website statistics

Archive for the 'Spam' Category

Expecting the Spectator

I don’t know why, but since last night, the American Spectator’s website at spectator.org has been blocked for being a “reported attack site”:

American Spectator website blocked as “attack site”

Fortunately, perhaps, Google provides diagnostic tools for those curious about where the site has gone:

Google diagnostics on the Spectator as an “attack site”

Alas, I don’t know enough about network security to make a diagnosis. (Dammit Jim, I’m a private eye, not a doctor.)

As of this morning, I can get the website to load in Safari but not in Firefox 3, albeit intermittently. The front page is accessible, but when I try to visit the blog, I get this instead:

American Spectator will cause “harm” to your computer

In the past, Google has been accused of removing conservative-aligned content from YouTube and from Google News, but I see no evidence that this is what’s happened this time. I’m not even quite sure why Google is responsible for making this call or providing these diagnostics.

What’s most likely is the Spectator’s webmaster left a security hole unplugged and the site was taken advantage of by opportunistic spammers, which is something of a tautology.

I’ve put an e-mail in to a contact at the Spectator, and if I find out what happened, I’ll provide an update in this post.

Update: Looks like I called it. The site still isn’t working for me in Firefox, but via Safari, they offer this explanation:

We have received a number of inquiries regarding the fact that Spectator.org has been designated a “harmful site” by Google, because of outside entities attempting to use our site to distribute malicious software. We have been working with our Web hosting company to address the issue, and believe that it has been resolved and that our site is safe to visit, though there is a lag time before Google can remove the “harmful site” status. In the meantime, if you normally find us via Google, you can still visit us by typing Spectator.org directly into your browser, or by entering our site via Yahoo. Thank you for your understanding.

Could Twitter Ads Help Stop Twitter Spam?

Twitter spam is back on my mind as I think about this morning’s TechCrunch report that Robert Scoble, the #5 most-followed Twitter user, has started tweeting paid advertisements. TechCrunch is shocked, shocked! to find out there’s advertising happening on Twitter, and alludes to speculation that Twitter’s founders will renege on a longstanding promise never to put ads on the Twitter website.

Some of this is driven by the fact that Twitter.jp, the Japanese-language counterpart, launched with advertising last month. Why ads on the Japanese version, but not the English? The conventional wisdom is that it’s harder to put advertising on the site later. That may be true, but of course, we already know that advertising happens on Twitter: many if not most of the accounts listed on TwitterBlacklist.com are primarily commercial in nature.

Which leads me to wonder, could Twitter ads be a partial solution to the problem of Twitter spam? After all, what these people are trying to do is reach many more people than their actual level of notability would attract. In lieu of other options, they’ve followed many more accounts than they could actually read, often using a bot to follow accounts automatically. How many of them would be willing to pay a small amount to place advertisements in the blank space underneath users’ left-hand sidebar? My guess is quite a few. In fact, so would quite a few others not presently engaging in spam-related activity.

One requirement for these ads could be that they must link to a Twitter account, which could then link out to where ever the advertiser wished. According to Valleywag, Twitter.jp ads do this, and it sounds to me like a fine way to keep the advertising conversational, like Twitter is meant to be. You know what isn’t conversational? A self-help guru whose promotions-only account follows 18,265 others with only 472 reciprocal followers.

Twitter advertising of this type would create an alternative to annoying other users with unwanted follower notifications while putting Twitter’s parent company Obvious on the slow road to profitability. Biz, Ev and Jack say they’ve been looking for a business model. Why not this one?

Everything in Moderation: A Closer Look at Comment Spam

At my ever more occasionally updated personal blog, I’ve long published a series of posts called “Great Spams of the Internet” wherein I highlight a particularly amusing bit of e-mail spam and even the occasional e-mail interaction. Once when a 419 scammer tried to get me to call him on the telephone, I replied:

Regrettably, I was born with no mouth.

He was very understanding, writing back the next day:

thank you sir thank for your mail all is understood well i can question you just of the condition you gave any please kindly make a way we can both talk

At least I think he understood. In any case, this is the long way around getting to my real point.

