the technologies
Abstract: Linux and Open Source has long (well, almost eight years now) been used for creating superclusters known as Beowulf systems. First it was complex and crude, but now with the addition of other software projects such as Oscar and Scyld, creating a Beowulf system is becoming easier and easier.

This community energy is now escaping to the concept of "The Grid", a way of sharing vast computer resources over the world. Clusters of computers can be configured to be "compute providers" or "compute users", to help utilize the vast amounts of computing power that we have. This talk will show just the surface of some of these projects and the potential they are showing for making computing a true commodity, with true Open Source standards. It will put forth arguments for the corporate, divisional, departmental and personal grid.
Biography: Jon "maddog" Hall is the Executive Director of Linux International (www.li.org), a non-profit association of computer vendors who wish to support and promote the Linux Operating System. During his career which spans over thirty years, Mr. Hall has been a programmer, systems designer, systems administrator, product manager, technical marketing manager and educator. He has worked for such companies as Western Electric Corporation, Aetna Life and Casualty, Bell Laboratories, Digital Equipment Corporation, VA Linux Systems, and is currently funded by SGI.

He has taught at Hartford State Technical College, Merrimack College and Daniel Webster College. He still likes talking to students over pizza and beer (the pizza can be optional).

Mr. Hall is the author of numerous magazine and newspaper articles, many presentations and one book, "Linux for Dummies".

Mr. Hall serves on the boards of several companies, and several non-profit organizations, including the USENIX Association.

Mr. Hall has traveled the world speaking on the benefits of Open Source Software, and received his BS in Commerce and Engineering from Drexel University, and his MSCS from RPI in Troy, New York.

In his spare time maddog is working on his retirement project: maddog's monastery for microcomputing and microbrewing

Prof. Peter Tarasewich
Date: Fall, 11.19.03
Topic: Evaluation of Visual Notification Cues for Ubiquitous Computing
Abstract: With increased use of mobile information technology and increased amounts of information comes the need to simplify information presentation. This research considers whether low-information-rate displays (such as those used in mobile devices) can provide effective information awareness. An experiment was performed to measure the performance/size tradeoff of visual displays ranging in size from two LEDs to nine LEDs, and using a number of display characteristics - i.e., color and blinking in various combinations. Results show a reliable tradeoff between performance (participant response time and accuracy) and display size (number of LEDs). However, even the full set of 27 messages can be conveyed with high recognition accuracy using only three LEDs by mapping the messages into color and position. Thus, mobile devices with micro-level form factors can be designed to convey critical information and provide effective notifications. Future work and a prototype developed from this work are discussed.

Ian Holland
Date: Fall, 11.14.03
Topic: Computer Science in Perspective
Abstract: Thoughts from a PhD alumnus examining the motives and opportunities for computer scientists and information systems professionals in 2003.
Biography: Ian Holland is the Vice President of Systems Engineering and Architecture at Kronos.

Dean Larry Finkelstein
Date: Fall, 11.05.03
Topic: Future Directions for the College of Computer and Information Science
Abstract: In this presentation, I will outline the long-term goals for the College of Computer and Information Science. An important focus will be on the College's desire to be a top 50 nationally ranked computer science program. I will also describe future plans for undergraduate education and new programmatic efforts at the graduate level. Finally, I will present an inside view of the new building, which shows how the space has been utilized to support the College's programmatic efforts.

Ian Melven
Date: Fall, 10.22.03
Topic: Exploits 101
Abstract: Programming errors that lead to stack overflows, heap overflows, integer overflows, string bugs, and Windows shatter attacks will be shown. Methods of exploitation will then be explained. Current techniques for payloads will be discussed. Real world working exploits for Windows 2000 will be demo'd and dissected. If there is time, ways of finding unreleased exploits will be discussed. Hopefully, there will be time for questions.

The talk will be slightly biased towards Windows but most of the techniques discussed are also applicable to Unix systems.

John Chamberlain
Date: Spring, 04.28.03
Topic: The Importance of Robotics Development to Our Future
Abstract: It is no secret that future productivity depends on automation. What is less recognized is the importance of physical automation--robotics--to our future. As programmers we focus on informational automation and tend to see robotics as a sideline, one of many specialities, and one that is even a little distant because it appears hardware dependent. This may be an error in perception. Robotics is increasingly at the focal point of human development and software development is the single most important component of robotics. This talk will detail the pivotal role robotics has in our future and why it is crucial for computer scientists to take the lead in physical automation as well as informational automation.
Biography: John Chamberlain is a 1993 graduate of Northeastern's masters program in Computer Science. He has worked for many of New England's largest firms including Gillette, Cabot Corporation and Fidelity. Mr. Chamberlain has spoken at JavaOne and written articles for leading trade journals such as Java Developer's Journal, JavaWorld, Visual Systems Journal and Visual Basic Programmer's Journal.
Slides: Available as a PowerPoint file.

