Whoa-- LJ spam!

[dragonlady7]

I just got a comment on an entry I posted in June. The entry mentions how a bird at a window scared a cat which scratched me in terror, leaving a scar. The comment is a link by an LJ user to several websites about window-watching birds and the like.
Obviously she (I say she because the lj name contains the word "girl") has searched for entries containing keywords and is just leaving the comment on those entries, to increase the backlinks to her site. Which is an old search-engine optimization trick.
Thing is, that's pretty fucking blatant spam. Where do I go to report it on Livejournal? I can't remember, and can't find the link.

And no, she pretty obviously hadn't read the entry. More interestingly, she left the comment using a different username than the journal she linked to. Probably because she expects the spamming username to get banned. I tell you what, I don't care if it's a research journal rather than porn: if their IP addresses are the same, LJ oughta nuke 'em both. There's a definite line between reading and commenting in other people's journals so as to increase your own readership, and spamming other people's journals, and you know what that line is? I bet you do.

Crossing that line is what makes the Internet a bad place. Anyone who crosses that line deserves to get fried.

Comments

[redstapler.livejournal.com]

Thoroughly unrelated, but this is a link (http://www.irandefence.net/showthread.php?t=29) I thought you'd enjoy.

Female soldiers from all over the world!

[dragonlady7.livejournal.com]

See-- thoroughly unrelated but actually relevant! I've been debating the difference between spam and not-spam, and how a machine's supposed to tell the difference. This comment *looks* more like spam than the other one, but isn't, because it's relevant to other themes in my journal. ...

That's a really weird forum, though-- who on earth are those people?
But interesting pictures. Overwhelmingly, the female soldiers of the world are apparently blond and hot. (Except the Asian ones, who are brunette and hot.)

That Croation one was a real hottie. (How could they tell that was a woman?)

And obviously other armies of the world have much less stringent requirements about hairstyles. ^.^

[mother2012.livejournal.com]

You're absolutely right. Anyone who is in someone else's LJ ought to be there because they find that person interesting. On a personal level.

She could easily have replied to you personally. She could even have said something about her own research journal, making a personal appeal as to why you might find it to be something you might want to do.

Spam is spam.

On the other side of the issue, I sent my daughter a news piece that absolutely inflamed her, so she sent it out to everyone on her AOL list. They froze the account because it must obviously be spam!

[dragonlady7.livejournal.com]

Yes: the problem is that there's a clear line between what's spam and what's not, but it's not a clear machine-detectable line. Which is really what has caused all the spam wars, both in email and search-engine spam (I used to do search engine optimization consulting, and there's a constant battle between "white-hat" and "black-hat" optimizers-- the black-hats are responsible for all those times you've done a search and found a whole page full of results that take you to pages consisting of utter junk that contains your keyword over and over and over.

Humans can spot it instantly. Machines must be rigorously programmed to spot it. I got one true comment spam on my journal months ago, that was just a big pile of links to a porn site, and LJ flagged it in the notification email and provided the abuse link-- Livejournal's machines were pretty sure it was spam, and so I followed the link and reported it, and it was neatly taken care of. This one, their machines didn't flag, because it looks like a genuine response to my post. It isn't, of course; it's an irrelevant response to a keyphrase occurring in my post, but it's possible it was done by a human or by a human-supervised program. It looks like a real person. But within two seconds I had recognized that it was spam-- because I could read *and understand* it.

They're still working on the machines' ability to do that, I guess.