Account Services

I have

Reportnet architecture

Uploading a report to CDR might look deceivingly like uploading to CIRCA library, but there are some important differences.

First you will notice that the description (or metadata) is not placed on each file, but on an envelope, which can then contain many files. This is because a report is often comprised of several files, and there is no need to repeat the metadata for each file. This is very helpful, as one only has to enter metadata once per obligation.

The second observation, is that files are expected to be unzipped in the repository. This makes it possible to inspect/download individual files and also harvest them for search engines.

What you may not realize, is that the search and navigation to get to the reports has been decoupled from the hierarchial structure. In CIRCA you must set up and maintain a rigid hierarchial structure of folders to indicate to the clients where you have put a specific report. In CDR all this is done with the metadata. Then the metadata is harvested by the Content Registry harvester on a regular basis. A special dataflow search form lets you see what a country has reported or which countries have reported CLRTAP data no matter what repository they have used. The only reason for a hierarchial structure in CDR is to let metadata be inherited from the collections to the envelopes.

If wanted, a MS can develop its own repository on other servers than this one. This is because the content registry centralizes searches, and only requires that a country supports the interface required by the content registry so it can harvest the metadata. There is no need for structure. The reports can lie helter-skelter on a webserver somewhere with meaningless filenames as long as the correct meta-data is reported to the content registry.