WebLicht is a service-oriented architecture (SOA) for creating annotated text corpora. Development started in October 2008 as part of CLARIN-D's predecessor project D-SPIN, and further development and enhancement of WebLicht is an important goal of CLARIN-D, aiming to make WebLicht a fully-functional virtual research environment.
WebLicht employs chains of RESTful web services. Each web service encapsulates a certain linguistic tool. For example, users can access as a web service the query component of a corpus, a format converter, a tokenizer, a tagger, or a parser. Translation between the input format specific to some tool and the WebLicht information interchange format TCF (see below) is performed by a web service wrapper. Each web service adds at least one layer of annotation encompassing the work of the tool encapsulated by that service. The output of a chain of WebLicht services is an automatically analyzed corpus in the form of an XML document.
To do this, each WebLicht service must be able to use a common interchange format that all the other services can also process. CLARIN-D's Text Corpus Format (TCF), serves this purpose. It is broadly compatible with existing related interchange formats like Negra, Paula, or TüBa-D/Z, and format-specific converters allow interchange between them.
WebLicht can be accessed only with a valid DFN-AAI/Shibboleth-based account or a local Tübingen account.
