How to Participate in THDL

The Tibetan and Himalayan Digital Library operates as an archive and publisher for scholarship and data on any aspect of the Tibetan and Himalayan geographical regions. It provides technical systems, technical support, standards, and maintenance for a wide variety of projects run by different individuals, institutions and boards. Thus rather than being one project based at any one place, it offers an umbrella to be used for diverse initiatives. The Library thus actively supports scholarly projects by providing technical assistance and digital publishing forums for their materials. In addition to projects conceived and created by the Library to solicit and organize collaborative input and materials from around the world, the Library also supports as collaborative enterprises projects that originated outside and which have independent institutional bases.

The Library's various publishing facilities can be divided loosely into four types.

1. Firstly are the broadly collaborative resources such as the Gazetteer and Dictionary. These are designed to be developed by a diverse range of scholars working on-line over many years, with each contributing granular bits of information to extensive essays into a larger joint resource which, however, maintains individual credit.
2. Secondly are the individual projects, such as specific literature collections, a specialized dictionary, a video collection, and so forth. These are run by specific individuals, institutions, or teams, and are presented as such. Thus while they may draw upon and utilize the larger joint resources, they are also profiled and presented independently.
3. Thirdly is the Journal, which enables scholars to incorporate multimedia into their publications, and provides academic credit for substantial essays that are then also incorporated within the relevant portions of the library.
4. Finally the Library offers a way to make digitally available articles and books which out of print under the heading of the "Publications Archive".

Our major priority at present is to open up collaborative input into THDL so that individual scholars can submit small resources, as well as open up new projects of their own design. The current document first details the long term plan for facilitating and managing collaboration, and then provides a summary of present possibilities for collaboration. While we are in the process of developing automated submission forms and guidelines, as well as an associated electronic journal, interested collaborators and contributors should contact us at thdl@virginia.edu.

THE LONG TERM PLAN

INTEGRATED LOG-IN SYSTEM

Our long term plan involves a single integrated system where users log-in, and their ID allows them to submit and manage materials in certain resources, but not in others. For example, a specialist in Tibetan language might be able to work on the dictionary, and control a specific thematic dictionary within the overall dictionary but be unable to submit items, for example, to the Gazetteer. Or, a scholar might have complete control over a particular project on a given place that they are running, but not necessarily be able to submit materials anywhere else in the library. Access privileges will be granted by the various individuals and boards that oversee the resource in question. This log-in system allows many scholars, librarians and others to control their own resources in tightly controlled ways within a very large library system characterized by diverse projects and administrations.

USER-FRIENDLY WEB-BASED MANAGEMENT INTERFACES

Submission and management of resources will be done directly over the Web using user-friendly templates that insulate the user from the underlying programming language or database structure. This is both to avoid data structures from being corrupted through inadvertent deletions or alterations of the underlying structural elements, as well as enable users to work with their resources without having to learn difficult and confusing technologies. We thus aspire to provide the same straightforward and streamlined interfaces that ordinary desktop software or commercial Web sites provide, such that individuals can submit, revise and manage data in powerful ways without having to learn complex new technologies. Thus each interface provides, as relevant drop down menus and preset options, but also is accompanied by detailed documentation and help files that orient users to the necessary standards to follow in entering data. In addition, these interfaces will provide the ability to specify that the submitted data is intended to remain "private" while one continues to work on it over a periods of days, weeks, or months, or that the data become immediately "public" so that upon submission other users of the library will have access to it immediately. These interfaces thus function to both restrict users' ability to corrupt data by limiting access to underlying structures, and enable their ability to work with data in powerful ways with a minimum of training by providing clean and clear access to complex structures.

CREDITING AND WORK FLOW TRACTING

These collaborative interfaces will also stamp every submission with the name of the submitter and date of submission, thereby taking care of issues of crediting as well as work flow management. Presentations of the data will thus clearly display the name of the data's creator/revision and the date of creation/revision. This is not only essential for general purposes of crediting, but also for the tracking and crediting of each submission within a resource collaboratively built by a number of different people. Thus, for example, a person who contributes a single etymology for a single term in the dictionary must have that contribution credited to them, even if others have worked on other aspects of that term. These interfaces thus ensure complex tracking of the authorship of work for crediting and work flow management.

