HyperDic is a web dictionary derived from the Wordnet database. HyperDic is neither a program nor a database, but a collection of web pages, that can be read on any system, with any browser that supports standard HTML. The online version of HyperDic 3.0 covers over 145,000 English words, while the HyperDic 1.0 CD-rom from 2001 covers 120,000 dictionary entries.
There are a few differences between HyperDic and WordNet, mainly related to the systems' overall ease of use. The main advantage of HyperDic is the versatility and ease of use provided by standard web text, directly accessible on any modern computer system. The interface is well known from the start, and well integrated with other programs. Expressing the conceptual links between related words directly as web links makes the word lookup easier, so the overall user interaction with the system can be a quite laidback experience. At the same time, the hyperlinks empower the reader to freely follow word associations, and facilitate learning.
Since HyperDic is a perfectly standard HTML text, any page can be easily converted to a wide variety of formats, by any good text processing program. You can import whole pages in your text editor, or just cut and paste text excerpts.
HyperDic runs in a browser, so you can take advantage of the features offered by standard browsers, such as the possibility to print pages, or to bookmark them for quicker access.
With all modern operative systems, it is possible to establish shortcuts on the desktop, to directly launch your favorite programs. This feature can be used to start HyperDic on any chosen page, by a single mouse click. If your browser is well integrated with the operative system, dragging this link to the alphabetical index, and dropping it somewhere on your desktop, should normally set up a permanent shortcut that can be activated at any time.
Since HyperDic is not a program, it does not use computer resources, beyond what is needed to run a browser. All information relevant to each word is stored in a separate file, and is only read when the user requests it. Thus the only resource used by HyperDic is storage space, which has always been much cheaper than memory or processing power. With the online dictionary, there is also a bandwidth issue, which is not dramatic, since the pages are soberly coded in XHTML Basic 1.0. XHTML is a subset of XML and, as a consequence, HyperDic interacts nicely with the new generation of XML-based semantic web agents.
Depending on the filesystem, the full version of HyperDic occupies between 500 and 700 megabytes of storage space. Thus, it is relatively easy to locate a copy of HyperDic on a harddisk, and access it from a small wireless device, such as a pocket computer, with very limited computing capabilities, and minimal strain on the server.
This is in sharp contrast to typing a word in a searchbox, which requires additional effort from the user, to ensure that the spelling is correct. An increasing number of computer users prefer not to use the keyboard at all. Also, a number of small computers do not even have a keyboard.
For these reasons, the best way to look up a new word is perhaps to surf into HyperDic's Alphabetical Index, which also lists alternative spellings, and derivations of common stems.
HyperDic was developed by Eric Kafe.
Starting with the WordNet 3.0 databases, intermediate databases were produced, using a number of traditional Unix-tools, in particular awk and sed.
During this process, the original WordNet 3.0 databases were slightly expanded formally, resulting in the introduction of some novelties. In particular, antonymy was changed from a relation between words to a relation between word sets, thus yielding many new antonym pairs. A few inverse relations were also derived, so that all links in HyperDic are now bi-directional.
The final dictionary was generated using a script written in the Python programming language.