As you may know, I run a blog here. As you can probably guess, I get my share of spam comments; most are caught by the Akismet plug-in for WordPress. But then, most are fully automated and advertise prescription drugs, gambling websites or sex acts that would probably boost my unique visitor counts if I mentioned them, but I don’t need that kind of traffic.

However, a small percentage of it manages to evade Akismet’s filters and find its way into my moderation queue. In some cases, they are only barely distinguishable from real comments. In some cases not listed here, I’ve approved comments that I am sure were intended only to improve the SEO of the website linked, but were interesting enough to allow through on their own merits.

Most are not, but this doesn’t mean they’re entirely without value. Some of them are clever, some are just amusing. I’ve been holding onto a few of them to discuss here, so let’s open up the queue, if for no other reason than now I can finally delete them:

Example spam comment received at Blog P.I.

Here, somebody is pushing what appears to be a YouTube clone, even using a joking nickname YouTube acquired once the site itself was acquired by Google. In fact, the site turns out to be a combination of Google’s input forms. Though the IP address indeed traces back to the United Kingdom, the author is not especially concerned with proper English spelling or punctuation. They also have no system for keeping track of which websites they have already hit, or they just don’t care. I’m leaning toward the latter.

Example spam comment received at Blog P.I.

Here is one that, at first glance, looks like a genuine comment: This was intended for a post that mentioned Ron Paul, just as the one above tried attaching itself to a post discussing Google and YouTube. But if you follow the link, it goes to a blog whose posts consist of only of one YouTube video and sometimes-relevant text copied from other websites — “scraped” as it’s called. And there’s a good reason why it sounds like a real comment: It was scraped from another comment from the same thread.

Example spam comment received at Blog P.I.

This one promotes yet another inscrutable blog, this time in a foreign language that I presume to be Turkish. I guess this because the IP address resolves to Izmir, Turkey. The one above resolves to Istanbul, Turkey. The two cities are not close by, so they are probably not the same person. But if Turkey is a hotbed of comment spam, that’s news to me.

Example spam comment received at Blog P.I.

Undoubtedly, this one is my favorite. Like the Wikipedia vandal whose edit summary consisted of “Blanked the page” or the panhandler who admits he needs the money for booze, “Sohbet” is admirably honest about his intentions. I might even consider throwing him a link, except that the website no longer exists — less than a month after he was trying to extract Google juice/build traffic for it. Also of note: the IP address resolves to Antalya, Turkey. Still, if Turkish comment spam is a known phenomenon, I can’t find any discussion about it.

Example spam comment received at Blog P.I.

Funny at first, but tedious. I get a lot of these, and it’s kind of similar to another common tactic I’ll get to in just a bit. Flattery will get you everywhere with some people, but not me. Also, the linked site is in Russian. Russian spam at least I am familiar with.

Example spam comment received at Blog P.I.

Better than YouTube! Quite a claim. Surprisingly, the website is well-designed, coherent and legitimate. For someone who just wanted to find videos related to a presidential or prospective VP candidate, it might actually be better than YouTube. So here we can start to draw a clear distinction: Some spam comment campaigns aim to promote fake websites that seek ad revenue or to promote another website. Others are spammy promotions for real websites; it’s very possible the creators of this website don’t know exactly what their SEO is up to. But I’m not particularly offended by this comment. It doesn’t add to the conversation so I won’t approve it, but it got the general subject matter of this website correct, it’s vaguely conversational, and it doesn’t represent itself as anything other than what it is: a pitch.

Example spam comment received at Blog P.I.

Lastly, this one I’m including not because it’s compelling, but because it’s so common. Also, because it represents the dishonest counterpoint to the previous example. Here, the commenter announces enthusiasm for the targeted website (in this case mine), then immediately starts pitching another website. Notice that his subject matter is completely off-base with what Blog P.I. is about. The targeted post — which I wrote in July, 2006 — included exactly one use of the word “wedding,” in a throwaway reference to New York Times announcements page thereof.