COLLABORATIVE MANAGEMENT SYSTEMS

A related issue pertaining to collaboration is the integration of work from multiple agents on a single subject in ways that both ensures the diversity and integrity of each submission, but also provides for coherent displays of the overall data as it accumulates for quick consultation as well as extended analysis. This is especially true when a resource within the library is intended to be broadly collaborative where multiple scholars may be submitting very granular pieces of information on the same item. For example, the dictionary – if forty scholars are working on building a rich, OED-style dictionary, each from their own angles, the potential for rich, scholarly conversations on given terms is clear. Yet the potential for messy, complex accumulation of data hard to digest or use is also painfully clear. This is without doubt the most difficult challenge that faces large scale collaborative projects using Web-technology, and the points outlined here are simply our current strategies.

Firstly, as pointed out above, every submission no matter how granular is stamped with the submitter's name and date of submission, and revision of those materials are restricted to the original submitter. This provides clear attribution and integrity for individual submissions of data and scholarship. Secondly, in large collaborative resources, our data structure allows for the creation and maintenance of separate projects within them, which can be extracted for coherent management and viewing as individual resources even while also usable within the broader resource. Thus a scholar may run a "X's medical dictionary" within the broader dictionary, even though the same terms might be worked on by other scholars as well. At any given point, however, the scholar will be able to extract not only just those terms s/he has marked as "X's medical dictionary", but also extract only his/her submissions on those terms and thus divest it of anyone else's work done on the same terms. The same applies to work on a particular place or region in the Gazetteer. To highlight this work, such projects can also be given their own home pages within the library that are controlled by the project director, and explain the goals, intent, status, and participants of that project. This is similar to how in a traditional library, one has thousands of books, but when one takes a single book of the shelf, it is presented in its own integrity. We will also enable filters that would allow one, for example, to view the dictionary with only certain submitters' entries viewable, and the rest suppressed. Another strategy is to have a group of scholars determining "standard" descriptions – whether for terms, places, or anything else in a given project – which are of reasonable length and synoptic scope. Marking these as "standard" allows a user looking for quick consultations to not be distracted by the wealth of other information. Finally, each project requires careful attention paid to the presentation of the data, since formatting and structure of the visual display can make a huge difference in striking the balance between recording diversity, and maintaining coherent integration. In these ways, we can work towards building new structures that provide proper credit, and enable coherent, integrated presentations of data while also recording rich, scholarly conversations between diverse perspectives.

QUALITY ASSURANCE

Finally, we would like to address the issue of quality assurance, an issue of compelling importance in the Web-era with its reality of free and immediate publication. How do we ensure high quality in the Library's resources and many projects, and especially in large scale collaborative resources in which many people are simultaneously working without any explicit coordination between them as individuals? This involves two distinct issues: quality of the data formally in the sense of its technical usability within the library, and the scholarly worth of the submissions as scholarship.

The first issue pertains to the need to avoid data corruption that breaks the underlying data structure, as well as irregular data, such as divergent phonetic schemes for representing a language, thereby rendering searches across different domains dysfunctional. Data corruption is dealt with primarily by attempting to tightly control direct access to underlying data structures, and offer collaborative interfaces that restrict contributors to bounded text boxes and/or preset drop down menus as outlined above. In addition, formal and regular review procedures attempt to catch corruption before it expands, while the commitment to carefully designed and implemented data structures from the beginning is essential. Irregular data is a more difficult issue because there one must rely upon the commitment of a wide number of people to adhere to standards. While one can to a limited degree make data entry templates reject certain types of data that do not adhere to the required formal structure, there is no way to do this across the board for all types of irregular data. The same applies for standard procedures such as having a different person proofread data entered by any given person. All communities are based upon the adherence to standards, and digital communities probably even more so. We thus are committed to thorough documentation of all projects providing comprehensive guidelines, introductions, and help documentation for contributors to rely upon. In addition, we are building remote training modules that we will insist each contributor proceed through prior to making any submissions. We are all also working hard at developing community standards, such as transliteration and phonetic rendering standards, through consultation, but also through then providing powerful software solutions relying on those standards. We thus utilize a mix of consultation and attraction; ultimately, people adhere to standards because involvement in the corresponding community is of compelling enough importance to justify the modest restrictions and learning curve/time that adherence to given standards entail. These procedures ensure that the underlying data adheres to standards that enable their long term persistence as well as integrated utility across the library's various domains.

