Future Plans

Functionality and implementation¶ ↑

An advanced search option would be nice. Users who are familiar with SQL can choose quite a few search options by typing wildcards into the search form. It’s a good idea to have an advanced-search page where a user can choose a number of options (like starts-with, ends-with, exact-match, contains) that are available using SQL wild cards. An advanced search can also allow users to specify a part of speech or perhaps search the contents of word definitions (as well as any other fields later added to the data model). This can be displayed as a separate page or perhaps as a sidebar that can be hidden.

Users will eventually want to print from Dubsar. That requires review and probably significant work on all fronts (main and mobile websites, native applications).

XCode 4.1 includes the Mac OS X 10.7 SDK for Lion. Dubsar for Mac will soon follow.

Spelling correction after a search, especially one that returned no results, might be a good idea. (“Did you mean…?”) I recently spent some time poking around the site for diaphonous. Finally I consulted a paper dictionary to find that it is actually spelled diaphanous.

I want to review Rails web servers (options with Rack), though it may make little difference when deploying with Passenger.

It might be interesting to write a Dubsar plug-in for looking up definitions and synonyms of key words in any text. The WordNet data set is well suited to that task. This could take the form of:

• A Rails plug-in for looking up words and providing things like synonym spans for template rendering or general-purpose access to the data model.

• A plug-in for applications and controls that perform spellchecking or similar text-handling functions.

See below regarding better lexical identification of words from context to narrow synonym lists.

Here are a few other longer-term possibilities:

• Support for other languages, possibly including translation

Content¶ ↑

I’d like to add more content to the application by further exploring the WordNet® data set or finding others. This kind of application is only as good as its data. Here’s a wish list of things to add:

• Etymology

• Pronunciation, including audio

• Remaining parts of speech (conjunctions, interjections, prepositions, pronouns)

• Illustrations (somewhat rare in online dictionaries)

• Inexact searching (soundex, metaphone, etc.)

• Map views for geographic entries, particularly since WordNet provides regional information. For example, when viewing dubsar-dictionary.com/senses/107650, the fact that domain of synset (region) says, “Great Britain, United Kingdom of Great Britain and Northern Ireland, Britain, U.K., UK, United Kingdom,” would eventually produce a map view showing that location.

• Better lexical word identification: If Dubsar can better recognize a word’s part of speech from its context (e.g., inflection, position relative to other words, etc.), its synonym list can be narrowed.