Predictably, the website being promoted is commercial in nature, but doesn’t offer anything for sale itself. What it does, though, is link to pages on a real wedding supply website, which presumably hired the spammer to boost their search engine ranking. A bit of rudimentary sleuthing reveals the SEO’s identity and company; he’s using his real name (which is something, I guess) and he didn’t even register the URL anonymously.

But I’m not going to single him out with a link or textual mention that could turn up in a search engine. He’s not doing anything illegal and, as noted above, similar practices are exceedingly common. I’ve been a critic of certain SEO practices, but I’m fascinated by also them, and clearly I think some tactics are better than others. The way I see it, if you’re going to do black hat SEO, why not do it with some style?

Also, the joke is on them: Every link in my comment section is automatically assigned a nofollow attribute.

No Follow

Not to turn this into The Twitter Spam Post (besides, that’s Stop Twitter Spam) but I believe I’ve just discovered a new spammer technique.

At 3:42 a.m. last night, I received a notification that a Twitter user going by the name Cardiophile was following me. Twitter Spammer: CardiophileBut when I checked out the account this morning, the sidebar looked as it does in the graphic to the right of these words. For readers who don’t know me, I am not one of the 5 accounts being followed.

So, mark this as the logical next step in the growing sophistication of Twitter spammers. Aware that they’re being identified by the obvious disparity in their following/follower counts, they’re now following an account just long enough — seconds, maybe — to send a notification e-mail and then unfollowing, so there isn’t the dead giveaway. I caught on, but then there are 350-some Twitter users who followed it anyway.

Earlier this week, a Twitter friend suggested that I simply uncheck the e-mail notification option. True, this would keep me from being annoyed. But there are two problem here. First, I would prefer to avoid changing my behavior because of spammers. But more practically, I wouldn’t know about followers I do care anbout and would want to follow back. Follow me?

Update: Stop Twitter Spam has also noticed this new technique, and has posted on the same subject.

Twitter Spam Gets Political

Last week, Mashable’s Adam Ostrow asked whether Twitter was facing a spam problem. I said it already does. Ostrow pointed to a Twitter account that seemed to be following far more people than anyone could know, and for purely promotional purposes.

As of today, that account follows (i.e. has friended) 13,000+ Twitterinos, only to tweet links to images of ho-hum abstract artwork. Amazingly, more than 800 people are still following this account. Shortly before reading his post, I found a teenager in Norway who seemed to be doing something similar. While he may in fact be using the service genuinely, he too was following thousands before he’d posted a single tweet. Today he’s following some 3,700+ others, but hasn’t updated for two days, when he was:

watching random crap on youtube :) — Flyaxe on Twitter

Now Twitter spam has taken a turn for the political. On Sunday, CQ’s Eric Pfeiffer told me that his account (which he updates only sporadically) had recently been followed by a horde of obviously fake accounts named for a current or former presidential candidate, plus a number. Most of the notification e-mails he had already deleted, but the others he forwarded to me. They are… interesting. For example, here’s the latest tweet from ChrisDodd53:

Twitter spammer rips off techPresident Daily Digest

I think we can safely assume there are not 52 other Chris Dodds on Twitter. But did you recognize the content of that tweet? I sure did: it was scraped from today’s techPresident Daily Digest. And this pattern is repeated across all the examples of spam accounts he sent my way.

Herewith, a list of these accounts, and a link to the blog whence its latest tweet was scraped:

You know, if these were simply attached to RSS feeds and genuine aggregators of political news, I wouldn’t mind so much. Yes, the aggressive, untargeted following is certainly annoying. But these accounts do not drive traffic to the sites where the words originated. This also makes the creator’s intent all the more inscrutable; they aren’t saying anything, they aren’t promoting anything, and they aren’t updated by hand. The only thing it’s good for, maybe, is souring users on Twitter. But I don’t believe Pownce or Google/Jaiku are really that underhanded. So I remain mystified.

Whatever the cause, this must stop. And it can. Unlike e-mail, which is traded from network to network across yon Internets around the globe, Twitter is administered entirely by Obvious, LLC. They have the same control over the Twitter network as Facebook has over its pages, and it’s within their power to stop it. I’ve previously suggested capping the number of users you can follow, relative to the number of users who are following you. Nothing too restrictive, but something flexible to keep Twitter accounts honest. Jack? Biz? Ev? Little help over here?

