The Social

Surprise! A full 40.5 percent of posts on Twitter--or tweets, as they're called--can be classified as "pointless babble," according to a new study from Pear Analytics. Coming in second was "conversational," which the company says makes up 37.55 of all tweets.

Pear Analytics published its investigation, which was conducted through a series of random samplings from the Twitter public timeline, into the different species of tweets on Wednesday. That means that only public tweets were indexed; the numbers could be different if friends-only accounts were taken into consideration as well. (Obviously, that would be much tougher to analyze.)

There's some interesting stuff in there. Despite some Twitter critics' insistence that the microblogging service is loaded with self-promoters, Pear Analytics only classified 5.85 percent of tweets as "self promotion."

The other categories were "news" (3.6 percent), "spam" (also lower than I'd expect, at 3.75 percent), and "pass-along value" (8.7 percent). Granted, sometimes there's plenty of gray area (is linking to a blog post you wrote "pass-along value" or "self-promotion"? shouldn't tweeting about breakfast too often be considered spam?) but it's pretty cool regardless.

"We thought the news category would have more weight than dead last," the report read, "since this seems to be contrary to Twitter's new position of being the new source of news and events."

That might be a bit of a buzzkill for Twitter's team, which is pretty vocal about wanting the service to be a ubiquitous communication standard. Regardless, the news about the relatively low levels of spam is interesting--for some perspective, about 90 percent of e-mail is spam.

On Sunday, I had an e-mail alert about someone writing on my Facebook wall--a college acquaintance with whom I hadn't spoken in quite some time. As it turns out, I was a victim of "wall spam," a recent phenomenon on Facebook in which automated spam posts show up on members' message walls. It's similar to a wave of profile spam that swept News Corp.'s MySpace a few years ago.

The message in question read, "Some thinks you are special and has a hot^crush on you. Find out who it could be!! ;)" with a link to a Flash file claiming to be hosted on the imageshack.us domain.

But by the time I navigated to my Facebook profile to get rid of the spammy (and possibly virus-ridden) message--within an hour or two of the notification showing up in the first place--the wall post was gone. This means one of either two things: someone else saw the message on my profile and flagged it, or Facebook is actively policing the site to keep it under control, probably by searching for duplicates of a known spam message.

Of course, an hour or two is still a big enough frame of time for people to click on the link and get their computers loaded with some nasty new malware.

I've asked Facebook for comment on exactly what their strategy is and whether any members' login credentials are getting compromised by this spam or virus. I'll update when I hear back.

"Wall spam" rose to notoriety earlier this month, when members started noticing the phenomenon, and security firms started flagging worms that were spreading via Facebook members' walls and installing malware when a link in the message was clicked. The company has recommended antivirus fixes and says it's acting fast.

The Silicon Alley Insider reported earlier this month that Facebook had been deactivating links in identified spam posts; removing the posts entirely is a more aggressive measure.

"If we get a report of a bug or a hole from a user, a security researcher, a reporter, blogger, or anyone, we check it out and fix it as quickly as possible," Facebook security chair Max Kelly wrote several weeks ago on the company blog in response to another virus. "In fact, we appreciate it when help comes our way from the many security experts and organizations out there."

Did you know that you can only follow 2,000 people on Twitter--unless there are at least 2,000 people who have opted to follow you?

This was one of the measures that the microblogging service formally announced Thursday as part of a new system to cut down on spam. The company acknowledged it only obliquely, but bloggers like David Risley picked up on the news and spread the word.

Twitter spam accounts are known for adding thousands of followers and then hoping that some of the unwitting Twitter users will follow the spam account in return--most don't, meaning that spam accounts tend to have a disproportionately low number of followers in contrast to the number of people they've added. But extremely popular Twitter accounts, from Web celebs like Jason Calacanis to the Twitter feeds for news outlets like CNN and political campaigns like Barack Obama's, Twitter still allows the adding of more than 2,000 followers. The rationale is that if people are willing to add them back, they probably aren't spam.

Risley suggested that Twitter could offer paid accounts to raise the limit, which could be a viable first step for a service that still has not put a business model in place.

There are very few rules on microblogging platform Twitter. But if you use it for unsolicited "tweets" about male enhancement products, watch out: Twitter has started to shut down accounts that it has flagged as "spam," reported blogger Jesse Stay.

Twitter employees confirmed the new tactic in their developer forum. "We've been considering this issue here at Twitter HQ, and we're planning on simply removing the accounts of users who have violated our Terms of Service, as opposed to freezing their account as we've done in the past," Twitter engineer Alex Payne explained.

Previously, "frozen" Twitter accounts were not removed, but the owners could not add any other users as followers. They also weren't notified of the freeze. Now, owners of accounts that have been flagged as spam will be alerted in advance and will have a chance to make a case for themselves.

So you'll be able to explain to the Twitterati, however unsuccessfully, that you really do just like to talk about Viagra and cheap Rolex watches all the time.

