Re: Hashcash for Blogs

John Honan <[email protected]>
Newsgroups gmane.mail.spam.hashcash
Message-ID <[email protected]>
Mitch Denny wrote:

>Hi Jonathan,
>
>I just read this post by James Moore:
>http://www.freelists.org/archives/hashcash/03-2004/msg00003.html
>
>It seems he had an implementation in JavaScript that was quite slow. So
>maybe for web-form comments (as opposed to track-backs) the best mechanism
>would be what Atom Smasher suggestion in the form of a dynamically generated
>image which is hard for a machine to parse. Does anyone know of how easily
>those are compromised? I'd prefer not to have to have people download a Java
>applet to make this work, and most browsers definitely won't let an
>externally loaded page query a HTTP server on localhost - that would be a
>security violation.
>
>OK, so the way it is panning out is this:
>
>	1. Web-form posted comments are filtered using a image verification
>system.
>	2. Track-back comments are posted using hashcash. The server can
>choose to produce the stamps for the posting client (especially if it is via
>the web-based admin interface), or the client can provide a set of stamps
>that match the referenced URLs.
>
>  
>
Some more suggestions;

If your intention is to slow down the spammer and deter them from 
spamming your blog then why not just apply a delay at the server end? - 
Make it pause 15-20 seconds after they hit the 'submit' button before 
actually posting the comment. Would this achieve the same thing?

Another form of flood protection I've seen on forum software is to limit 
the amount of posts coming from any one IP address. If you try and post 
again too quickly, it won't allow you (in most cases you have to wait 2 
or 3 minutes before you're allowed to post again).

Actually stopping spammers from placing adverts on blogs or forums is 
quite difficult. As you state on your webpage, they only have to post 
once, and they end up getting indexed by google and viewed by possibly 
hundreds of people. If there is one posting coming from a person, and 
assuming they're not trying to flood the database, then how do you 
figure out if it's a genuine poster or a spammer? - In fact, the 
definition of 'spammer' kind of falls apart in this scenario. Instead of 
sending out millions of emails, they just make one posting and get it 
read by hundreds of people!

Another solution is filtering (does the post look 'spammy'?) - Or 
moderation, where each new post is sent to you first for review/approval 
before getting added to the blog.

John.
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.