Twitter will just block it. Scraping should be done from multiple random addresses from the networks all around the globe, at random intervals. I guess, the onl

Twitter will just block it. Scraping should be done from multiple random addresses from the networks all around the globe, at random intervals. I guess, the only case is voluntary botnet, where users sympathetic to the cause would run limited (so, not prone to abuse) proxies. (To avoid users messing with tweets, just proxy TCP connections to the mothership and let it verify TLS.) However, this would hinder the service, as with the polling model it may not always notice all the tweets.