Robert Familiar
Date: Spring, 04.14.03
Topic: Microsoft .NET Architecture Patterns and Best Practices
Abstract: Designing distributed applications is no simple task. Many decisions need to be taken at the architecture, design, and implementation levels. These decisions will have an impact on the "abilities" of the applicationsecurity, scalability, availability, and maintainability, to name a fewand will have an impact on the architecture, design and implementation of the target infrastructure. This session will delve into the choices that arise when designing the layers of a distributed application, and will present these choices as a set of layers of components that you can use to model your application. An overview of .NET, Tightly Coupled and Loosely coupled architectures, XML Web Services and .NET and J2EE interoperability will be covered.

John Manferdelli
Date: Spring, 04.07.03
Topic: A Model and Method for Erecting and Maintaining Security Perimeters around Distributed Trusted Computing Bases on Open Computing Platforms
Biography: Available at Microsoft here

Ryan Perry
Date: Winter, 02.31.03
Topic: WebObjects: An Intuitive and Powerful Web Application Development Environment (and Apple's Best-Kept Secret)
Abstract: With WebObjects, Apple has provided a multi-platform, J2EE-compatible set of tools for the creation of dynamic web content and client/server-based applications. Originally created by Next in the mid-90's, WebObjects is already a proven, robust system that competes favorably on performance and features with more widely-known web application development environments like IBM's WebSphere and Microsoft's .NET web application development tools.

Even though WebObjects is a great technology that stands up well to comparisons with other web application development platforms, this presentation will not feature comparisons with other technologies. Instead it will offer a developer's point of view on the technology, including an introduction to the system, an explanation of the various features and development tools, and a concrete example if time permits.
Biography: Ryan Perry is a recent graduate of Northeastern University's Undergraduate College of Computer Science and currently does web application development as a Senior Software Developer for NBH Solutions LLC, a small IT consulting firm specializing in IT management and custom web application development.

Prof. Robert Futrelle
Date: Winter, 02.24.03
Topic: The Challenges of Building Practical Knowledge-Based Systems
Abstract: The biomedical literature increases at a rate of 600,000 papers per year, about a terabyte of content. This compares with the human (your!) genome of about a gigabyte. It is possible to do a reasonable search of this important literature using PubMed. But no good ideas or systems exist to answer conceptual queries or to find particular diagrams of interest from the literature.

This talk discusses three aspects of these problems as worked on in my research group, the Biological Knowledge Laboratory (BKL).

• How do we analyze the meaning of text?
• How do we analyze the meaning of diagrams?
• How can we use this understanding for query answering and knowledge mining?

Miguel de Icaza
Date: Winter, 02.17.03
Topic: The Mono Project, Ximian Inc.
Abstract: The Impact of the project is found in the possibilities for software development as well as the controversy of an open source project implementing Microsoft s technologies. Although the CLI and the C# language are ECMA standards ADO.NET and ASP.NET are Microsoft technologies, and Microsoft s CEO Steve Ballmer expressed his doubt that mono will succeed without copyright infringements to a German industry magazine. On the other hand it s advantageous to the global software community if a superior infrastructure like .NET is available on any platform Microsoft itself released a version of the CLI for FreeBSD (aka Rotor). As a founder of the GNOME desktop project Miguel de Icaza will most likely lobby to deploy the mono infrastructure within the GNOME system, this in turn will increase the economies of scale for any .NET application to run on Windows and Linux desktops.

The Mono Project is an innovative open source approach to implement the same infrastructure found in Microsoft s .NET for the next generation computer software architecture on operating systems other then windows. Mono itself is focused on implementing a virtual machine, justin- time compiler and a class loader; all three are forming the CLI. The class library will be designed to run all languages supported by the CLR. Finally mono will provide a compiler for the C# programming language. Additional efforts close to the project are aiming to provide other parts of the .NET framework as ASP.Net and ADO.NET as well as Windows. Forms and GTK# as GUI API's.
Biography: Miguel de Icaza is the CEO of Ximian, the company hosting and founding the mono project. He s also a founder of the GNOME Linux Desktop, for which Ximian develops solutions as well as provides services and support. As the founder of mono he has been criticized by some open source representatives for allowing Microsoft to enlarge it s reach to Linux. Mr. Icaza clearly focuses successfully on combining business goals with traditional open source ideas.

Prof. Agnes Hui Chan
Date: Winter, 02.10.03
Topic: Authentication and Key Exchange Protocols for Wireless Imbalanced Networks
Abstract: Low-power wireless devices, such as Personal Device Assistants (PDAs) and cellular phones, are characterized by their limited memory capacity, low computational power, small and monochrome screens. In addition, the wireless environment is limited in bandwidth and is subject to erratic changes such as weather, terrain and external interference. These constraints have prevented a simple migration of cryptographic protocols that are widely adopted in wire-line networks to wireless networks for authentication and security. Due to the mobility of wireless devices, authentication of both the user (client) and the base station (server) becomes paramount importance. In this talk, we present our efficient mutual authentication and key exchange protocols between a low-power wireless device and a powerful base station. The aim is to reduce the computational burden on the client while maintaining similar level of security and scalability as expected by users. We will discuss both device-oriented authentication as well as password based user-oriented authentication.

