cdave: (Angry)
[personal profile] cdave
Sigh.

I'm fed up with this J-spam. If you've had any recently could you go comment on this support request IPs and idea's especially welcome.

If you don't know what I'm talking about by 125.14.xxx.xxx, 125.15.xxx.xxx or Lewis then casadetodd has a good explanation.

Date: 2009-05-27 02:19 pm (UTC)
andrewducker: (Default)
From: [personal profile] andrewducker
I'm not sure there is a good solution to this, except to screen anonymous comments - or at least put a captcha on it, both of which are already options on the setting page.

IP blocking is notoriously unhelpful in these situations - spammers just switch IP addresses, and normal users behind the same proxies get shafted.

Date: 2009-05-27 02:53 pm (UTC)
From: [identity profile] chiller.livejournal.com
I had my anon comments screened with catcha and they continued to come.

Date: 2009-05-27 03:01 pm (UTC)
From: [identity profile] chiller.livejournal.com
Well, I don't know - I only know that I had it enabled and the bloody things kept coming. Have you tried enabling it on your LJ to see what happens? Since I went 100% private and then re-emerged, I've not had a dicky-bird.

Date: 2009-05-27 03:31 pm (UTC)
From: [identity profile] chiller.livejournal.com
See, some people maintain this is a bot. If it is, it's a bloody good one, because the JS has made direct (albeit slightly batty) responses to things I have posted, and [livejournal.com profile] offensive_mango has had the same thing happen when she posted a photograph (the spam referred to something in the photo - not even to something that had existed on the page in text, IIRC).

The batty-ness of the responses could be put down to the spammer using translation software to read and/or create responses to posts. Tokenisation (the identification of individual words) in Japanese automated translations is notoriously appalling and is one of the problems I run into in my work all the time. It gives rise to a slight or profound context drift in any translated document. Many of the spam messages I received made no sense whatsoever, but the ones that did appeared to be the result of one of the following: a) someone reading a translated version of my post or b) posting a poorly translated (Japanese to English) response to it, or c) of a bot with a pretty good language algorithm doing it.

The problem is, the photograph incident on Mango's journal rather kiboshes the bot-with-language-algorithm idea.

I don't know if bots can deal with captcha text. It's just not my area. I know some humans can, although frankly I find it a bit challenging. ;) I'm still of the belief that this is more likely to be a nutter than a bot, but whatever/whoever it is, I'm interested to know why taking my journal out of its reading list for a week stopped it dead.
Edited Date: 2009-05-27 03:32 pm (UTC)

Date: 2009-05-27 03:55 pm (UTC)
From: [identity profile] chiller.livejournal.com
So it looks more likely that my "nutter" theory may have some weight to it.

Maybe we should call their bluff and ALL turn up?

Date: 2009-05-27 04:46 pm (UTC)
From: [identity profile] chiller.livejournal.com
Totally. *knock*knock* LEWIS! HI! Mwa! Mwa!

I suspect that alcoholism is involved.

Date: 2009-05-28 08:03 pm (UTC)
andrewducker: (Default)
From: [personal profile] andrewducker
Did you get anywhere with this?

Profile

cdave: (Default)
cdave

June 2018

S M T W T F S
     12
3456789
10111213141516
17181920212223
2425262728 2930

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Jul. 13th, 2025 07:36 am
Powered by Dreamwidth Studios