If the virus has a highly variable DNA sequence, different RNAi sequences will need to be designed, validated, and grown up in lab for each viral strain. Viral strains will need to be identified upon infection so the right RNAi sequences are used, which can increase time until treatment. We won't be seeing this technique used on a highly variable virus like HIV or the common cold anytime soon, because these viruses mutate incredibly quickly within the host.
And while yes, RNAi has been around for a while, nothing like this has been done with the technology. This is a breakthrough.
http://bytesizebio.net/index.php/2010/05/29/a-cure-for-ebola...
The key for me is this part
"this is the first time that siRNA treatment has been shown to work in primate models of a human disease."
That's a big deal under any circumstances.
The investigators plan to do safety tests in humans next, but they have some preliminary data on safety this is positive.
This is a breakthrough not only because it may give us a cure for an uncurable, incredibly nasty virus, but also because the same method might work for other viruses, and because we have woefully few effective antiviral treatments.
The exciting thing is not that we have a potential cure for a rare[1], exotic[2] virus. The exciting thing is the technique and its potential applicability to other viruses.
[1]http://en.wikipedia.org/wiki/List_of_Ebola_outbreaks
[2] Part of the reason they're probably using Ebola is to get DoD funding (ie in the anti-biowarfare space)