The more complex issue is to ensure the scholarly worth and integrity of data with so many scholars and projects. There is no simple answer, anymore than one exists in the conventional world of print journals and book publications. Rather, there is the reliance upon a complex set of checks and balances which ultimately stands, or fails, upon the decision making process of the individuals who run it. To begin with, just because a ;given project such as the dictionary or gazetteer is potentially open to contributions from a large number of people, does not mean that anyone can participate. Each resource is governed by a small, rotating board of individuals with the appropriate expertise, who make decisions as to who is given rights to contribute within that resource. Other boards also consider applications for the opening up of individual or institutional projects within the library, thereby ensuring that new projects have the appropriate intellectual integrity and necessary expertise behind them. These boards also continue to monitor work, and as necessary revoke such rights when they are abused or used inappropriately. The log-in system then keeps track of these rights, and ensures that individuals and institutions can control their own resources, but not make submissions or alterations in areas to which they have not applied for, and been granted, permission. In addition, for collaborative projects, the data structures, collaborative interfaces, and delivery systems build a strong and explicit coordination of work into the system, even when the scholars in question are not necessarily in direct verbal communication with each other. This also helps support the integrity of collaborative scholarship on a broad scale. These procedures ensure that the scholarship and other submissions of the Library are of the highest possible intellectual and artistic integrity through a complex system of checks and balances that govern other systems of publishing and archiving intellectual and artistic productions.

THE STATUS OF PARTICIPATION TODAY

As we work towards this long term goal, the status of the ability to collaborative on-line, as well as off-line, is in a constant state of flux and varies from area to are within THDL. The current section provides a simple overview of where things stand today in general, but is in no sense exhausts the possibilities for collaboration, while much is happening behind the scenes each month that is not necessarily recorded herein. People interested in possible collaboration – whether or not a viable framework is listed here – are thus encouraged to contact us at thdl@virginia.edu, and we can pursue it in discussion.

SUBMIT IMAGES, VIDEOS, OR AUDIO RECORDINGS TO MULTIMEDIA COLLECTION

Media submissions are open to scholars, professional photographers, and the wider public in dependence upon the quality of the images, as well as the character of their documentation.

At present we have a complex image cataloging template we have not put on-line yet, but can be used off-line. We are also still finalizing documentation. Please contact us directly at thdl@virginia.edu to participate in the image database. In 2002, we will provide facilities for users with permission to upload images from their computers into THDL, as well as catalog those images on-line after uploading; we will also provide an off-line means of pre-cataloging your images before they are uploaded. We are also creating detailed documentation in our Digital Ethnography site by February 2002 that will discuss necessary criteria to be used in digitizing photos, guidelines for use in the field when taking photographs, and the cataloging system we are using. Depending on the nature of the collection and available staff resources, we are open to applications for help in digitizing and processing significant collections.

We are currently refining our interactive video/audio work flow management database, which allows for videos/audio and transcripts to be uploaded, as well as displayed in coordination with each other, in addition to detailed cataloging entries. However, videos/audio are far more complex to manage than images and their data, and hence at present we ask interested parties to contact us directly at thdl@virginia.edu.

Every submitted media is marked with the creator's name, and copyrights remain with the creator(s), though THDL has permanent non-exclusive rights to display the media within its Multimedia Collections, and participants in THDL are allowed to incorporate the media within their THDL projects. Creators can specify individual specifications in terms of the copyright details within the catalog template. This includes whether people are allowed to use them in Web-pages outside of THDL, and whether people are allowed to use them in print publications. For large collections we can also mark them as belonging to a named collection. Collections can be given home pages within THDL where the creator or administrator of the collection can detail the nature and contents of the collection, and other relevant issues.

