Konstantin There are several ways in which you can use the type value pairs inside handles. This off course includes the storage of any values including other persistent identifiers. The persistent of these values is guaranteed by the update function performed by the handle owner and their actionability can also be defined by this owner or other applications. For example: some applications include configuration details, rights _expression_, TCP IP addresses or even some very simple versioning information as is the case with the Advanced Distributed Learning Registry Schemas. The following proxy request : http://hdl.handle.net/2000.2.1/ADL-R-Reg-T?type=Vdata can actually be used to request the XML document that describes the different versions of this schema and therefore use an application to traverse this information and make sense of it. An example of such application for instance is the CORDRA proxy that allows XML validators to request a particular version of the Schema to perform validation ie: http://hdl.cordra.net/2000.2.1/ADL-R-Reg-T?VData=version&id=1 . This is only one of the possible uses of the Handle system. Other applications include using the handle system as a high speed distributed authentication database and even as an operation rules resource. Certainly there are many other users of the Handle system and its architecture is generic enough so that you could fit any type of information on it. The only issue is to try to keep it as lean and fast as possible so that you can keep acceptable performance on your local servers. This means that you should try to avoid overloading the information you store on the system in order to keep query responses small. I hope this helps Best regards Henry Henry Jerez Senior Research Scientist Corporation for National Research Initiatives ---------------------------------------------------- This email is considered confidential and should be restricted only to the parties originally intended as its recipients. Please do not distribute or quote this email without specific consent of the author. Unless otherwise stated the contents of this email are not the official position of CNRI. --------------------------------------------------- On Mar 28, 2007, at 7:42 AM, Konstantin Rekk wrote:
|