A database for digital assets

Question

I work at a print shop that has a decent sized archive of old print jobs. Currently, to find a job we have to search the SMB share (on win2000 server), which looks through a few hundred thousand files.

Our job data is organized by: Year>Month>Customer_Name_Job# and then the contents of the job are inside the last folder.

Is there a way to create a database so that we can query a job desc. or job number? Basically just a simple search that is faster than searching with windows search would be great. I tried the windows index service but it only finds .doc file, not folder names or PDFs.

In the future we are planning on making a SQL database where complete info will be entered when the job is created so it is easily queried in the future. I was hoping to throw together something that will make it easier to search older orders.

Answer 1

I personally wouldn't try to design a database from the ground up for this one -- what you're looking for in the library and archives field would be called a 'digital repository' (one step abstracted from a 'digital library', which usually is only tracking bibliographic items). In the business world, they talk about stuff like Digital Asset Management, but sometimes that's done through registries, not repositories; the different being that in a registry, you only track information about the object, but it only points to the object, rather than storing the authoritative copy of the object.

See if one of these projects could be adapted to fit your needs:

• http://fedora-commons.org/
• http://www.dspace.org/
• http://oodt.apache.org/

Or, see a larger list of free & open source digital repository software

Answer 2

You could use a Document-oriented database for this. You could then create a program in your preferred language to import the existing documents into the db, parsing the folder structure for the metadata (customer, job#, etc).