Prof. Matthias Felleisen
Date: Winter, 02.03.03
Topic/Abstract:

The
 Why 
  When 
   Where
    Whom 
     What 
      and How of a PhD

Michael Oh
Date: Winter, 01.27.03
Topic: NewburyOpen.net - Boston's Free Wireless Initiative
Abstract: Michael Oh, President and Founder of Tech Superpowers, Inc. - creators of Boston's Free Wireless Initiative, NewburyOpen.net - will talk about the benefits and perils of Public Wireless. NewburyOpen.net has been covered by the Boston Globe, CNET, Slashdot, ABC World News Tonight, London's Financial TImes, and Investor's Business Daily because of its innovative way of bringing WiFi to the masses.

802.11b is by far the most popular wireless standard since the cell phone, but its limitations, security risks, and widespread deployment have been the discussion of techies and non-techies alike. NewburyOpen.net is Boston's largest public wireless installation, so it's been through all the trials and tribulations of a public WiFi, and Michael will talk about his experiences as one of the city's and nation's wireless evangelists.

Prof. Karl Lieberherr
Date: Winter, 01.06.03
Topic: Why Should You Care About Aspect-Oriented Programming?
Abstract: A common problem in software is that programs are a tangled artifact of several design concerns. This makes them hard to understand and maintain.

In this talk we will present an approach to programming in the AspectJ programming language that allows many (not all) concerns to be expressed loosely coupled rather than tangled together with other concerns. This leads to programs that are less tangled and that are easier to understand and maintain. AspectJ is currently the most popular aspect-oriented programming (AOP) language.

The Inventor's Paradox is about problem solutions becoming easier when we solve more general problems and it plays an interesting role in mathematics and in AOP.

We apply the Inventor's Paradox to programming and show that programming becomes easier if we write the programs for generalized object/program structures rather than specific object/program structures.

The presentation will be sprinkled with AspectJ examples that illustrate the concepts.

For a related conference at Northeastern, see http://aosd.net/conference.html

Mark Logan
Date: 12.02.02
Topic: Anomaly: A special purpose language for system and network management.
Abstract: Heterogenous networks are inherently difficult to manage. This talk presents a special purpose language, called Anomaly, that will ease the burden on administrators by allowing them to explicitly model contextual relationships (such as physical locations). After learning about contexts and environmental acquisition, which are the principles that Anomaly is built on, you'll hear about the implementation of Anomaly as well as several interesting examples.
Slides: Available as PowerPoint file

Rebecca Neeson
Date: Fall, 11.25.02
Topic: The TiVo-ization of America: Trusted Systems, the Digital Millenium Copyright Act, and Your Own Personal Computing Appliance
Abstract: Trusted systems technologies and heavy-handed laws are the content industry's tools for the gargantuan task of putting the cat back into the bag after the Napster/P2P debacle. We'll address the technological advances related to so-called trusted systems and the legal mechanisms set in place to back them up by the U.S. Digital Millennium Copyright Act. We'll ask ourselves, "What's a trusted system?" from the perspective of the average PC user...and why is it called "trusted"? We'll explore the drawbacks of these technologies, and the possibility that trusted systems that aren't perfect may still be good enough for their creators' purposes. I'll argue that the current trend points to a future where your (or at least your mom's PC) will more closely resemble a TiVo than a ThinkPad.

Matte Elsbernd
Date: Fall, 11.18.02
Topic: Life as a Web Designer
Biography: Matte has been designing web pages since 1994 and has bounced form company to company working on web design. He is best known for his extra-curricular projects.
Abstract: If anything, my niche on the web is the realm of "do it yourself". From my own experience, there's nothing that one individual can't do that equals or exceeds what dot.com businesses have been doing. The Firm List has exceeded and outlasted sites run by teams of people with millions of dollars of VC financing.

Ultimately, the power of the web is that all of the tools are available to anyone. It doesn't take a lot of money or even a lot of people to get your ideas & projects online. In addition, there's no quality difference between a "spare-time" project and a "business". It's all a matter of how hard one works.

There are a lot of resources, tools, and programs out there that aid the independent web publisher. There are shareware web packages. There are different types of revenue generators, etc. I've had experiences with many over the years. I've figured out the advantages & disadvantages of them all.

I am known in the industry for my spare-time projects more than my corporate work. I think that spare-time projects are the perfect place for those interested in the web to learn & develop their skills as well as show off their abilities to the industry. I've gotten all of my jobs from these projects, and they are an essential portfolio piece.

David LaPorte
Date: Fall, 11.4.02
Topic: Network Security
Abstract: David LaPorte, a Senior Network Engineer at Harvard University and maintainer of a large network intrusion detection system (NIDS), will discuss real-world use of the technology. Topics covered will include an overview of NIDS deployment, a discussion of the tools and equipment necessary to implement a NIDS yourself, and - of course - a few war stories.

