People posting "dodgy" links

Hi guys,

I need a bit of help from you regarding ethics, data protection, and suchlike.

As i've mentioned before, my website http://zipurl.co.uk, (a service to shorten urls) is attracting a fair amount of attention from spammers, scammers, and wrong'uns...

Last week it was links to be distributed in spam emails, this week it's people (or "person") creating links to Google searches for "Paedophilie" and "kinderpornographie".

What should I be doing about it? I already track the IPs of the people posting links, but should I also be tracking the people clicking the links?

Should I be reporting this kind of behaviour to the authorities? If so, who specifically should I be contacting?

Cheers,
Karl

web design, development & seo by DigitalVibe

In reply to:
As i've mentioned before, my website http://zipurl.co.uk, (a service to shorten urls) is attracting a fair amount of attention from spammers, scammers, and wrong'uns...

Karl, I don't know any way you are going to be able to effectively prevent that; I'm convinced that it is just "the nature of the beast" with that kind of a service. In my experience, people will *always* abuse/misuse that kind of a service (much like an image storing or file sharing service). frown

In reply to:
What should I be doing about it? I already track the IPs of the people posting links, but should I also be tracking the people clicking the links?

I think you would have even less success trying to track "clickers" than you are likely to have trying to track (and control) the posters. It could easily become an almost full-time pursuit, and at the end of the day, is not lilekly to be effective in actually curtailing abuse/misuse.

In reply to:
Should I be reporting this kind of behaviour to the authorities? If so, who specifically should I be contacting?

I guess it would depend upon what it was, but I, for one, would not want to get involved in that unless it was a particularly heinous instance, or I felt particularly strongly about a given instance.

It would be a massive timesink, and the argument can easily be made that, if you don't want to facilitate such behavior, you probably shouldn't make it available as an "open" service.

Of course, all of that is just my opinion, and YMMV, but that's how I perceive it. wink

--rlparker

Good points well put smile .

I started it originally as a personal project... like an "I wonder how tinyurl do it", and it fast became a reasonably busy service.

I'm going to keep it running as it does pull in a fair amount of traffic to my main website... although it does need some work to improve the logic.

All things considered, i'll keep it running and keep logs. If people report inappropriate or spam links, i'll ban the url and if needs be the user's IP, but I definately will not be religiously monitoring the activity.

I'll cooperate with any legal requests, so i think it's best I keep logs, but that's as far as i'll go.

Thanks for the input smile

Cheers,
Karl

web design, development & seo by DigitalVibe

I just used your site and it made the URL longer. http://zipurl.co.uk/3x/ tongue

It would be easier to just take the site down then go through that trouble, unless it's somehow generating a bunch of revenue to make it worth it. Otherwise, if it doesn't jeopardize your hosting account, I probably wouldn't worry about it too much.

If it's clear what the site is for, I don't see how you'd be responsible for the links. I'd think that if one of your links was reported to DH, they'd figure out what was really going on.

Just leaving the redirecting page up should tip most people off that the link they clicked isn't where they're being sent.

In the end, the site the link leads to is what should be dealt with, and/or the person emailing the link, since he's the spammer. So, one of those would be the user's host's responsibility, while the other would be the ISP's. If they do their jobs, your short link becomes useless on its own.

Maybe require an account to make links? That way, it shows a human took time to make the account, then confirmed it through a valid email address.

Then, if they use it for spam links, you might be able to kill 20 spam links by simply disabling one account.

It's a hassle for a spammer that needs to keep making links, but a quick easy process for those that actually have legit uses for it and can keep adding links without breaking the TOS.

If you didn't want to require an account, you could have restrictions for those that don't use them. Examples would be that the link expires in a short time, limit to 1 (or whatever) links per IP per day. Or anything else that would annoy spammers, but encourage legit users to create an account.

Good points that you raise there.

My current process logs the IP of the person submitting the url, so i effectively use "anonymous" user tracking - anonymous int hat i don't collect user data - just the simplest way to identify a "user".

That way if one person from an IP submits a dodgy link, then their IP is banned, along with any previous urls submitted by that IP. In addition, their IP is blocked from even using the service.

I'm taking a very hard line towards people abusing the service. Naturally i want to protect myself as much as possible, and it's literally a 30 second job to block an IP and ban all an IP's submissions at the same time.

Cheers,
Karl

web design, development & seo by DigitalVibe

Someone created links to Google searches?

I'd report them to the International Ergonomics Association.

It's logical... someone wants to email a link and bypass a content filter in a corporate system. Disguise the link.

Cheers,
Karl

web design, development & seo by DigitalVibe

If I were doing what you're doing I would build in a keyword filter. If you wrote the existing software yourself then it won't take long to write.

It seems that you keep a close eye on logs as it is, so taking a few extra seconds every time you come across something you don't like will be of benefit.

From a tactical point of view, if you take a slightly more draconian approach you'll discourage the (I'm guessing small) number of people who are abusing your service.

Good point, however the problem comes in that most of the problem links are perfectly legitimate sounding domains, or non-english search queries... that makes it all the more difficult to maintain a list of suspect phrases.

I could use the keyword list to produce a report of any new links with suspect content (such as the words "porn" and "sex"), and the option to ignore or ban them.

Another problem is evident in that most porn/"wrong" related words can also be used in a perfectly legitimate and proper context... such as "ratio of consumers by sex", "campaign against porn", etc etc... i'm sure you get the idea.

Using the same logic, that Draconian approach could well turn away genuine people.

I've actually noticed a drastic drop in service abuse since i've blocked The Philippines from accessing the site wink . Obviously these abuse cases are a distinct minority, but as i mentioned previously, the large part of the abuse is coming from user(s) on one Philippines based ISP, and i've not had *any* legitimate users from that IP range.

Cheers,
Karl

web design, development, pay per click marketing (PPC) & Search Engine Optimisation (SEO) by DigitalVibe

People posting "dodgy" links