[guice]

I agree, ball is in WM's court. And if they chose, they can fight. If they choose to fight, WoWI will surely loose. You can't win "banning" on the internet. That's what makes the internet so great! Everything goes. No boundaries, no rules, free reign. That's why the internet has grown so fast so quickly. WoWI set itself up for failure the moment is dropped the gauntlet and attempted to ban WM. "Force" is the word used. I don't believe in that. Other routes, now apparent, could have been taken. They were just unknown until recent discussions.

As for bandwidth; it's two fold here. The old days, it was a bunch of text files through SVN co's. People went directly to the svn they wanted and exported everything out; all through text medium. Text is very low bandwidth. Then you have WAU which essentially took svn snapshots and bundled them into a single download package. This convenience factor did two things: it increased bandwidth from scanning -- something that can be removed if the APIs were just there. It also increased bandwidth from people downloading *more* add-ons. When it's easier to update 30 add-ons, people tend to download 30+ add-ons.

WM did the same thing; but this time it was from multiple sources and this gave a wealth of options for the user. So what does the user do? Download even MORE add-ons! That, of course, causes even more bandwidth usage. And, natually, without an API, WM is force to website scrape.

If you think it over, you can see the bandwidth used wouldn't be that much more. It seems like a lot because of a few factors: a lot of users, and WM is doing a full scrape, short periods of time.
Now, add in this factor: if a user did all this by themselves, they'll be download every webpage they get through each link; ever JS file; ever image; every offsite reference. In reality, the two should be equal if not WM actually *saving* bandwidth. Saving simply because WM is dealing with text-only and not download images, js or css files in it's page scraping.