Creating a Virtual Center as an International Web-Based Interactive Infrastructure for Research and Teaching in the Language Sciences: A new Research and Library collaboration.

“Creating a Virtual Center as an International Web-Based Interactive Infrastructure for Research and Teaching in the Language Sciences: A new Research and Library collaboration.”

María Blume Cornell University mb48@cornell.edu Elaine Westbrooks Cornell University elw25@cornell.edu Cliff Crawford Cornell University cjc26@cornell.edu James Gair Cornell University jwg2@cornell.edu Tina Ogden The Ogden Consulting Group tina@tinaogdtina@tinaogden.com Barbara Lust Cornell University bcl4@cornell.edu

We describe here the web-based Virtual Center for the Study of Language Acquisition (VCSLA) now under development at Cornell University as a close collaboration between the Cornell Language Acquisition Laboratory (CLAL), Cornell's Albert Mann Library and a set of national and international partners. The purpose of this center is to foster and facilitate active and continuing interactive research on shared data involving many different languages. Taking advantage of the potentialities of the web, we bring together in a truly interactive way the expertise of researchers at CLAL and that of others at a number of scattered institutions, with the experience and capabilities of the library in information technology applied to storage of and access to shared data This first stage of development, financed by an NSF Planning Grant, has a number of crucial features, described here, that in assembly make it an innovative creation that may also serve, as both an ongoing enterprise in language acquisition research and a model for other fields as well.

1. While centered in CLAL, the VCSLA involves the active participation of national and international researchers at other institutions, incorporating a multi-directional flow of information and a perpetual linkage to the library.
2. It incorporates the Virtual Linguistic Lab (VLL), a new web-based interface for data transcription, analysis and access. This interface serves as a common but flexible framework for data entry and access in an interactive manner. This is well under development, and designed to be applicable to both research and teaching.
3. The enterprise embodied is cross-linguistic, including data and research on a widely spread set of languages that will be constantly augmented. VLL furnishes a detailed framework, assuring cross-language comparability, and facilitating both entry and access. It will include the considerable resources amassed in CLAL, and that provided by the participants physically located elsewhere. Thus it will necessarily be under constant revision as information flows in both directions.
4. The presence of the Mann Library in VCSLA is a special and unique element of its design. Mann Library brings its considerable expertise in the areas of data preservation, data archiving, and metadata management into force on the goals of the enterprise. This effort is consistent with the Cornell Library’s long standing active commitment to outreach activities that more firmly engage the library into faculty research and instruction in non-traditional ways, including several of its current digital and Virtual Library initiatives (See for example, “Reinventing the Humanities: Cornell Librarians and Faculty Members Create Electronic Collaborations”, Cornell Library READ!, Winter 2002).
5. The cooperative effort with the library is crucial to the functioning of VCSLA. Among other things, it links the VCSLA to the Open Language Archives Community (OLAC), an international partnership of institutions and individuals who are creating a worldwide virtual library of language resources, by agreeing to use The OLAC Metadata Set (OLACMS), a set of metadata elements for describing language resources. This component thus extends the informational and accessibility resources of the VCSLA.The OLACMS Standard uses XML to represent metadata descriptions, and metadata librarians at Mann are engaged in making the Cornell Language Acquisition Lab and associated VCSLA a metadata provider as well as a service provider within OLAC. That is, the library has built the infrastructure that allows the CLAL to share its metadata with other OLAC participants and it also has provided the interface that allows CLAL to harvest the metadata from other OLAC participants.
6. There are in existence other forms of electronic data sharing for child language research, notably CHILDES (Carnegie Mellon). VCSLA differs from and is complementary to these in both scope and design, most notably in the incorporation of the VLL format for comparability of data entry and access, and in its essential interactive component. That is, it does not represent simply a databank that allows researchers to access and enter data, but a facility that allows interactive and cooperative communication and research by active scholars who are physically separated by location. It is also designed to be a useful element in the training of students, especially potential researchers in language acquisition Also, as an International Web-Based Interactive Infrastructure, it is designed to incorporate new participants and laboratories worldwide, including areas and countries where resources for training and research in the field are limited or nonexistent.

In our presentation we will sketch each of the above components of the VCSLA and its progress to date. We present concrete examples of its applications to both research and teaching in the language sciences.