Recently, I've been needing to get data out of minidumps on several occasions, and I found it unwelcoming. First, and this should be an entirely separate bug, when you're on the "Raw dump" tab and you're not logged in, you obviously can't download the raw dump. But after you login, you get back to the details tab, not the "Raw dump" tab. Once you're logged in, you get two links: one for a .dmp, one for a .json. The .json looks very much like the data already in details and metadata tabs, so it's questionable why the link requires being logged in in the first place. And the minidump is... well, to do anything with it, you need tools, and most people don't have them. So you have to either download them from somewhere (where?) or download the breakpad source and build them. The most commonly used tool, I think, is minidump_dump, because it displays all the data in the minidump, include address space mappings or stack contents that don't appear in the raw dump json in the iframe. A link to the output of minidump_dump would be, IMHO, very valuable. Or some sort of json like the one in the iframe, but with every single bit of data from the minidump. (The minidump, though, can still be useful because it can be used with debuggers)
See also: bug 977778, bug 818069. FYI the .json is behind auth because those are the raw annotations submitted by the client, so that file will contain any sensitive keys like email and URL. We do have some docs on getting useful info out, but it's always easier to get things out of the web interface of course: https://developer.mozilla.org/en-US/docs/Mozilla/Debugging/Debugging_a_minidump#Using_other_tools_to_inspect_minidump_data
Ted, Just like we have on the Symbols page, there's some instructions and sample (curl and python) code to get your started. We could use the space underneath there on report/index to either show what you can do/need to do with these files to get started. Or just a link to that MDN page perhaps.