People who have been around Enterprise Vault for quite a while will have heard of the Enterprise Vault Converters. In this post I'll explain a little bit about the converters, what they do, and why they're needed in a product like Enterprise Vault.
A long time ago, in a far away place, the converters were also known as Outside In Converters. I *think* (but a Bing search just now didn't help) that it was a separate company that was eventually purchased by Oracle. Oracle is certainly the home now:
http://www.oracle.com/technetwork/middleware/content-management/oit-all-085236.html
Essentially what the converters do is to take content like Microsoft Word documents, PDF files and so on, and convert them to HTML (or text). For a product like Enterprise Vault, and it's archiving activities, what this means is that Enterprise Vault can 'obtain' from a message, or file system, or SharePoint, etc, a file, pass it to the converters and get back either HTML or text parts/equivalent of the file. This can then be added to the Enterprise Vault index, for our friendly users to search against. OIT can convert about 600 different file formats, according to the Oracle web site.
From time to time there are updates to the converters. This usually happens from Enterprise Vault version to version, sometimes from service pack to service pack, and *very* occasionally more 'needy' updates like this one:
http://www.symantec.com/docs/TECH167455
You normally see the converters running as a process (or a number of processes) called EVConverterSandbox. Take a look on your Enterprise Vault server, during an archiving run, and you'll see them busily converting content for your items.