It’s Friday, and a slow one at that.
As an exercise in milking the clock I checked the referral statistics to find that Airbag is being swept by Garst.com. The domain forwards to Net Identity which tells me that Garst.com is a shared domain and that I can get an email address using said domain.
Of what business does an email company have in lurking through all the entries of this site?
Awkward.
Join the fray by reading through and commenting at the end.
They are harvesting email addresses from comments so they can spam us. Joy.
Good thing MT encodes them....
MT encodes email addresses? This is the first I've heard of that.
Is there any way to block a service from doing this? I can block comment spam, so what about spiders?
Greg: Sure it does; MT replaces the colon in the "mailo:" pseudo-protocol, and the at-sign as well as periods in the address with their respective html ascii entities.
As far as blocking spiders go, there is of course nothing MT can do about that, but you can block them in your .htaccess file.
Actually, that was the wrong url, I meant this page.
A google search yielded this: Link.
A good read, I think. Do you have permission to edit .htaccesss and are you familiar with mod_rewrite?
MT replaces the colon in the "mailo:" pseudo-protocol, and the at-sign as well as periods in the address with their respective html ascii entities.
Ooo... that's sure to stop'em in their tracks. Like spambot owners haven't figured that one out by now.
Bob: My site statistics suggest they haven't, or at least that many of them haven't, because robots follow those links like they were files on my site and result in quite a bit of 404-requests.
From my knowledge gained by watching the Terminator, the machines will grow stronger. The line must be drawn here.
Tomas, that's encouraging news. Unfortunately, as we all are aware, the web evolves so quickly...
BTW, the derisive tone of my comment was not intended for you; rather, it was a jab at MT's (IMHO) too-simplistic encoding method.
If you know php... which it looks like you do since this URL I'm looking at has a *.php file in it...
You could add a line to the top of all of your pages that looks to see if the client coming in is from "*.garst.com" and if so, before anything is written to the page, just have the location headers change to anywhere else that you like. My personal preference would be manhole.com.
If you need to know the exact php code to get that, I can go into more detail - or just e-mail me if you would rather do that. Should be fairly easy (just make a file with the data and then include 'thatfile.php'; at the top of every page.
Greg,
I basically have stopped putting any email addresses on my site. All mail to me is done through formmail and l only provide attribution links in comments to authors with websites.
It doesn't stop the bots or the refer log spoofers, but at least no email addresses get harvested from my site. You can muck around with a robots.txt file in your home directory, but the spamming folk don't heed such roadsigns.
I'm using Dean Allen's Refer to track my incoming referers, and I've had hundreds of hits from 211.152.14.93, .94, .95, .96, .97, and .98. This is spam pure and simple. I've had luck using some extensions to refer ( http://www.textism.com/article/780/ReferSpamming ) to block the IP address of pesky site spammers like this guy.
This is referrer log spam from a chinese bot. Some Blogs are publishing their referrer logs, so these spammers are trying to get higher google ranks.
You can add these lines to your .htaccess:
RewriteEngine on
RewriteCond %{REMOTE_ADDR} ^211\.152\.14\.
RewriteRule .* - [F]
Hum, that makes me think... why not write a program that download and check the referrer page has actually a link to this site? Not much spam is going to get through that filter don't you think?
If you're lazy and aren't looking for the greatest solution ever you can just change the file name of your refer file. A couple weeks and they will be back on you... but then one could just change it again. Worked alright for me.
Check out this cool little javascript email encoder:
Hiveware: Enkoder Form 6.0
"The Enkoder Form will encrypt your Email address and convert the result to a self evaluating JavaScript, hiding it from Email-harvesting robots which crawl the web looking for exposed addresses."
- H
Shouldn't all this get fixed with MT 3.0? I've been waiting to do any major changes since it appeared that MT 3.0 might solve many of these issues.
While we're on the subject of referrers, I've a question. Recently, I've been getting a flood of referrers like the following:
[ed. - extremely long url]
...with www.theforeignembassy.com being my site. It seems to be the result of an images.google search, but I'm getting an awful lot of 'em. Anyone know what would cause that kind of hit?
Oops, sorry about the long URL.
No worries about the long URL, the CSS didn't like it.
I meant to reply sooner to say that I have been seeing the exact same thing as the result of a search for images of Iconfactory's xScope application. No matter how many times I try to follow the logic and the link I can't find the relation.
That was an interesting read, it made me think which is a true rarity these days.
I am packing up my bags as I write this to you all here at the lovely Airbag, getting ready for a journey into the past, present, and future. To stop by the graves of the innovators, and legends, and visionary minds who I am looking to now for my salvation.
Hope is a funny thing... I suppose it's the same thing with religion and carrying a continued faith and believe in something or someone that we never truly know even exists. It's that unknown, that mystery, that so many of us cling to when all else in this world has let us down, including human nature, at least in my case.
Thank you AIRBAG, I smiled here just now, maybe you'll stop into my LABYRINTH and smile as well.
My Love 2 U...
Jamie Leigh
AIM: xanAmericanGrrlx
On my site, my email address is written using Javascript. That should stop bots dead in their tracks, unless they start doing things like supporting Javascript (which seems unlikely). Downside is that you can't read it when Javascript is disabled, but you get a notification of that so you can enable it for a short while if needed.
http://www.students.cs.uu.nl/~lholst/
Take a peek at the source. It doesn't work with documents served as XHTML by the way, but if you're a bit into writing javascript for XML it shouldn't be much effort to adapt. If you want an XHTML-version (only neded when you serve a content-type of application/xhtml+xml to accepting browsers such as Mozilla), feel free to ask me for one.
~Grauw
Thanks, Greg. It's a weird type of link, but I could see an image search app making sense. I think I'll start uisng Refer, so I can just filter those hits out of my stats.
I would try using this great script on your site. It's able to block all sorts of fake refferrers and such.
http://www.skyzyx.com/projects/blocker/
Enjoy!