SAS 9.x Windows 2000 Professional

Summary: Staff in my deparment, who don't have SAS installed on their computers, have need to read SAS data into Excel for further analysis.

Details: For the sake of brevity, I will only summarize what I've done. I can do what I want if I create a User or System DSN, then open that DSN (in C:\Program Files\Common Files\ODBC\Data Sources) in Excel. This pops up a window where I can select the dataset of interest. Alternatively, I can open My Documents\My Data Sources\+Connect to New Data Source.odc in Excel, create an ODC file, and accomplish the same thing. I've gotten this to work with both a local server or SAS/Share server (running on my local machine).

Problem: My colleagues will want to do this without creating any DSNs on their machines. To test this, when I copied the DSN file to a network drive and accessed them from another computer, it failed. Perhaps that's why they are called "(non sharable)"? If I try to create a DSN under the File DSN tab, I cannot create settings in the Servers and Libraries tab of the SAS ODBC dialog.

I have read the SAS ODBC Driver: User's Guide and Programmer's Reference doc.

Questions:

1. Can this even be done, or are local DSN definitions required for every user that needs to access the SAS data?

2. What's the difference between a DSN and ODC file? They seem to accomplish the same thing.

3. How can I create a File DSN where I can specify the Server and Libraries during definition?

4. The ODC creation process (using Microsoft OLE DB provider for ODBC Drivers), in the Data Link Properties Window, Connection tab, has the Connection String field. Can someone provide a typical connection string without using a DSN?

5. Bonus: we have a SAS/Share server running on a Unix box (on the other side of the world). It would be cool to connect to that SAS/Share server and download the data from there into Excel. What do I need to do to connect to that server? Do I need to know the port it's listening on? Do I need to edit my local services file? Etc.

Thanks, Scott