David LaPorte is a class of 2000 CCS alumnus and past ACM president.
Slides: Available as PowerPoint file

Prof. Kenneth Baclawski
Date: Fall, 10.28.02
Topic: Ontology-Based Computing
Abstract: An ontology is a theory about what entities can exist and how entities can be related with each other in a domain. Ontologies are emerging as the most effective means for enabling flexible communication between autonomous computer systems. The notion of ontology is at the center of the "Semantic Web", proposed by Tim Berners-Lee and featured in a recent article in Scientific American. This talk will discuss the research and development efforts that are attempting to make ontology-based computing a reality:

• The DARPA Agent Markup Language (DAML), a standard language for expressing ontologies and ontology-based knowledge representations.
• Tools for ontology development, consistency checking and mediation between ontologies.
• Scalable, high-performance storage and indexing of ontology-based knowledge representations.
• Integration with other computer-based technologies such as programming languages, databases, CASE tools, Web servers, and so on.

Nat Friedman
Date: Spring, 05.20.02
Topic: GNOME Project, Ximian
Biography: Until recently, Nat served as both CEO and President of Ximian, and he is excited to now return to programming and product development full time. A long-time open source developer, Nat was one of the architects of the GNOME Foundation's proposal and charter. He has also made significant contributions to the development of the Bonobo object model. Both a hacker and an entrepreneur since childhood, Nat's dynamic personality has made him a natural leader and an important spokesperson for the free software movement. Nat earned bachelor's degrees in Mathematics and in Computer Science at MIT. (taken from the Ximian Executive Team page)

Joel Solon
Date: Spring, 05.06.02
Topic: InterSystems Caché and the Caché Campus Program
Abstract: Caché is a new generation of an established database technology that addresses the needs of today's transaction processing applications (large networks, thousands of users, high performance, fast development) by combining a multidimensional data server with a versatile application server. Featuring rapid Web development, and a unique fusion of advanced object technology and SQL, Caché provides high levels of performance and scalability. Caché Campus is a program developed specifically to aid academics, researchers, and students in learning about the latest, cutting-edge technology being used in business today. The Caché Campus program gives you the opportunity and all of the materials you'll need to use it in your own work, for free. Our offer extends to not only traditional computer science departments, but also to information management-intensive schools such as those specializing in business, healthcare, and industrial management.

Prof. Harriet Fell
Date: Spring, 04.29.02
Topic: Babble Recognition and Related Applications
Abstract: The Early Vocalization Analyzer (EVA) is a computer program that automatically analyses digitized acoustic recordings of infant vocalizations. Using the landmark detection theory of Stevens et al for the recognition of phonetic features in speech, EVA detects syllables in vocalizations produced by infants. Landmarks are grouped into standard syllable patterns and syllables are grouped into utterances. Statistics derived from these groups and the underlying features are used to derive a "vocalization age" that can clinically distinguish infants who may be at risk for later communication or other developmental problems from typically developing infants in the six to fifteen month age range.

The EVA software is also being used to locate stuttered segments in speech for automatic smoothing and to classify types of dysarthria. Two new projects using the EVA software are just getting underway. VisiBabble is a system to provide real-time visual reinforcement of vocalizations produced by infants or preschool children who are delayed in their production of pre-speech vocalizations. We will also be using the underlying technology of EVA to automatically classify emotion in speech.
Pictures of Lecture: here

Jon 'maddog' Hall
Date: Spring, 4.22.02
Place: 50 Dodge Hall (where?)
Topic: Free Software, but no Free Beer
Abstract: Everyone has heard about the operating system called Linux (who some people call GNU/Linux). Some even know that is runs on things as small as embedded systems and as large as supercomputers. Few, however, know WHY it is such a good operating system, and why it has the range it does. In this talk maddog (as his students called him, and as he likes to be called) will discuss both the technical aspects of the operating system and the business aspects that follows from it. Neither technical person nor business person should be afraid to attend, as he will endeavor to be gentle with both parties without boring either.
Biography: Jon "maddog" Hall has spent thirty years in the computer field as a programmer, system designer, product manager and technical marketing person. As an educator at the college level, he was proud of his ability to make "complex things simple" rather than the other way around. He is known to be able to explain business issues to engineers, and engineering issues to business people without either group falling asleep.

maddog has his BS in Commerce and Engineering from Drexel University, and his MSCS from RPI.
Pictures of Lecture: here

Paul Graham
Date: Spring, 04.15.02
Place: 50 Dodge Hall (where?)
Topic: How Startups Work
Abstract: The word "startup" dates from the 1960s, but the idea is much older. Since the Industrial Revolution, technology startups have been the most straightforward way to get rich. If you think you might one day want to be involved in one, there is a lot you need to know. This talk will attempt to cover the most important 45 minutes of it, including the pros and cons of doing it at all, how to structure the company, who to hire, what to make, how to get funding, what to spend it on (and not), how to deal with competitors, and how to get bought.
Biography: Paul Graham is the designer of the Arc language. He most recently worked for Yahoo! Inc. Previously he was president of Viaweb, which became Yahoo! Store when Viaweb was acquired by Yahoo! in the summer of 1998.

