A number of the hypermedia spaces documented here predate "the web" and "HTML" by several years. They include what may be the first GML tag-based hypermedia document, the first preloaded hypermedia on-line documentation system, the first application specifically written to invoke hypertext in a context sensitive manner, the first hypermedia book, the first client browser display of hypermedia from a server, and the first distributed hypermedia server system to run on any network anywhere. See the description of HELP for more details of how this work has arguably contributed to the development of the web as we know it.
"Hypermedia space" does not, then, necessarily mean "web site" here. A hypermedia space is a collection of interconnected and thematically related web pages, generally sharing a common look and feel, that can be regarded as a unit. A single web site can contain many hypermedia spaces. A single hypermedia space may be spread across many web sites and/or contain other hypermedia spaces. Note that, perhaps somewhat uniquely, this resume is itself a hypermedia space, and that, in its documentation of my development work, it also documents my software implementations, agent software implementations, architectures and systems, and efforts associated with the development of new communication media.
Rationalized the disparately located web resources of the NCA Human Communication Technology division into the hctd.net url. Moved a substantially updated primary HCTD web site to my server space at www.hctd.net. Integrated divisional NCA call for papers at callforpapers.hctd.net. Similar server references integrated other divisional resources. Created an interactive collaboration site for the division at wiki.hctd.net.
Global PartnerInfo is a large hypermedia site that encompasses over 250,000 documents, all operating within a single managed navigational structure and look and feel. My involvement with "GPI" has been largely an architectural relationship that is covered elsewhere in this resume. I have, however, spent large amounts of time on GPI system performance and look and feel issues. I didn't design the sites look and feel. Most of the credit for that goes to Olgivy and Mather and IBM's Enterprise Web Management team. I have, however, had a significant role in translating their design into an operational look and feel that performs well across the network and under heavy loads.
Specified the API's for the back end HTML service that supported the NetFlare internet diagnostic user interface. Combined HTML, JavaScript, JAVA applets, and a Visual Basic client application. This user experience operates within the overall NetFlare diagnostic testing system at several levels and from a variety of entry points. First, it actively participates in the testing process by providing test probes. Second, it supports a progress indicator that is maintained in collaboration with the test server. Third, it prompts the user through the the diagnistic process through scripted dialogue involving questions, forms, informational pages, results pages, and diagnostic repair pages. Fourth, it allows initiation of the configuration test procedss from either Windows clients (using a VB Windows launcher application) or cross-system clients (directly from the web interface. Finally, it provides detailed help and diagnostric information from its associated web site.
Quite possibly the fastest interactive computer conferencing site on the web, IBM's Toolsrun/2 based web infrastructure was designed to enable the movement of IBM's forums from its traditional home on VM to the web world. Implemented using a combination of Javascript, Java applets, Java servlets, and C++. Hosts several thousand interactive forums that are publically accessable within IBM. Provides subsecond response time access to most content, even across a 28.8 modem connection. My architecture and user interface design based on an initial concept by Terry Heath. Developed and operated by the Globenet team, most notably Peggy Frisch and Mai Pham. This pointer to one copy of this site won't work except inside IBM.
A specialized web site that allowed users to schedule teleconferences and view coordinated web displays while teleconferencing. More details of the capabilities of this site can be found in my description of TurboConferencing. This site was largely built by others, but I was responsible for several key elements, including its underlying design.
A special java-driven website that allowed MMDC clients to automatically check for, and install, software updates. Entailed JAVA application code on the client and a special java server. Very cool.
My company web site. Operated using secured Wiki collaborative composition technology such that the content of the site changed from any web browser. Solves a lot of site content maintenence problems.
