I have written NooJ from scratch, with the exciting C# / .NET / Visual Studio computing environment, using all benefits of the "Component Programming" methodology, as well as the free automatic memory management!
Using my 10-year experience as the INTEX author, I have redesigned every single aspect of INTEX’s tools and methodology:
1. A New Linguistic Engine
-- NooJ dictionaries are a unified enhancement of INTEX’s DELAS, DELAC, DELAF, DELACF and DELAV dictionaries. They can encode syntactic information and thus are capable of storing lexicon-grammar tables: internally NooJ does not make any difference between DELA-type dictionaries and lexicon-grammar tables; for instance both can be displayed in a list form or a table form.
Thanks to its unified inflectional and derivational morphological engine, NooJ dictionaries represent all inflected and derived forms as one lexical entry: no more multiple and independant lexical entries for derived forms such as: launch, relaunch, lanchable, relaunch and launchability!
-- NooJ grammars are protected structured sets of graphs. Now users can share and adapt a grammar, or part of it, without fear of destroying it. NooJ grammars are associated with a contract that garantees that a grammar cannot be accidentally broken.
-- Because NooJ grammars are not destructive, they are very easily combined in cascade (no more MERGE/REPLACE mode).
-- NooJ morphological and syntactic engines are integrated: this allows linguists to program morphological operations that are performed during a syntactic transformation.
2. NooJ for teaching linguistics and second languages
-- NooJ includes Labs that are little demo applications which can be used to demonstrate a linguistic phenomenon, or a computational issue. Examples of labs: inflectional morphology, text encoding, perl regular expressions, etc. The dynamic .NET architecture allows any programmer to build new labs for NooJ in a couple of hours (Teachers: Design your labs!).
-- NooJ allows teachers to save their current working configuration, as projects. Projects include all opened windows, lexical and syntactic resources (with their priorities), as well as a documentation.
-- Students can build, accumulate, filter and export complex concordances.
-- Grammars’ contracts can be given to students, who can then test their own work.
3. A New Set Of Functionalities
-- NooJ is multilingual
-- NooJ can process corpora, i.e. sets of (hundreds of) text files
-- NooJ understands over 100 file formats, including four variants of UNICODE, all DOS/Mac/Windows/ISO/EBCDIC variants, all DOS/Windows/Mac MS-WORD documents, HTML, RTF, etc.
-- NooJ understands XML texts: NooJ can process XML tags exactly in the same way as NooJ’s own lexical and syntactic information. NooJ can also export its information to XML documents.
-- NooJ various types of linguistic analyses are non destructive. No more MERGE/REPLACE mode! All linguistic analyses’ results are stored in an annotation database that is synchronized with the corpus/text. In consequence, it becomes much easier to perform complex analyses in multiple passes (i.e. apply grammars in cascade).
-- NooJ has been designed to be used to develop and maintain large numbers of dictionaries, morphological and syntactic grammars. Several functionalities, such as file history (unlimited undo/redo), contracts, support for various levels of comments, etc. should help linguits develop, share and maintain their resources.
Using my 10-year experience as the INTEX author, I have redesigned every single aspect of INTEX’s tools and methodology:
1. A New Linguistic Engine
-- NooJ dictionaries are a unified enhancement of INTEX’s DELAS, DELAC, DELAF, DELACF and DELAV dictionaries. They can encode syntactic information and thus are capable of storing lexicon-grammar tables: internally NooJ does not make any difference between DELA-type dictionaries and lexicon-grammar tables; for instance both can be displayed in a list form or a table form.
Thanks to its unified inflectional and derivational morphological engine, NooJ dictionaries represent all inflected and derived forms as one lexical entry: no more multiple and independant lexical entries for derived forms such as: launch, relaunch, lanchable, relaunch and launchability!
-- NooJ grammars are protected structured sets of graphs. Now users can share and adapt a grammar, or part of it, without fear of destroying it. NooJ grammars are associated with a contract that garantees that a grammar cannot be accidentally broken.
-- Because NooJ grammars are not destructive, they are very easily combined in cascade (no more MERGE/REPLACE mode).
-- NooJ morphological and syntactic engines are integrated: this allows linguists to program morphological operations that are performed during a syntactic transformation.
2. NooJ for teaching linguistics and second languages
-- NooJ includes Labs that are little demo applications which can be used to demonstrate a linguistic phenomenon, or a computational issue. Examples of labs: inflectional morphology, text encoding, perl regular expressions, etc. The dynamic .NET architecture allows any programmer to build new labs for NooJ in a couple of hours (Teachers: Design your labs!).
-- NooJ allows teachers to save their current working configuration, as projects. Projects include all opened windows, lexical and syntactic resources (with their priorities), as well as a documentation.
-- Students can build, accumulate, filter and export complex concordances.
-- Grammars’ contracts can be given to students, who can then test their own work.
3. A New Set Of Functionalities
-- NooJ is multilingual
-- NooJ can process corpora, i.e. sets of (hundreds of) text files
-- NooJ understands over 100 file formats, including four variants of UNICODE, all DOS/Mac/Windows/ISO/EBCDIC variants, all DOS/Windows/Mac MS-WORD documents, HTML, RTF, etc.
-- NooJ understands XML texts: NooJ can process XML tags exactly in the same way as NooJ’s own lexical and syntactic information. NooJ can also export its information to XML documents.
-- NooJ various types of linguistic analyses are non destructive. No more MERGE/REPLACE mode! All linguistic analyses’ results are stored in an annotation database that is synchronized with the corpus/text. In consequence, it becomes much easier to perform complex analyses in multiple passes (i.e. apply grammars in cascade).
-- NooJ has been designed to be used to develop and maintain large numbers of dictionaries, morphological and syntactic grammars. Several functionalities, such as file history (unlimited undo/redo), contracts, support for various levels of comments, etc. should help linguits develop, share and maintain their resources.
after INTEX