SCHOLARSHIP ON THE COLLECTED TANTRAS OF THE ANCIENTS (RNYING MA RGYUD 'BUM) OR THE ORAL TRANSMISSION OF ZHANG ZHUNG (ZHANG ZHUNG SNYAN BRGYUD)

Both of these Tibetan literature thematic research collections offer published deep level catalogs that are open for submission of editions of texts, critical editions, summaries of chapters, translations, and other scholarly analysis. They are only open at present to scholars with competency in literary Tibetan. At present, we are only partially automating submissions as we gain experience – please contact samantabhadra@virginia.edu and bonpo@rice.edu respectively to participate.

CULTURAL GEOGRAPHY

There are three ways users can currently participate in the broader project: submission of textual documentation of places to the Gazetteer, submission of maps and media visualizing places, and essay length data of text and/or images/audio/videos presenting places. While we have not yet implemented on-line means of submitting data for these resources, nor have we actively sought materials while we complete designs, we are ready to work with interested individuals on integrating their materials manually. Please contact us at thdl@virginia.edu

1. Gazetteer: this involves short synoptic descriptions of places or regions, as well as documentation of place names (in different languages, including specification of source and time period), features a place contains, is contained by, or intersects with, georeferencing data (latitude, longitude, altitude), and so forth. Submissions can thus be as simple as a name and paragraph long description.
2. Maps, media and associated text: this involves arranging maps, images, videos, audio recordings and/or textual data about features within a place into a coherent presentation. This can range from a simple sketch map and pictures to complex projects.
3. Essays on places: we are developing analytical templates for recording essays on places that focus on particular subjects – artistic traditions, economy, etc. In the meantime, we can publish such essays in association with the places.

It should be noted that focus to date has been on culturally Tibetan areas now controlled by China, but that we are in 2002 focusing on other areas of Cultural Tibet, as well as other southern Himalayan regions such as Nepal, Kashmir and Bhutan.

DICTIONARY

We have designed and implemented a powerful system modeled after the Oxford English Dictionary that allows for a rich documentation of words, including their use in literature over time. This is now viewable on-line on thdl.org under "dictionary" as implemented for Tibetan. However, we are currently redesigning it to (i) make the various categories more keyed to definitions, and also produce a cleaner, less cluttered view of the dictionary. During this long process of design and redesign, we have deliberately chosen to not input extensive data. We anticipate beginning extensive data input in the first half of 2002. This design will accommodate individual projects, such as a medical dictionary, music dictionary, and so forth run by particular scholars or teams which can be presented individually. Please contact us at thdl@virginia.edu.

We have not at present implemented this for other languages such as Nepali, Dzongkha, and so forth, but several discussions are under way. Please contact us with suggestions or offers of participation/leadership.

BIBLIOGRAPHY

We are now putting the final touches on a powerful bibliographical system that allows bibliographers from around the world to manage thematic bibliographies from anywhere in the world using easy to understand Web interfaces. Thus a librarian in the US might manage a bibliography on Nepali textiles, while a scholar in Paris might manage a bibliography on Tibetan history. It includes materials in any language, whether primary texts, translations, or secondary scholarship. The system should be ready for use in the first quarter of 2002, and thus we are actively soliciting offers of participation. Each bibliography will be accessible from a general search page, but will also have its own independent home page where the bibliographer presents a digest of the resources available. We have decided to build the broader range of themes gradually in conjunction with offers of participation, so please contact us at thdl.@virginia.edu.

MISCELLANEOUS SCHOLARSHIP

Scholars with miscellaneous editions of Tibetan texts, translations, word indexes or any other such useful materials languishing on your hard drives, please contact us at thdl@virginia.edu to discuss how we could help publish these for a broader audience.

NEW PROJECTS OR BROADER COLLABORATION

If you have a project of any type that you would like to direct, or are interested in a broader collaboration with any of the extant projects, please contact us at thdl@virginia.edu

SOFTWARE

We welcome contact from anyone interested in collaboration with us on software initiatives we already have under development, or other projects. Please contact us at thdl@virginia.edu.