Paul is the author of On Lisp (Prentice Hall, 1993) and ANSI Common Lisp (Prentice Hall, 1995), now the standard college text. He has worked as a consultant to the US Department of Energy, DuPont, and Interleaf. He has an AB from Cornell and a PhD in Computer Science from Harvard, and studied painting at RISD and the Accademia di Belle Arti in Florence.

The above biography was taken from his website at http://www.paulgraham.com.
Pictures of Lecture: here

Prof. Will Clinger
Date: Winter, 02.25.02
Topic: Mathematical Models of Object Lifetimes
Abstract: After forty years, it is no longer necessary to argue that garbage collection---automatic reclamation of unreachable heap storage---is a Good Thing. What we need now are better garbage collectors: faster, more space-efficient, and less disruptive.

Alternative algorithms for garbage collection are often compared by running or simulating them on a set of benchmarks. The usual result is that the best-performing algorithm depends on the benchmark. To make sense of these empirical results, we need a quantitative theory of garbage collection. To obtain this theory, we must develop models for object lifetimes that are both realistic and tractable. Insights obtained from these models have already led to a new family of algorithms for garbage collection.

Mark Logan
Date: Winter, 02.11.2002
Topic: Web Development with Zope
Abstract: Zope is an application server written in Python, and offers an alternative to the more common script/database web development paradigms in use today. The talk will include an introduction to Zope and a brief tutorial on web application development with Zope, in the form of several "Cookbook" style solutions to common web development problems.
Slides: Located on his Zope website (html).

Scott Miller
Date: Winter, 02.04.2002
Topic: Statistical Models for Natural Language Processing
Abstract: The talk will consider several problems in human language processing including speech recognition, optical character recognition, information retrieval, and language understanding.
Slides: Available as PowerPoint File

Homer Pien
Date: Winter, 01.28.2002
Topic: Bioinformatics
Abstract: Bioinformatics deals with the computational aspects of molecular biology. Bioinformatics plays a critical role in the understanding of virtually all complex biological processes, and is rapidly becoming a cornerstone of the drug discovery process. Interestingly, the complexities of bioinformatics problems can be reduced to simple computer science concepts, involving little more than programming, algorithms, data structures, and databases. In this talk I will provide a very cursory overview of molecular biology, followed by a quick survey of a few applications of bioinformatics. Lastly, bioinformatics in the context of biotechnology and pharmaceutical drug development will be discussed.
Biography: Homer Pien received his undergraduate in math from the University of Illinois, and his MS and PhD in computer science from NU CCS. Homer is currently the CEO of SRU Biosystems in Woburn, MA (http://www.srubiosystems.com/).
Slides: Available as PowerPoint file

Jennifer McDonald
Date: Winter, 01.14.2002
Topic: Interactive Pushdown Automata Animation
Abstract: This lecture will present the Interactive Pushdown Automata Animation for use in an Automata Theory class. It will present the features of the IPAA as well as the algorithm and data model used. Finally, this article will outline the necessary pieces of a good visual tool and show how they are implemented in the IPAA.

Eric S. Raymond
Date: Fall, 11.19.01
Topic: Freedom, Power, and Software: What the Internet Teaches Us About Ethics and Politics
Biography: Eric S. Raymond is an observer-participant anthropologist in the Internet hacker culture. His research has helped explain the decentralized open-source model of software development that has proven so effective in the evolution of the Internet. His own software projects include one of the Internet's most widely-used email transport programs. Mr. Raymond is also a science fiction fan, a musician, an activist for the First and Second Amendments, and a martial artist with a Black Belt in Tae Kwon Do. His home page is here.

James M. Atkinson
Date: Spring, 05.21.2001
Topic: Technical Surveillance Counter Measures and the Modern Eavesdropping Threat
Resources: Biography, Company Information.

David N. Blank-Edelman
Date: Spring, 05.9.2001 and 05.16.2001
Topic: Perl for System Administration
Abstract: Two-part workshop.

Prof. Rajmohan Rajaraman
Date: Spring, 05.7.2001
Topic: Tracking Globally Distributed Data
Slides: Available as postscript file
Abstract: (N/A)

Bruce Maggs
Date: Spring, 04.30.2001
Topic: Global Internet Content Delivery
Slides: Available as a PowerPoint file.
Abstract: (N/A)

Richard Stallman
Date: Spring, 04.23.2001
Topic: Copyright vs Community in the Age of Computer Networks
Talk: Available on two pages at the GNU website: Reevaluating Copyright: The Public Must Prevail, Freedom-Or Copyright?

Justin Richer
Date: Spring, 04.9.2001
Topic: Collaborative Technology
Slides: Available at this website
Abstract: (N/A)

Kevin Fu
Date: Winter, 02.26.2001
Topic: Computer Insecurity
Talk: Cookie Eaters Project

The talk will cover UNIX insecurity, Web insecurity, and common security flaws. Live demonstrations of exploits will accompany the talk. By the end of the session, the audience will have a greater appreciation for the security implications of subtle assumptions made in network protocols and operating systems.

