Daniel X. Pape

dpape@dpape.com
http://www.dpape.com/

OBJECTIVE

SKILLS

• Languages: Ten years of C++. Additional experience with Perl, C, Java, bash.
• Internet Technology: Multi-threaded client/server programming with ACE; HTML; shell scripts; rdist/rsync; XML and DTD development
• Systems: Linux, Solaris, FreeBSD, Windows NT
• Databases: Interfacing to Oracle with PRO*C
• Research: familiar with neural network and machine learning algorithms for data mining applications; text and image analysis for search and retrieval;
• Proven ability to investigate new areas of research, create new programs and algorithms, and improve design and performance of older ones.

WORK EXPERIENCE

August 2000-present
Senior Software Developer, MongoNet, Inc.

I am one half of the engineering team responsible for creating Mongonet's Fax-to-Email conversion service. This service is comprised of ten distinct subsystems, mainly written in C++. I was involved in all aspects of our service.

• Design
• Created initial design of multiple job servers coordinated by central job manager servers.
• Wrote initial specs for all the services and helped define multiple milestone goals for release.
• Researched available technologies to base our services on.
• Designed flexible messaging interface for server communication (XML).
• Development
• Made prototypes in perl of all separate services.
• Created generic server code on which all concrete servers were based. This code ran on linux, Solaris and Windows NT.
• Established and became maintainer of repository for all third party libraries our services depended on.
• Defined standard error and status logging practices for tracking and measurement of servers.
• As concrete servers became more mature, took ownership of image cleaning and conversion servers, optical character recognition server, central job manager and message router, central logging server.
• Testing
• Created generic unit testing code. Made initial unit testing programs for most existing classes.
• Designed and implemented the automated unit testing scripts.
• Became the company enforcer of all future unit testing.
• Wrote scripts for service and load testing of entire system.
• Made several log parsing programs for analysis of server correctness, performance, and concurrency.
• Distribution
• Wrote perl and rdist/rsync scripts for automatic pushing to linux, solaris and Window NT machines.
• Created startup scripts for all services on all OSes.
• Maintenance
• Made web page for reporting current server status. All servers could be started, stopped, installed and deleted from this page.

March 2000 - August 2000
Senior Software Architect, Tank Hill, Inc.

• Responsible for managing a small group of five software engineers and a UI designer.
• Involved with design and initial implementation of Tank Hill's peer-to-peer file sharing application, Ohaha. Ohaha was meant to be an interface to both the napster and gnutella networks.
• Researched file sharing protocols of napster and gnutella, and began design of meta-protocol that would let our application query and talk to both networks.

1996-2000
Research Programmer, Community Architectures for Network Information Systems (CANIS) Laboratory.
University of Illinois at Urbana-Champaign

• Designed and implemented a self-organizing neural network for the categorization of collections of text documents and image databases. Designed the computational core of the neural network and the 3D visualization of the network's output. The network was parallelized to operate on an eight-node Beowulf Linux cluster, and the OpenGL visualization runs on either a Linux or SGI workstation.
• Conducted research on measuring the ability of self-organizing neural networks to recover cluster structure within a data set. Implemented one program to generate artificial data sets, and another to compare different clustering solutions of a data set and measure the correspondence of the clusterings with each other.
• Created an application which allows the user to interactively browse through a number of images by specifying "texture samples." The program searches through a pre-computed thesaurus to find images that contain textures which are related to the specified samples.

1995-1996
Research Programmer, National Center for Supercomputing Applications, Mosaic for X Window System team.
University of Illinois at Urbana-Champaign

• Participated in the early development of NCSA Mosaic for the X Window system. Most of my C programming work centered around Motif GUI design for the application's front end, and some cryptography work for the application's meager security options. Gained experience developing under many UNIX environments, as Mosaic supported nine UNIX variants.

1993-1994
Software Consultant, Equilibrium Technologies, Inc.

• Developed graphics plug-ins for Equilibrium's flagship program, DeBabelizer. These plug-ins dealt with game platform (Sega and Nintendo) graphic sprite utilities and file format handling, specifically, TIFF v5.0, CCITT Group 3 and 4 compression standards. All work was done remotely.

1992-1993
Research Programmer, Imaging Systems Laboratory, SmartForest Project.
University of Illinois at Urbana-Champaign

• Created interactive 3-D forest ecosystem modeling environment with OpenGL. The visualization displayed three-dimensional tree symbols scaled and colored to represent numerous aspects and dimensions of tree stand data. The trees could be queried directly using the mouse to display data from the underlying database. Various dynamic models were incorporated, such as indices of crowding, tree-to-tree competition, and pest and fire hazards. These models could be used to evolve the forest through time to investigate the effects that changes in the environment would have on the forest ecosystem.

EDUCATION

1998-2000
Master of Science, Library and Information Science.
University of Illinois at Urbana-Champaign

1991-1997
Bachelor of Science, Computer Science.
University of Illinois at Urbana-Champaign

PUBLICATIONS

Refereed Journals

K. Powell, C. Chang, Y. Chung, D. Pape, Q. He, L. Tyrrell, N. Bennett, and B. Schatz (1999) The Interspace Prototype: An Analysis Environment Based on Scalable Semantics, Science, submitted.

Technical Reports

C. Chang, K. Powell, Y. Chung, Q. He, D. Pape, L. Tyrrell, N. Bennett and B. Schatz (1999) The Interspace Prototype Overview: Semantic Indexing and Retrieval in a Persistent Environment, Technical Report UIUCDCS-R-99-2094 (UILU-ENG-99-1715), Department of Computer Science, University of Illinois at Urbana-Champaign.

D. Pape, (1999) Useful 3D Virtual Environments, Videogames, and the CANIS Interspace CANIS Tech Report 0009, Community Architectures for Network Information Systems, University of Illinois at Urbana-Champaign.

Refereed Conferences

D. Dubin and D. Pape (2000) Clustering Applications and Validation: A case study with the Kohonen SOM, Classification Society of North America 2000 Conference, Montréal, Québec. June.

D. Dubin and D. Pape (1999) Validation Strategies for Large-Scale Clustering Applications, Classification Society of North America 1999 Conference, Pittsburgh, PA. June.

D. Pape (1998) Visualization and Navigation of Document Sets: Information Spaces Using a Self-Organizing Map, 1998 Joint Annual Meeting of The Psychometric Society and The Classification Society of North America, Urbana, IL. June.

Other

B. Orland, D. Pape, J. Obermark (1992) SmartTrees: Forest modeling and visualization. Imaging Systems Lab., Dept of Landscape Architecture, Univ. of Illinois, Urbana, IL.

INVITED TALKS

D. Pape (1998) New Term Pruning Algorithms, CANIS Brown-Bag Lunch Series, CANIS, University of Illinois. Urbana, IL. August.

D. Pape (1997) Categorization and Visualization of Information Spaces with a Self-Organizing Map, Seminar on Information Retrieval in Digital Libraries: The Technologies of Scalable Semantics, University of Illinois. Urbana, IL. October.

PROFESSIONAL SOCIETIES

Member of:
Classification Society of North America.
Mathematical Association of America.
Association for Computing Machinery.
American Society for Information Science.