This post was updated at 4:49 PM PT with a clarification from Matt Mullenweg.

MIAMI--"I'm Matt Mullenweg, and I'm famous for eating 108 Chicken McNuggets and surviving," the eccentric 24-year-old WordPress founder said in his talk at the Future of Web Apps conference, explaining that he's no longer continually the No. 1 "Matt" in a Google search because the dancing viral-video star "Where The Hell Is Matt?" gives him a run for his money.

At FOWA, Mullenweg was slated to talk about both the physical and psychological "architecture" of WordPress, which has gained both positive buzz and popularity for being simply constructed, easy to use, and remarkably efficient.

"Scale is what separates us from the other industries of the world," he explained, saying that it's only in the technology business that a tiny entrepreneurial team can create something used by millions of people. WordPress, Mullenweg said, powers 2,523,000 blogs, gets 135 million global unique visitors, and has only 19 full-time employees.

Matt Mullenweg

(Credit: Wordpress)

"All these old-media companies are adding blogs like it's going out of style," he said, talking about how WordPress now powers blogs for The New York Times, CNN, and Fox News ("unfortunately," he added on that last one).

Mullenweg added later in a conversation that he didn't intend "unfortunately" to sound the way it did and that he meant no offense to WordPress' major media clients.

He had quite a bit of advice for the audience. "Be the person in the support forums who's answering everybody's questions," Mullenweg advised start-up entrepreneurs in the room. If you don't look like you're hard-core about your company and its users, he said, you won't build up a following.

It was a pretty geek speak-intensive talk, with Mullenweg explaining to the developer-filled concert hall how WordPress handles server and bandwidth demands, and how to take advantage of systems like Memcached, which was originally developed for social-media pioneer LiveJournal. But he also expounded a bit on the Web 2.0 landscape and some of the issues it faces--like spam, the ugly side of the open-social Web. WordPress has deleted more than 800,000 "splogs," or spam blogs, for example.

Spammers are "the terrorists of Web 2.0," Mullenweg said. "They come into our communities and take advantage of our openness." He suggested that people may have moved away from e-mail and toward messaging systems like Facebook messaging and Twitter to get away from spam. But with all those "zombie bites" showing up in his Facebook in-box, he explained, the spammers are pouncing on openness once again.

He also has a pretty nontraditional view of ad revenues, the supposed cash coffer of new-media sites. "Most of you have never, and will never, seen an ad on WordPress.com," Mullenweg said, referring to WordPress.org's free blog-hosting arm. "We decided to show ads only on certain pages, only to the people who were sort of random drive-by visitors...if you use Firefox, you'll never see an ad, no matter what, mostly because I like Firefox."

Ever come across one of those Facebook Platform applications that required you to spam a dozen of your friends with invites before you could access the results of your "Vampire Jedi Zombie Personality Quiz"?

They're annoying. And now Facebook has done something about it. Developer applications must "offer some navigation option to leave the friend invite process," according to a change in the social-networking site's platform policy. If an application's friend-invite page doesn't contain one of Facebook's in-house "Skip This Step," "Cancel," or "Skip" buttons, it has to contain an alternative way to navigate away from the friend invite process.

Developers whose applications ignore the new regulations reportedly receive warning letters that threaten shutdown if they fail to comply.

This is a big step toward cleaning up the cluttered Facebook app directory. Ideally, it will cut down on some of the "app invite overkill" that's led many Facebook users to groan every time they're invited to the corny application du jour. And it'll likely mean that Facebook members will probably only be passing on invites to applications they actually like, rather than spamming their friends just so they can learn what character from Hannah Montana they most resemble.

It will also mean your friends won't have to know that you even installed such an abhorrent application.

Despite the obligatory missing vowel, bacn (pronounced "bacon") isn't a hot Web 2.0 start-up. It's "the middle class of e-mail," the stuff that isn't really spam because it's not totally unwanted, but isn't really wanted either. Case in point: Pownce messages, Facebook friend requests, Amazon "recommendations."

Unlike many dorky tech terms, the origins of bacn aren't especially apocryphal; we've got a real (electronic) paper trail. The term arose during a discussion at Podcamp2 Pittsburgh earlier in August and slipped onto my radar via Twitter feeds from friends who were attending that conference--Fearless Cooking video blogger Grace Piper, for example, who clarified that "steak" is e-mail you always want to read. Fellow video blog personality Bill Cammack added that "FakinBacn" would refer to e-mail that's really spam but attempts to gussy itself up in the guise of bacn. Those video podcasters are a clever crowd.

It wasn't until a conversation with digital marketing strategist Rachel Clarke at last night's first-anniversary party for gadget blog CrunchGear when it occurred to me that bacn was deserving of a spot in the lexicon of trendy tech terms. Unfortunately, BuzzFeed had already beaten me to the punch. Ouch.

So what do you think? Will this one make it to the dictionary or will it remain restricted to tongue-in-cheek use in geek circles?