Kevin is a doctoral student at the MIT Lab for Computer Science, coordinator of the LCS Applied Security Reading Group, and member of the MIT Network Security team. His research interests include secure file systems and user authentication on the Web.

Bronislav Kavsan
Date: Winter, 02.19.2001
Topic: Integration of VPN and Public Key Infrastructures
Abstract: VPN and PKI are two very powerful and promising technologies for securing Internet. Even though PKI is enabling technology for securing various applications (Web, Digital Signatures, E-mail, etc), VPN presents some unique challenges for effective utilization of PKI. This presentation addresses tremendous challenges in integrating these very sophisticated and complex technologies to achieve seamless interoperability, usability and strong security as well as offers specific architectural approach for solving some of these challenges.

Bronislav Kavsan is a Vice President at RSA Security, where he is leading an engineering organization developing Advanced PKI family of products.

Mr. Kavsan joined RSA Security in 1999. His areas of interest include Data Communication Protocols, Network Security and Public Key Infrastructure. He has more than 25 year of information systems data communication experience, including architecture and implementation of VPN/IPSec systems (SafeNet, Inc) and research/development in the area of Data Communication Protocols (Bell Laboratories, AT&T/Lucent Technologies.

Richard M. Smith
Richard's Hompage
Date: Winter, 02.12.2001
Topic: Security and Privacy Problems in Wireless Web Devices and Cell phones
Abstract: Cell phones are increasingly taking on computer-like capabilities. Most digital cell phones in use today already have unique email addresses assigned to them and can receive short text messages. The next generation of WAP phones include small Web browsers for surfing the Web. All of these new technologies being introduced in phones raise new privacy and security issues which wireless phone companies have not had to deal with before. In Richard's talk, he will discuss potential problem areas in cell phones such as cell phone spam, buffer overflows, and malicious scripting code. He will compare and contrast these security and privacy threats to similar problems with desktop computer systems.
Richard is the Chief Technology Officer of the Privacy Foundation. The Foundation is a not-for- profit organization based in Denver, Colorado dedicated to research and public education of privacy issues.

John Zukowski
Date: Winter, 02.05.2001
Topic: Visions of the Future of the Java Language
Abstract: In this lecture we'll examine the evolution of Java the platform, the effect of lawsuits on that evolution, and what the future holds for Java. We'll focus on areas of interest to Java developers (vs. managers / business-type people) and see where those developers should focus in the future as Java continues to evolve.

Robert Silverman
Date: Winter, 01.29.2001
Topic: Cryptographic Key Sizes and Security
Abstract: Robert Silverman is Senior Research Scientist at RSA Laboratories. His interests include computational number theory, mathematical cryptography, design and analysis of algorithms, complexity theory, and large scale parallel computing. He is a leading expert on algorithms for breaking public key cryptosystems.

Vassilis Tsaoussidis
Vassilis' presentation
Date: Winter, 01.22.2001
Topic: TCP/IP for Everything, Over Anything
Abstract: The TCP/IP protocol stack is the core of the Internet. As the Internet expands its services to other domains (e.g., telephony) and devices (e.g., phones, TVs, handhelds), traditional protocols require modifications. Furthermore, the new applications entail increased heterogeneity of the underlying networks: traditional wired, high-speed, and wireless domains are now components of the Internet infrastructure. The TCP/IP stack needs to integrate efficiently these components and the service requirements of Internet applications, and to exploit network resources in an application-oriented and device-specific manner.
In this talk, I will justify some design choices of the TCP/IP protocol stack and discuss some design limitations, recent proposals and work in progress of IETF. I will argue that proposals are not always evaluated with the appropriate performance metrics, methodologies and goals and I will indicate a potential direction of the TCP/IP stack in the Next Generation Internet.

Prof. David Lorenz
Northeastern University College of Computer Science
Date: Fall, 11.13.2000
Topic: Software Components
Abstract: Components are "units of independent production, acquisition, and deployment that interact to form a functioning system" (Szyperski, 1997). The decoupling of software production and deployment (as exemplified by markets for third-party components) and the centrality of large-scale composition are largely why component-based programming lies "beyond object-oriented programming". These attributes bring with them many new and largely unexplored issues in software design. This talk underscores several of those fundamental design decisions in component-based programming and the design space of component-based design beyond object-oriented design. We illustrate how a good object-oriented design does not necessarily make a good component-based design, and we identify design decisions that are not a concern in object-oriented design. Throughout the talk, we use Java Beans to illustrate the concepts concretely.