The pages you are looking at right now. Originally developed in 1994 using HTML as a single file with a large number of named anchors. Evolved, in 1997, into a comprehensive academic curriculum vitae with a look and feel that is still visible as the non-frame variant of this space. Created a version with the kind of complex look and feel that it has to day using JavaScript and frames in 1999. This is an ambitious JavaScript web space design with a distinctive underlying document architecture. Professional recruiters have called this the best looking hypermedia resume they've ever seen. It has evolved over time. Prior versions of this resume web site illustrate how the look and feel as evolved over time. You can view the look and feel of the site as it looked in 1999, 2001, 2002, 2003, and today.
A web portal for finding people with the surname Foulger who have sites on the web. Mostly done for fun, but this site also provides the new host location for my personal web site.
Created a full web site in support of my students at SUNY Oswego. The top page contains references to news and schedule information. Pointers to all of the classes I have taught and am currently teaching are maintained. For each class there is a full on-line class syllabus, including a detailed day by day outline of the course. For each lecture there is a set of available lecture notes. Additional pointers provide information on my research program. Encompasses hundreds of pages of content.
Created a collaborative composition (Wiki) site in support of my collaborative research activities. Several discrete interfaces to this site are provided, including interfaces in support of student access to class materials (in a large font interface appropriate to teaching) classroom display of those materials. The full range of interfaces to this system is described in a base page.
A complete HTML and JavaScript rendering of my 1990 dissertation on the evolution of computer conferencing in IBM. The realization of a longstanding dream.
Intended to act as a set of pointers to other hypermedia spaces, my personal web site wouldn't be worth mentioning if it didn't mix a variety of web site construction materials to build an entertaining whole. The pointers remain, but they are now accompanied by a Java game (also visible from its own page) I developed in 1996 and some JavaScript that allows you to exert some control over the Java program. This version of puzzle allows you to solve puzzles of variable dimension and has been programmed to ensure that all puzzles are solvable.
A limited-time only site, in terms of interest value, that has some interesting graphical features. Note, in particular, the graphical overlay on the first page of the site against a half-toned background. The use of backgrounds, like this one, to give a site thematic consistency, is a common tactic in web site design. The selective overlay of meaningful elements of the background into the site foreground is far less common, and kind of cool. Portions of this site point detail community service work I've done for the Wappinger's Central School District, where I've been active in promoting the use of the Internet both in classroom instruction and as a means of promoting school district activities and acheivements.
I first heard about HTML in 1989 when my friend Charles Wilkes told me about a small group of folks he was working with at CERN to build an Internet hypermedia system. We had a long conversation about how it was similar to, and different from, my HELP hypertext system. I didn't hear much about it again until 1991 when, as a participant in the National Academy of Science symposium on "Rights and Responsibilities of Participants in Networked Communities, Murray Turoff told me that it was the next big thing on the internet. I didn't actually get around to using HTML until 1993, when I set up my own server. That server, named "rogue", provided my principle personal presence on the web until 1998, when its 70 Meg hard disk proved inadequate to continued service. My initial experiment with HTML, and the base on which a number of other sites were evolved over time.
OS/2 help manager is a direct design descendent of HELP and is almost indifferent, syntactically, from the first versions of HTML. Although I cowrote the requirements and design specification for OS/2 Help Manager, and the software accepts HELP content without modification, I have generated only a couple of OS/2 Help Manager HyperMedia spaces:
Written with Dave Slauson as the on-line documentation deliverable for ForaBrowse and Talklink. Distributed both electronically and on diskette. Later updated to encompass ForAgent.
