The core of the PPCA is three interdependent components: tough regulations for health insurance providers, an individual mandate, and subsidies to lower-income individuals. By conceding that the regulations have to stay, Republicans are essentially conceding the
entire framework of the bill.

As Paul Krugman has pointed out, PPCA is a balancing act. If you aren’t going to repeal the new regulations, there’s no way insurers will allow the repeal of the mandate. If the mandate is repealed but the regulations are left, many people won’t buy insurance at all, and will wait until they get sick to buy
coverage. This will drive the insurers out of business. So, unless Republicans are looking to go to war with the business community, they’ll keep the mandate intact.

At this point, Republicans have already conceded two-thirds of the structure of the bill. The one component we haven’t discussed yet is the subsidies, and how to pay for them, but last time Republicans were unopposed in power, they weren’t able to make a dent in spending on
social safety net programs. The best they were able to do is keep spending the same and shift the funding burden from taxes to deficit spending. Doing this to PPCA would be unhealthy (see also: the squandering of the Clinton surplus), but wouldn’t come remotely close to undermining the bill as a whole. As far as I can tell, once a Republican concedes the new regulations, they concede 90% of PPCA is here to stay.

I’m starting a count of what I’ll call the “Cornyn bluffers”, the Republicans that bluster about repealing health care reform while implicitly admitting that they will, at most, tinker around the edges. John Cornyn was the charter member of this coalition, and Matalin joined last night. Who will be next?

There are two broad categories of issues here: problems that arise from the “legitimate” use of the service in unwise ways (generally privacy issues) and people taking illegitimate or unauthorized action through the service (generally security issues)

Some issues that can arise with people just posting to Twitter:

1. Spam.
2. It’s very easy to trick somebody to visiting a URL you control via Twitter, due to the use of link shorteners, and the culture of sharing links. Tricking people to visit sites they don’t know the provenance of is a frequent first step in attacks such as phishing or cross-site request forgery (XSRF).
3. Impersonation (especially of celebrities)
4. People inadvertently sharing too much information about themselves (leading to employers, family members, or criminals finding out more than they should).
5. Other members of one’s social graph sharing more than they should about you (the Facebook equivalent of someone tagging you in photos you don’t want to be associated with).
6. People being able to infer more than they should from your social graph alone (from my graph, you could guess where I live, where I work, which groups I’ve worked in at my employer, where I went to college, for starters…)

Twitter also has a number of security issues that almost all web applications are vulnerable to:

1. As you mentioned, ‘phishing’ attacks which aim to steal people’s passwords. These are compounded by the fact that people tend to re-use passwords across services, so even if a non-Twitter password is compromised, this can often be leveraged into access to Twitter accounts.
2. Worms use Facebook as a method of transmitting (send a copy of yourself to all your friends). Worms in the future will certainly broadcast themselves via Twitter. (ed: and probably already do.)
3. Twitter has to be careful about cross-site scripting (XSS) and cross-site request forgery (XSRF), which are essentially ways for an attacker to take over your browser by posting malicious content to a third-party web service you’re viewing, such as Twitter.