Is Spamming a Security Breach?
victor | 23 January, 2006 11:22
I received some interest emails from readers regarding the relation between Spamming and Security breaches. In summary, some of these emails thought about spamming comments as a security breach related to Denial of Service attacks where the space is "filled" up by bogus messages.
I tried answering some of them but it turned out that a lot of typing and explanation is required so I figured out that the best thing to do would be to post the anwer in here as an article.
The question to answer is the following:
How To Avoid Blog Spamming?
Read more for the answer.
In Denial of Service attacks, a system is usually busy serving bogus requests sent by attacking hosts. In the case of Spam, this becomes a problem if all the disk space on the website is filled up with spam and, thus, no more comments are being accepted.
In the case that I faced lately, the website was not full yet (luckily) since my space is almost unlimited here. Thus, the main concern was not about whether this was a security breach or not (since it failed filling up the space) but was whether these spamming comments were annoying to visitors or not.
Spamming IS Annoying to Users
Spam is not a security breach. It is simply an annoying action done on public systems by kiddos.
You avoid spam to maintain your visitors, not to avoid security breaches.
If DoS occurs, then spam might be considered as a security breach. In
that case, PUBLIC systems must be taken private to prevent spam (such
as Email Spamming). In my case, I would like to keep my blog a public
system (this is called a business constraint) to maintain a secure anonymous system.
This creates a two-weight balance between anonymity and trust. Thus, I will have to balance between spam and normal visitors.
Action taken? Visual
Code Confirmation
The reasoning is that spammers use tools to attack websites.
Tools fail visual code confirmation. Thus, tools are stopped.
Spam Tools vs. Human Being Spammers
The question now becomes: what if human beings attack the comments section?
In
this case, the visual code countermeasure fails.
Another countermeasure
becomes required here and that is to train the system to identify spam and block
spammers. This feature is already built in but was time-consuming
in the earlier case due to the large amount of spam messages that was recevied. The
ones that appeared on the website were approx. 30% of the real messages that
were received.
Now that human intervention is requried, the effectivity rate becomes
bigger and the IP Blocking feature becomes more effective ;)
If they
all fail, the only solution will be private blogging and this is
exactly what I don't want to achieve.
How Long Will This Last?
The real challenge now becomes whether the spam identification and IP-Blocking countermeasures will last long before someone comes up with an attack that succeeds at both levels. My answer is normally to wait and see. Needing a solution is at the basis of all Inventions. We will wait and see and, when these countermeasures start failing, I will worry about getting something new. Meanwhile, I don't see any reason for this headache ;) This is a personal blog afterall and I don't see aliens flying towards it yet with pretty advanced security attacks :) If you ever see one, please let me know and I will make sure to solve it. Meanwhile, I will be working on a private version of this blog just in case it took me a long time to come up with a solution.
Related Articles: