There are basically 2 responses:

1. This is not a problem, and if it is we throw money at it
2. This is a solved problem because we know how to do HTTP

As for #2, I just sent mail to Bill Venners at Artima about this, because the Artima crawler is pounding my feed. That's because it's not doing ETag, If-Modified-Since, or gzip. I know that when I bugged Brent Simmons about getting Gzip support into NetNewsWire he told me that he wasn't finding many feeds that were compressed. Just because we know how to do HTTP doesn't mean we actually are doing HTTP.

One post also said that the success of RSS is due to using HTTP. That may be true, but I don't see that as the major point. The reason weblogging is successful in my eyes is that I no longer have to manually navigate to a zillion sites, and then try to figure out what I have and have not read. It has nothing to do with the protocol or RSS (although I am aware that they are necessary to enable the end user benefit).

Right now weblogs are a niche phenomenon. They have incredible reach, but the number of deployed aggregators is far less than the number of deployed browsers. I believe this is going to change. I believe we are going to see some uses of e-mail replaced by RSS. The bandwith usage is only going to increase. Its not just raw bandwidth usage either, it's the time it takes to download the feed as well.