Oh I don't know, I remember plenty of people on usenet who would quote a fragment--deliberately eliminating relevant context--to create a straw man that they could tear into. Some people just like to argue, and the only way to deal with them in an unmoderated forum is to ignore them. Kill files were useful for persistent offenders.
Of course this did also happen on usenet. Call me old but a good flamewar beats a good shitstorm any time (and the names might even be an indicator why).
But one design element of twitter might make it easier to stumble into the realm of people who will wildly disagree with a poster's opinion.
Just replying to some tweet you can not know what kind of domain you are entering. With usenet groups the name has always been a strong indicator what people you will encounter.
If you post "Ruby sucks and is dead slow, Perl rulez!!1eleven" into comp.lang.ruby, you shouldn't be surprised if you get flamed. With twitter accounts, this is not that obvious.
Of course, this is a core design element of twitter that it is concerned with accounts and not thematical groups and both have their advantages and disadvantages but I think this is one of the reason that you can much more easily and unintentionally anger the wrong people on twitter.