An experiment, done as I neared the completion of my doctoral dissertation and the OS/2 Help Manager entered broad use. Over half a megabyte of cross linked materials. Unpublishable in this format given the usual practice associated with doctoral dissertations in 1990, but later used as the basis for my HTML hypermedia dissertation (see above)
I developed HELP in 1984 as an online documentation tool in support of the Yorktown PC User's Workbench. HELP was one of the first PC Hypertext systems anywhere. It debuted at almost the exact same time as Ben Schneiderman's Hyperties (late 1984), but had very different structural characteristics. Hyperties was a control-based Hypertext system (as were several subsequent offerings from other sources including the Brown University system, OWL Guide, and Apple HyperCard. Help, by contrast, was designed to use text markup (e.g. intent based tags in the tradition of GML (which predates HELP) and SGML (which postdates HELP). It borrows a number of tags directly from the GML starter set (credits to Geoff Bartlett), including list and paragraph tags, but introduces some unique tag elements including the anchor tag and the href (originally help reference) argument. Note that both of these elements are reproduced (perhaps by convergent design) in HTML. The big difference between HELP and HTML is in the server file structures. HELP puts many HTML pages in a single file and uses an index file to rapidly pull out individual pages. HTML simply breaks things out into individual pages. There are arguments to be made for each approach, but faster storage media and processors make the HTML approach more viable.
To be useful, HELP required content, and I developed a number of HELP hypermedia spaces. Initially those spaces provided the online documentation for the Yorktown PC User's workbench, but I later experimented with those spaces in interesting ways. I have documented some of the HELP hypermedia spaces I wrote in the 1980's here, focusing on those that are interesting in one way or another.
Quite possibly the first true hypermedia book (although other Hypermedia systems may well have developed a book equivalent earlier), this hypermedia space provided a complete reference to the REXX programming language. Has since been ported to HTML (a trivial task for HELP tagged hypermedia spaces). Other people (names now forgotten) did most of the writing. I simply organized things and put the result into distribution (sort of a webmaster role).
Possibly the first use of hypermedia across a distibuted network. This large hypermedia documentation space, which included both site, educational, and application related hypermedia content, was distributed across several servers on the premises of the IBM Corporate Technical Institutes rather than being stored directly on an end users machine. I had very little to do with generating the content for this hypermedia space. That was the work and vision of Larry Margolis. I was, however, a key player in making sure the software would work in referencing, and pulling content from, a distributed network of file servers. I also supplied some class materials, in hypermedia format, related to an Intel 808x Assembly Language class I taught there from time to time.
Notable not so much for the content of the hypermedia than for the fact that the application was written to call a hypertext program in a context sensitive manner. Possibly the first time context sensitive hypermedia help was designed into another application. I note, in this regard, that Hyperties was initially used to create online books (its first big public use was in Kiosks at a museum), the Brown University and OWL Guide systems were oriented towards providing hypermedia access to educational materials, and that Apple Hypercard was designed to allow people to create multimedia applications. Probably the program that comes closest to providing what Fileman provided on PC's was UNIX man pages, but man pages were never structured as hypermedia and have always been run as a distinct application from a unix command line.
A comprehensive overview of a broad variety of DOS filters, including several written by myself, that documented various ways of building and using filters under DOS. Hardly the first online collection of documentation for a set of filters (that honor clearly goes to UNIX documentation utilities), but almost certainly the first to do so using hypermedia, including hyperlinks between related filters and detail sections for important options. Probably also the first hypermedia documentation that allowed execution of a program from within a hypermedia panel.
Possibly the first time hypermedia, tag-based or otherwise, was displayed on a client while being retreived from a server in real time. This hypermedia space didn't involve any new content. It was simply an image of the Yorktown PC Users Workbench that could be run on a PC from a disk image on a remote VM system using IBM's PC/VM Bond software product (developed by the group I belonged to at the time).
Over 1 megabyte of end user documentation, covering every application distributed with the Yorktown PC User's Workbench. In its initial release, this included context sensitive help for a DOS version of the VI editor, hypermedia documentation for Personal Editor, context sensitive help for FILEMAN, documentation for the Easymenu menuing system, and, perhaps uniquely, context sensitive documentation for a variant of the classic Unix VI editor. Subsequent versions added hypermedia documentation for STP, various members of the E editor family, an array of DOS filters, and the REXX programming language. Probably the first comprehensive hypermedia documentation delivered as a preinstall on any computer.
Quite possibly the first GML tag-based hypermedia document ever written. Provided details on how to use the HELP Hypermedia software and generate HELP hypermedia documents.