Merrill Warkentin
Date: Fall, 10.30.2000
Topic: eCommerce Strategy
Abstract: Firms are facing new challenges in the quest for real success from the online revolution. Some companies have realized millions of dollars of actual profits, while others have yet to earn any. The real winners are likely to be in the business-to-business eCommerce realm, rather than the crowded world of B2C e-tailing. Beyond the technical infrastructure, there is an organizational business infrastructure evolving to fulfill the new requirements of companies going online. This eBusiness network allows firms to link their database systems effectively, facilitate efficient content delivery, support order fulfillment and payment, service supply chain processes, and enhance all aspects of business interaction. These evolving marketspaces will be discussed and presented with current examples. We will have a discussion of the impacts of these evolving trends on existing companies, on the economy, and on your career.

Merrill Warkentin is Associate Professor of Management Information Systems (MIS) and MIS Area Coordinator in the College of Business Administration (CBA) at Northeastern University in Boston, MA, where he teaches graduate and undergraduate courses in MIS, including systems analysis and design, electronic commerce, database management, decision support systems, information resource management, electronic research methods, and other information systems related subjects. He has authored over 80 papers, books, and book chapters. His research, primarily involving electronic commerce and Web strategies, computer-mediated communication systems and virtual teams, artificial intelligence and expert systems, and computer security has appeared in such journals as MIS Quarterly, Decision Sciences Journal, Information Systems Journal, Journal of Electronic Commerce Research, Expert Systems, HEURISTICS: The Journal of Knowledge Engineering & Technology, PC AI, AI and Medicine, Journal of Computer Information Systems, ACM Applied Computing Review, ACM SIGICE Bulletin, and The Journal of Intelligent Technologies. Professor Warkentin has served as a consultant to numerous companies and government agencies, and has been a featured speaker at over one hundred industry association meetings, executive development seminars, and academic conferences. He holds BA, MA, and Ph.D. degrees from the University of Nebraska-Lincoln (UNL).

Rick LaRowe
Baltimore Technologies
Date: Fall, 10.23.2000
Topic: PKI and it's applications
Slides: PKI and it's applications   Positions at Baltimore Technologies
Abstract: (N/A)

Yiannis Tsiounis
Date: Fall, 10.16.2000
Topic: Electronic Cash. What is it, and how does it work?
Abstract: Electronic Cash - What it is, How does it work? Just like their physical counterparts, there are several forms of electronic payments ranging from credit cards to debit cards to checks to electronic cash. One of the most important potential electronic payment methods is electronic cash. The reason is simply one of volume: cash accounts for the majority of consumer purchases in the Brick & Mortar world; teenagers alone have spent US $140 Billion in cash purchases in FY 1998 alone. The question is, what are the qualities of physical cash that make it so compelling as a payment tender, and how can these qualities be captured by an electronic cash system? Clearly there are some compelling reasons to use cash on everyday purchases:

• Convenience - cash is easy to use, it is always handy and it is fast, especially for small purchases
• Universal access (cash is available to all consumers - in contrast to, e.g., credit cards which are only available to 65% of US households)
• Universal acceptance (cash can be used everywhere - well, except on the Internet!)
• Privacy - cash is an anonymous and untraceable form of payment
• Security - a cash payment does not create the possibility of fraud (like a credit card or check does)

• Privacy - at least some degree of anonymity must be present
• Security - clearly, if fraud ever becomes an issue, any e-cash system will fail
• Convenience - this is simply a rule of the market: make something

Weld Pond and John Tan
The L0pht and @Stake
Date: Fall, 10.2.2000
Topic: Current State of the Art in Computer Forensics
Slides: Computer Forensics
Abstract: As crime moves increasingly into the computer realm is necessary that proper forensic techniques are developed for civil and criminal court cases. We will discuss the details of the 4 phases of a computer forensics process : acquisition, identification, evaluation, and testimony. We will cover the technical details as well as the legal details that need to be observed during the process.

David Juitt
Date: Fall, 9.25.2000
Topic: A History of Information Security
Abstract: The science of information security has matured significantly over the past three decades. Evolving from a host-based model to a network-based model; precipitated by the adoption and widespread use of the Internet, InfoSec has solved and created many challenges. This discussion will provide a grounding in some fundamentals and provide a brief tour o f how we've gotten to where we are today. The tour will conclude with some potential realities for the future of the field.

David LaPorte
Undergraduate Northeastern University College of Computer Science ACM
Date: Fall, 11.22.1999
Topic: Securing your Linux PC
Slides: Securing Linux
Abstract: David LaPorte, a Network Security Engineer at the Harvard University Network Operations Center (as well as former ACM President), will discuss the necessary precautions that need to be taken to prevent your Linux machine from being hacked. Special attention will be paid to machines on ResNet.

Jeff Ladino and Stephan Chenette
Undergraduate Northeastern University College of Computer Science ACM
Date: Fall, 11.08.1999
Topic: CCS Unix systems basics
Slides: Unix Basics
Abstract: Are you a freshman and new to Unix? Are all of your classes in PC labs? Do you want to learn the basics of the Unix operating system so that you can feel comfortable using the computers in 201 CN? Jeff Ladino and Stephan Chenette are both upperclassmen at CCS and will lead you through the basics of the Unix operating system. Come and bring your questions.

