Welcome to my digital library research (DLR) page. This is my professional page. Currently I am working as Head of Digital Library Research with General Libraries at Emory University. Previously I worked at the Virginia Tech Digital Library Research Lab (DLRL) as a Graduate Research Assistant on the CITIDEL project. I left when I finished my Master's, and came to Emory in late October 2003.

Here I've placed materials related to my current and past work and research. You can contact me at akrowne@emory.edu if you have queries.

• "Quality Metrics" - A project to determine in detail how intuitive user notions of quality relative to digital library record attributes bear on results set presentation. The outputs of this project are user study data, a theoretical model, and a prototype search system sensitive to quality metrics.


• MetaCombine - My central focus at Emory. MetaCombine is a Mellon-funded project about "combining" digital library resources and services (and digital libraries themselves) in an improved, more meaningful way.


• PlanetMath - My own project (and thesis topic), a mathematics site with an encyclopedic focus and pedagogical slant. Think MathWorld but done collaboratively and with peer-review. This project is currently unfunded but very active, and support is being sought.


• OCKHAM - A collaborative project we're a part of at Emory. The goal is to build a decentralized network of digital library services, to help the spread of useful resources and service tools.


• CITIDEL - Computing and Information Technology Interactive Digital Educational Library. This is meant to be a CS/IT education "mega-site". My GRA work revolved around CITIDEL. This project is currently unfunded and inactive, though the site is still kept running.

• The MetaCombine Software - A clustering system, a visualization system, a scheme editor, a focused crawling system (FCS), semantic clustering (OCKHAM) web services, an OAI repository insta-copier (OAICopy), and much more. All outgrowths of the MetaCombine project (and to some extent, OCKHAM).
  
  
• Noosphere - The software behind PlanetMath. A collaborative encyclopedic digital library framework, supporting LaTeX.
  
  
• Vinstall - A collection of scripts which gives the file-level illusion of a linux system within a linux system, allowing projects to be "sandboxed" from each other. This eliminates library conflicts, package system dependency conflicts, and yields other benefits.
  
  
• ESSEX - Efficient Scalable Search Engine for XML. Developed to be a light, featureful search engine for CITIDEL. Coded in C++, with some use of STL. Follow the link to read more about its capabilities. Note: This is in working condition currently, but check the README in the archive for caveats. This should be considered "alpha" software; there is a TODO list in the README that we could use help with.
  
  
• LaTeX::TOM - Where TOM = "TeX Object Model." This is a native perl LaTeX document parser and handler, inspired by XML::DOM. Created for both the CITIDEL and PlanetMath projects, it is potentially useful to very many people in the Digital Libraries field, due to its ability to extract and distinguish true "plain text" portions of LaTeX documents.
  
  
• XML Logging for Digital Libraries - Java logging server and Perl, Java clients included. I helped design this system and wrote the Perl client.

• The FUD-Based Encyclopedia - A defense of Wikipedia and commons-based knowledge resources from a rising onslaught of fear, uncertainty, and doubt. I propose some basic principles of CBPP production in here, to illuminate why (and when) it works, and demystify the phenomenon. I also discuss how integrating respect for expertise is not antithetical to CBPP — PlanetMath includes it implicitly with its ownership system.
  
  
• Combined Searching of Web and OAI Digital Library Resources (and extended version; JCDL 2004) - About building a combined search service over both native digital library resources and web resources via exposing the DL resources through the web.
  
  
• The Effectiveness of Automatically Structured Queries in Digital Libraries (JCDL 2004) - About increasing search precision for metadata-rich documents by inferring metadata field clauses for queries.
  
  
• An Architecture for Math and Science Digital Libraries (master's thesis) - My thesis, about PlanetMath and Noosphere.
  
  
• Authority Models for Collaborative Authoring (HICSS 2004) - About authority models (owner-centric and free-form, as well as a mix of the two) for collaborative content creation. This is an empirical study, aiming to determine if one of the models is "better". This is now a chapter in my thesis.
  
  
• Building A Digital Library The Commons-Based Peer Production Way (D-Lib mag. Oct 2003.) - Discusses commons-based peer production of digital libraries, using Noosphere/PlanetMath as a case study. Good intro to my thesis.
  
  
• An Architecture for Multischeming in Digital Libraries (extended version of ICADL 2003 paper) - Discusses issues with multiple classification schemes in digital libraries in a federated scenario and how to solve them. CITIDEL uses methods discussed in here.
  
  
• The XML Log Standard for Digital Libraries: Analysis, Evolution, and Deployment (JCDL '03) - About the XML log standard developed in the DLRL. I had a part in designing the deployment (basically the socket architecture and protocol, as well as the client-side implementation in Perl, for CITIDEL).
  
  
• Building Digital Libraries from Simple Building Blocks (Online Information Review, Vol 27 No. 5, 2003) - This article is basically about Open Digital Libraries (ODL). I contributed some sections, talking mostly about the use of actual ODL components in CITIDEL, and things in CITIDEL which could potentially be componentized.

• Noosphere: An Architecture for Collaborative Math and Science Digital Libraries - My thesis defense presentation.
  
  
• CITIDEL and Resource Discovery - Presentation I gave to Dr Fox's ISR class on resource discovery in CITIDEL (none of which is yet implemented, though we will be carrying the CiteSeer collection).
  
  
• PlanetMath Automatic Reference Linking - Nearly all of the interlinking of PlanetMath entries is done automatically, and this document discusses how it works.
  
  
• Fast Query Repair in PlanetMath - Discusses the method I adapted of correcting search engine queries. This method should be useful especially in niche search engine scenarios.
  
  
• A CITIDEL Overview - HTML "slides" of a presentation I gave to the DLRL on April 1, 2002.
  
  
• Exploring CITIDEL - This was constructed for attendees of the PKAL conference (June '02) as a walk-through exercise to get acquainted with CITIDEL. This makes a good place for anyone to start, however, and doubles as a demo.
  
  
• Crawlifier Research Report - By Eric Tressler, a student who worked with me on my "crawlifying" research. Good intro document if you are curious.
  
  

• Virginia Tech Home Page - The school where I did my master's and undergrad.
  
  
• VT CS - Computer Science dept. at Virginia Tech.
  
  
• DLRL Homepage - The Digital Library Research Lab at Virginia Tech, where I work.
  
  
• The Open Archives Initiative - Digital library interoperability.
  
  

• Dr. Fox - My former MS advisor and PI of CITIDEL. Also director of the VT DLRL and a bunch of other stuff.
• Marcos Goncalves - Phd student at the VT DLRL whom I still work with.
  
  

• citidel/
  
  
• papers/