Publishing Your Data to the VO
If you have data that you think would be interesting for the astronomical community, and you are not a journal editor, data center manager, or database expert, then this page is for you.
There is a more comprehensive guide to data publishing called How to Publish to the NVO, that elaborates on what is on this page, and gives more specific directions.
In order for your data to be useful for others, you will need to provide not only the data itself, but it must also be documented with metadata. In the VO-world, communication is as much between machines as it is between humans and machines; therefore, this documentation is more useful if it is formally structured (XML). The VO provides web forms and pages to make this simple for you.
Once this descriptive metadata has been published to a VO-compliant registry, your data will be visible from every VO-enabled application around the world.
The next question concerns the publication of the data itelf. If the data is a collection of FITS images, for example, the VO will provide tools to catalog that data—meaning harvesting the FITS headers so that your images will be properly catalogued, registered, and available through VO-compliant services.
Spectra can already be published, where they can be uploaded, hosted, managed, and exposed through the VO registry along with existing large collections.
Catalogs will be published in the XML-based VOTable format. In order to make this data useful to others, they must be able to understand your data. In addition to column names, physical units, etc, the VO has a formal vocabulary (UCD) for describing the meaning of data—there are 100 ways to express a magnitude that should be translated to a common language. The UCD system has been derived from a careful analysis of thousands of published astronomical datasets.
When data is published, it may be your own machine that is exposing data through VO-compliant services. You can build your own implementation of the services, or download code for this from VO.
Alternatively, we are exploring ways for astronomers to avoid the need to maintain a data server. Small datasets could be uploaded to a community archive server. We are currently building alliances with archive centers, University libraries, and journal publishers to enable this model.