Donald Henrich
CEO of shoeboxx.com
Date: Fall, 11.01.1999
Topic: From NU Engineer to CEO
Abstract: Don Henrich is the CEO of shoeboxx.com, an Internet startup company. This is Mr. Henrich's third company. He will speak about his personal road from an NU engineering undergraduate to CEO.

Roger Brissenden
Date: Fall, 10.24.1999
Topic: Computer systems involved in the Chandra satellite
Abstract: (N/A)

Jeffrey Schiller
Network Manager, MIT
Date: Fall, 10.18.1999
Topic: Computer Security
Slides: The Barbarians Are Here
Abstract: This talk will discuss how computers are being broken into. Who are the "bad guys" and how are they doing it. What you can do to help prevent their intrusion in the first place and what to do once they are already in! We'll talk about password sniffers, buffer over-runs and common denial of service attacks.

Philip Graceffa
Technology Coordinator, Boston Athletic Association
Date: Summer, 07.21.1999
Topic: Tracking Boston Marathon Runners with the ChampionChip
Website: www.bostonmarathon.org
Abstract: The Boston Marathon uses a runner recording device called the ChampionChip. This chip tracks the athletes from Hopkinton to Boston by feeding data to 11 locations in real time. This data is then posted to the Boston Marathon web site in near real time.

Lars Hansen
PH.D Graduate Student Northeastern University College of Computer Science
Date: Summer, 07.21.1999
Topic: Modern Garbage Collectors
Abstract: Garbage collection (GC) is a strategy for reclaiming dead heap-allocated storage without aid from the programmer or her program; programming languages have had GC since before 1960. GC makes some classes of program bugs disappear entirely, makes program module interfaces smaller and reduce module coupling, makes higher-order and object-oriented languages and programs tractable. In some languages, GC is also instrumental in guaranteeing that object manipulations do not violate certain safety properties.

Many programming languages require GC, including Lisp, Basic, Java, Perl, Dylan, Modula-3, Simula, Smalltalk, Prolog, Postscript ML, and Haskell. Some, like Ada and Eiffel, allow it but do not require it. Garbage collectors even exist for unfriendly la nguages like C and C++, though the languages remain unsafe even with GC.

Garbage collectors have had a reputation for poor performance, high memory consumption, and outrageous pause times, but persistent work on design and implementation has to a large extent solved these problems. Current work on garbage collection focuses on making further improvements in these areas.

Lars Hansen will talk about basic garbage collector technology and elaborate on the software engineering advantages of GC. He will also talk about recent and current research on uniprocessor GC (including his own Ph.D. research at the College of Compute r Science).

Walter Eykel
Foxboro Company
Date: Summer, 07.04.1999
Topic: NU Coop to the Foxboro Corporation, Making the Transition
Abstract: Walter Eykel, of the Foxboro Company, will discuss his post-graduate studies and his transition from the co-op program to the corporate environment. Mr. Eykel will also discuss the opportunities and challenges he encountered at Northeastern University and at the Foxboro Company.

Prof. David Kaeli
Northeastern University College of Engineering
Date: Spring, 05.10.1999
Abstract: (N/A)

Paul Guglielmino and David Laporte
Undergraduate Northeastern University College of Computer Science ACM
Date: Spring, 04.26.1999
Topic: Linux Part 2 (Beginning Linux)

Brian Perry
Date: Spring, 04.12.1999
Topic: Creating a Successful High-Tech Startup

Prof. Harriet Fell
Northeastern University College of Computer Science
Date: Winter, 01.25.1999
Topic: Automatic Babble Recognition for Early Detection of Speech Related Disorders
Abstract: We have developed a program, the Early Vocalization Analyzer (EVA), that automatically analyzes digitized recordings of infant vocalizations. The purpose of such a system is to automatically and reliably screen infants who may be at risk for later communication problems. Applying the landmark detection theory of Stevens et al., for the recognition of features in, adult speech, EVA detects syllables in vocalizations produced by typically developing six to thirteen month old infants. We discuss the differences between adult-specific code and code written to analyze infant vocalizations and present the results of validity-testing. This work is done jointly with Linda J. Ferrier Department of Speech Language Pathology and Audiology, Northeastern University and Joel MacAuslan and Karen Chenausky, Speech Technology and Applied Research.

Jack Price
Director, Environmental Health & Safety, Northeastern University
Date: Fall, 11.16.1998
Topic: Healthy Computing
Abstract: The title of my talk will be "Healthy Computing". I will provide an overview of health and safety concerns for working at computer stations and what one can do to prevent repetitive strain injuries. This should be of keen interest for computer science majors who spend countless hours at their computers.

Prof. Judith A. Perrolle
Northeastern University College of Arts and Science
Date: Fall, 11.02.1998
Topic: Virtual Communities for Real People
Website: www.ccs.neu.edu/home/perrolle/fieldtrip.html
Abstract: Virtual communities exist in real life as geographically disbursed groups who manage to maintain themselves using communications and transportation technologies. The Internet and the web have given people new ways to build and sustain community, but they must deal with national and international struggles to control, own, and censor the net.