The Decline of Internet Anonymity?
I professed my love for 750words.com a few posts back but found something else pretty interesting about the site: you can only log on with 3rd party accounts. There's no native user account system; you're forced to use Facebook Connect, Google Friend Connect or Yahoo!ID.

This is the first time I've ever seen a web app configured this way and it has both its upsides and downsides:
The pros: Less code to maintain and one less username/password for users to remember
The cons: 3rd party dependency/limitations and a lack of user anonymity
The most interesting part is the lack of user anonymity, that is, forcing users to use their real names or ID's (though people can still create fake handles at those sites). It's probably not a huge concern for most 750words users since the application is not heavily community-centric (by default, your posts are private), but I wonder...how would this approach affect other sites and apps?

Here's the dilemma: for community sites, there's "an inverse relationship between the level of anonymity and quality of conversation". But at the same time, when you disallow anonymity, people are less likely to participate, meaning you might not have a conversation anyway.
Both approaches come with consequences. But with the growing popularity of Facebook, Google and other sites that let users take their accounts elsewhere, maybe the trade-off is worth it.
For community/social sites where real names are the norm (e.g., LinkedIn), why create a native user account system when most users already use those other sites? And for sites that advocate dialog and conversation, could losing anonymous posters be written off as collateral damage if the health of the community's conversation has an overall upswing?
There will always be a place for anonymity because some sites--especially those related to anything taboo--just can't survive without it. However, I wouldn't be surprised if the number of sites without their own user account systems continue to increase in the future.