Matt Biddulph

Curriculum Vitae

Location
Europe (flexible)
Email
mbcv@hackdiary.com
Date of Birth
9 February 1976

Education

1994-1997: Oxford University (Balliol College)
Computation BA Hons 2:1
1997-1999: Leeds College of Art and Design
Film Studies A Level evening class
1987-1994: Matthew Arnold School, Oxford
French, Maths, Computing A-Level, 11 GCSEs

Professional Experience

October 2005 - Present: Freelance Software Designer
hackdiary.com
A range of projects for clients including the BBC and Nature Publishing Group.
BBC Programme Catalogue - a huge searchable database of the BBC's archive of programming. Written using MySQL and Ruby on Rails.
More information
January 2004 - September 2005: Radio and Music Interactive, BBC
Architecture Software Team Leader
Managed a team of four software developers and was responsible for technical leadership on projects such as:
June 2001 - December 2003: Interactive Operations, BBC Broadcast Ltd.
Senior Systems Architect
Designed and implemented the input and output tiers of a scheduling and control system for interactive television. Worked with a third-party supplier on the design of the middle tier of the system, creating XML document standards for messaging to allow data exchange between the tiers. Wrote Java code to transform and route messages to a system of adapters that control playout systems for all the BBC's channels on Freeview, Sky Digital, NTL and Telewest Cable. Optimised the system for the realtime performance requirements of broadcast television.
Developed and supported the BBC website's promotional banner serving system.
Researched statistical techniques for web traffic data mining using Netgenesis. Analysed server logs for the BBCi website (including a case study of unusual traffic around September 11th, 2001) for patterns of user behaviour with particular attention to site-to-site comparisons. Created sampling methods to allow processing of smaller amounts of input data while retaining results representative of the complete data set.
February 2001 - June 2001: Pharmalicensing
Software Developer
Worked on a short contract to create a new phase of features for this pharma/biotech licensing portal. Using MySQL and a mixture of Python, PHP and Perl I implemented new sections of managed content, added user customisation features and improved the security of the authentication system.
Working with Debian Linux servers and ISDN hardware, helped to plan and install an office network with print server, firewall and fileserver to support Windows clients over wired and wireless ethernet.
July 2000 - January 2001: TW2.com
Senior Developer
Led a team of five developers to produce Printpotato, an online self-service printing service, using J2EE and Microsoft SQL server technologies.
Technical lead on a team of eight developers building a managed-content website for Aegis based on the SilverStream J2EE server and Oracle.
Planned and taught developer training classes on XML and object-oriented design.
August 1998 - June 2000: Ananova (previously PA NewMedia)
Technical Architect (January 2000 - June 2000)
Software Developer (August 1998 - December 1999)
Worked on a variety of web projects involving news feeds and syndication, initially as a developer and graduating to the role of Technical Architect, responsible for system designs, R&D and mentoring of developers.
Built the Java- and XML-based content management system for the first generation of the Ananova website and its computer-rendered audio/video streams.
Created search engines for TV and events listings, designed and built a library of generic perl objects for reuse in projects across the development team, and implemented a migration of the company's web-hosting cluster of SGI servers from Netscape Enterprise Server to Apache.
1997 - 1998: Information Today
Developer
Developed the first web version of Index to Theses as an online search engine using Linux, WAIS, HTML, perl CGI and mod_perl.
Prototyped an online web search engine for Microcomputer Abstracts, a large electronic journal of computer magazine articles.
Created a Microsoft Access database with multi-site data replication to manage marketing mailing lists for Rubicon Communications
1993 - 1997 (part-time): Expert Information Ltd
Developer
Wrote text processing, indexing and installation code for Index to Theses CD-ROM, a commercially-available academic text database, using Visual C++, Microsoft Access, Turbo C, TeX, and Omnis 7.
1992 (part-time): Learned Information Ltd
Technical Support
Acting as assistant to head technical support officer, responsibilities including PC hardware maintenance, diagnosis and repair; administration of NetWare network; telephone assistance to office workers.

Personal Projects

Planet RDF
An aggregator site for writing about Semantic Web technology.
The Daily Chump Bot
An IRC bot for collaborative weblogging.
RDF Crawler
A Java crawler for the semantic web.
Picture Diary
Photo journal, run with semantic web markup and RSS.
hackdiary
Writeups of personal technical work.

Presentations and conference papers

The Application of Weblike Design to Data - Designing Data for Reuse
"The World Wide Web is a database, but it's not tabular and relational, with a defined schema and referential integrity. That's why it works. However, there's no way to run something as simple as an SQL SELECT statement across all the data on the web. It's an open world, and you never know whether you've found all the facts; few questions can be answered categorically.
"There is a strong trend towards putting real machine-readable information on the web in open formats, and building public web APIs on top of existing website functionality. Without careful designs, data on the web is going to struggle to reach its potential. By applying principles of good website design to the process of modelling and deploying data on the web, your information stands a better chance of being discovered, queried, reused and built upon.
"In this paper, we will demonstrate how careful information modelling and URL design in the relaunch of radio websites at the BBC leads to a useful and useable data model."
Given at XTech 2005, where I was also a member of the programme committee.
Reinventing Radio: Enhancing One-to-Many with Many-to-Many
"How could you enhance a one-to-many national radio station by building in the many-to-many-style interactions of Flickr or the weblog community? How might lessons from social software further blur the distinction between listeners and broadcasters by pushing interactivity beyond the phone-in or the online poll?
"The session presents work from BBC Radio & Music Interactive's Technical Architecture and R&D teams, including demonstrations of existing software and working prototypes of new projects."
Given at Emerging Technology 2005 with Tom Coates, Paul Hammond and Matt Webb.
BBC Programme Information Pages: An Architecture for an On-Demand World
"What would become possible if every episode of every program that the BBC broadcast had a unique ID and URL?"
Given at Emerging Technology 2005 with Tom Coates, Gavin Bell and Margaret Hanley, and at Euro Foo with Tom Coates.
Crawling the Semantic Web
"This presentation examines the problem of semantic web crawling - following links from document to document and gathering the results for searching. Unlike centralised web search facilities, semantic web agents will be distributed, personalised and often highly domain-specific. How can we hold the entire world inside our laptops?"
Given at XML Europe 2004.
A Semantic Web Shoebox - annotating photos with RSS and RDF
"RSS 1.0 is a well-known XML format commonly used for syndicating news headlines. It is designed as an extensible format. This presentation discusses some interesting applications that can be built using RSS and RDF to annotate collections of photos."
Given at:
RDF and the Semantic Web
Lightning talk given at London Perl Mongers techmeet.
Designing J2EE applications
Introductory talk at the SilverStream Users Conference, London 2001.