In the meantime, there is is already a website carrying the banner against this annoying menace. That’s Stop Twitter Spam, which is currently tracking complaints about spam on Twitter, including my post from last week. The site only barely gets into solutions, and mostly serves to highlight the problem. Most interesting of all is the Twitter Spammer List.

This list includes most of the candidate-based accounts I’ve noted here, and some others I hadn’t. It also mentions the examples from last week, but separates them into two apparent categories of problem Twitter accounts: outright spam and overactive followers. It also notes the number of follows vs. the number of followers and shows the difference as a ratio. The greatest disparity is HillaryClinton5. When the list was last updated, the account followed 2905 others, yet only 25 others followed “her” back. It’s not that HillaryClinton5 has friended the most people — that’s probably the design/art company mentioned above — but that she has the fewest followers. These numbers are a few days out of date, but still give a useful snapshot of the problem.

I’ll admit, I’m a bit nostalgic for the days when Brian Shaler was just following everybody in sight, like it was a game. But then, Shaler is an honest Twitterino. And almost everyone he follows also follows him back.

Twitter Already Has a Spam Problem

Yesterday, Adam Ostrow at Mashable asked, “Is Twitter About to Have a Big Spam Problem?” Well, I wouldn’t yet call it “big,” but the problem is already here. Ostrow wrote:

[L]ately, I’ve been getting an influx of new followers that resemble this character to the right – someone who is following thousands of people, with only a couple hundred following back. In this case, the new follower seems to be a web design studio in Beverly Hills. While I can’t prove it, I have a feeling that this person used a bot to automatically follow me (and a lot of other people) in an effort to take advantage of the fact that a lot of people will simply return the follow – in turn giving this person a new platform to pump their marketing message.

Sidebar to Flyaxe, a suspicious account on TwitterI knew the account he spoke of; I am one of those also being followed by the “Tripix Designs” Twitter account he mentioned. Like Ostrow, I’ve been followed by a handful of these accounts. Aside from inflating my follower count, I didn’t consider it a problem. But this morning I’m convinced.

At right is the sidebar for “Flyaxe” — a Twitter account that added me sometime last night. That’s what it looked like at about 6:30 this morning, Eastern time. Just a couple hours later, Flyaxe is following more than twice as many. Unlike Tripix, it hasn’t even updated once, so it isn’t clearly promotional. Flyaxe appears to be a “19 year-old dude from norway,” as the matching, recent and similarly empty Digg account shows. At least Tripix was honest about its intentions; Flyaxe could be a Trojan horse for just about anything.

The only solution is for Obvious (the under-funded Twitter-makers) to impose restrictions on Twitter accounts. Facebook imposes all kinds of restrictions on its users, and the result is a better experience — at least for those of us who prefer it to MySpace. So let’s say, you cannot follow more than 75% of those following you. Add more followers, and you can follow more people. But we know already that Twitter doesn’t scale well, so anybody following 6,000 people is doing something other than keeping tabs on that many friends. If you want a macro view of Twitter, Twitter tracking exists and so does Tweetscan. The Twitter API and the myriad tools built using it obviate the need to create one account following thousands of other accounts.

Unless, of course, you’re trying to promote something. However, as I’ve written before, Twitter is not especially useful for broad marketing. Thanks to tracking, one could hand-build a targeted list that could be worthwhile for the marketer and the marketed-to. Flyaxe, on the other hand, is wasting my time and his.

There’s a Spam on the Presidency, and it’s Growing

I know this is nothing new, but I still got a kick out of this spam comment, which showed up in my Akismet spam filter earlier today:

Anti-Bush spam

The link goes to a parked domain pushing a number of presumably illegitimate travel agency websites. So, even if this spam comment does indeed originate from Russia, they can still move to France if that whole impeachment thing doesn’t work out.

P.S. Bush has twenty-three months left in office — isn’t it time to start thinking about impeaching somebody else?