Thanks for the replys.
The files are only in OnDemand for storage and will be retrieved by a user process that will open the file to determine the contents, and then use the data to create attachments for reports. Each indivdual file can contain different types of data and we will not know what the types are, they can have mime type data, XML strings or mixed data, in other words it will be blobs of data. The data will be fully indexed and will not be compressed when received, we will compress at load time.
I just need to make sure that loading it as NONE it can be easily opened by the retrieveing process.
Thanks,
Russell