Some design considerations for Xena 3.0:
The Table Of Contents is a known problem, and funnily enough fixing the TOC became the basis of our very major Xena 3.0 upgrade. The TOC right now is a static toc.xml file that is around 800KB, which takes way too long to render for modem users. My spec for the new TOC will be a maximum 5 second render time over a 56K modem - I haven't seen the technical spec for how this will be implemented, but I assume that they'll chunk layers of the TOC and pre-cache in some way.
With respect to cross-browser compatibility, this is also a priority 1 feature for the new site. This has been a hotbutton issue for me for some time, and my requirement is to have broad cross-browser compatibility; I believe we'll test on IE, Mozilla, and Opera.
Now, about File Transfer Manager: FTM is a client/server application that provides robust download, encryption, prioritization, geo load balancing, and reporting. Because we're dealing with pretty big (and growing) file sizes, FTM allows us to post a 4.7GB Visual Studio DVD image with very high confidence that it is being downloaded correctly. We are constantly making improvements, and in the Xena 3.0 timeframe I believe we'll be working on multi-threading the client to improve download speeds further. However, since we maintain the data centers (as opposed to using an edge network like Akamai) for Xena, a user who's 1000 miles from the nearest Xena server is going to have some latency issues. So, in a broad sense we'll be working on performance, but I don't anticipate using anything like BitTorrent or some edge routing to deliver the several terabytes of content that we need to securely and reliably publish.