Print

Print


DICT is a protocol, like HTTP, FTP, SMTP, NNTP, LDAP, yada yada yada.
dict.org is the home page for the group responsible for developing the
protocol.

In particular, dict is for allowing dictionary data to be exchanged
between software systems.  It's designed to be used by other programs,
not by end users.    This allows the creation of many sorts of tools,
such as rollup pages that let you query multiple dictionaries at once
(see: dictionary.com).  Basically, you stick a dict server in front of
your database, and you get a query system customized for dictionary
data that is safe to expose to the world (unlike SQL, which is so
general you daren't expose it; see: SQL injection) and there are lots
of tools that will know how to talk to your dictionary and do data
analysis and such automatically.

Even if you go the dict route, there's nothing to keep you from
writing your own custom front-end web interface (in Java or PHP or
Ruby or Python or Perl or ...) to do what you like with the data.  You
could have the web app use the same raw data source as the dict
server, or you could have the web app go through the dict server,
which would be looser coupling and therefore more resilient in the
face of future change. (In terms of MVC patterns, the model class
representing the dictionary would be an Active Resource instead of an
Active Record if the app talks to the dict server instead of a
database.)