HowTo: Adding to the Dataset List

From OPeNDAP Documentation

Goto the Dataset List now.

About the Dataset List

This wiki-based dataset list is an attempt to harness the power of the community of data providers to build and maintain a list of all OPeNDAP data servers. All "OPeNDAP" servers can be listed here; this is not a list for just Hyrax! If you are running a public server that DAP clients can access, we would like you to list it here! Similarly, if you are here looking for data and see a link that's broken or a site that is missing links, please use the contact information to let someone at that site know their information here needs to be updated. If all else fails, drop us a line at support at

Goto the Dataset List now.

Information to include with all datasets


For each dataset you list, please include contact information! This is the single most important piece of information from the standpoint of keeping the list accurate and your work to a minimum. If you have contact info bound to your datasets, then it's easy for people to get in touch with you, which means they can help you keep your datasets current. When you include this information, please try an use an email address that won't go stale quickly!


The info link references a web page that describes your data. You can organize this anyway you would like, however, be aware that this provides often very important information to both people and web crawlers (that build indexes like Google that people then us to find your data) so it's potentially a very link important!

A GCMD entry link

This link is similar to the info link in that it provides a general description of a particular data source. Unlike info which is a (link to a) web page, this is a link to a page/entry in NASA's Global Change Master Directory (GCMD). The GCMD is a list of data sources where information about each data source is monitored and maintained by specialists in different research areas such as physical oceanography. Include this information if you data set(s) have been cataloged by the GCMD staff!


The link labeled dir should point to a directory page for your server. Usually when DAP is used to serve data, those data are stored in collections of files that are stored on disk. The value of this link should be the top directory in the file system that holds those files. It's fine to list several links if that's how your data are organized.

The HTML for interface

The HTML is not for a web page that describes the entire data server's contents (use info for that) but instead is a link to the HTML form interface that all DAP servers provide. This provides an easy way to build URLs that access data. Hyrax servers provide ways to get data in NetCDF files and as ASCII among other things. This might not be appropriate for some really large sites (servers with many data sets) because the page will get cluttered very quickly.

Summary of suggested links

email or other contact information for the person in charge of the server.
A link to an HTML page that people can read (and crawlers can grovel).
A link into the Global Change Master Directory that describes this dataset.
A link to the directory page provided by the server.
A little confusing, this is not a page of static HTML but a form that a person can fill in, see a DAP URL built up in response to changes in that form, and then get data as binary, ASCII or a netCDF file. A useful tool for exploring a dataset.

Adding/editing an entry is easy

A site with one server

If your site runs only one data server it's best to list each major dataset under a heading named after you organization. Use a heading made up of two equals signs like:

== OPeNDAP's Test Server ==

Under that heading, include an info link if you have a web page that explains your server/site. This is a great way to help people looking for data and a also an easy way to get web crawlers like Google to index your site. If you prefix the link with a colon, the entry will be nicely indented. Like this:

 : [ info]

For each major dataset that you want to list individually, prefix them with an asterisk and then include info, dir, and other links as you see fit. It's a good idea to choose a name that short but informative (of course!). However, keep in mind that most people who use this list will not be familiar with your specific data. Here's an example for some of our test data:

* Freeform Test Data: [ dir]

Note that we don't have an appropriate info page for this, it is a simple directory listing, so we've just included a dir entry for the data set (which is really a collection of things we use for testing client applications). Here's a complete example and following it, the resulting page as it appears in a browser:

 == OPeNDAP's Test Server ==
 : [ info]
 * Freeform Test Data: [ dir]
 * NetCDF Test Data: [ dir]
 * COADS Climatology Test Data: [ html]

And what it looks like in a browser:

OPeNDAP's Test Server

  • Freeform Test Data: dir
  • NetCDF Test Data: dir
  • COADS Climatology Test Data: html

Sites with more than one server use subheadings

If your organization runs several servers (and it's fine to list several different kinds of servers like Hyrax, TDS, PyDAP, GDS, ..., here) then use two headings, one for your organization and separate subheadings for each server. To do this, make the organization entry using two equal signs (==) and the serve entries using three (===). Here's a partial example from NASA's JPL facility. Note that for clarity I added newlines between the various links in the listing below. It looks more polished if you leave those out in a real entry, but it makes the listing harder to read on this page.

== NASA/JPL - Physical Oceanography Distributed Active Archive Center (PODAAC) ==

=== SSM/I derived global heat and momentum fluxes (Atlas, Jusem, Ardizzone); Product #004 === 
: [ dir] 
[ gcmd] 
[ info] 
=== Ocean Wind === 
* SSM/I Pathfinder gridded ocean wind speed level 3 (NOAA, NASA, MSFC, Wentz); Product #057: 
[ dir] 
[ gcmd] 
[ info] 
* NSCAT scatterometer science product, levels 1.7, 2, 3 (JPL); Product #066: 
[ dir] 
[ gcmd] 
[ info] 

Note that if you want, you can use the heading and subheading format with one server - there's no rule against that - and, similarly, you can use this in other ways to organize your data.