This must work in some instances as they keep trying.

I would have more of a problem with it if it were in a public access environment (such as a library) where the users couldn’t change the setting themselves.

The community driven aspect of the filtering list doesn’t really change my opinion of the concept of filtering in general; if it’s not an optional choice by the user then I’m opposed to it flat out (so I’m OK with OpenDNS doing it as long as the use of OpenDNS is optional). The reason is that even if it is truly community driven (which most are not - the company usually still has the final say) then it’s still an opportunity for discriminatory censorship. If something I’d like to read is blocked, I don’t really care whether the larger community agrees that it should be or not.

The above doesn’t account for the censorship issue but from what I see the companies they use are community driven. So if there is a problem with an incorrectly tagged site then it would should be changed fairly quick. Similar to the approach of open source security.

You’re “concern” is understandable but I prefer prevention rather than a speculative concern.

If it does get out of control there’s always the option of stopping your use and/or contributing to the communities they use or any other company uses.

Anyway, I’m not too worried about them blocking phishing sites that actually are phishing sites, but they do mention “phishing and malware” related sites in their description.

I’m not accusing OpenDNS of anything like this, but to me any content filtering is a slippery slope towards censorship. For example, what are the criteria used to determine whether a particular page / site is “malware” related? What if I (or an anonymous commenter) am discussing a particular security vulnerability on my blog which could be exploited to do some damage by people with foul intentions? Is that enough to get my site blacklisted? If so, what do I do if I complain and they tell me they feel that they were correct and it should be blocked?

Since those criteria are not disclosed it’s basically arbitrary as far as we know. They might have business relationships with companies that allow undue influence in their filtering decisions as well. For example, if they’re in bed with Apple and Apple all of a sudden decides they aren’t down with the AppleTV hacks anymore, then your blog could be banned for distributing “illegal hacking instructions”.

Again, I’m not saying OpenDNS would ever do something like that, it’s just a general concern I have with all net filtering systems. Then again, they are a company where management can change hands at any time, so you never know.

For some who aren’t bothered by that, it’s perfectly fine to use them, as long as you understand and are willing to accept the potential limitations.

How does OpenDNS decide if a site is a phishing site?

How do I report a phishing site to OpenDNS?
How do I tell OpenDNS about a mistakenly-blocked site?

What is phishing?
How does OpenDNS decide if a site is a phishing site?
How do I report a phishing site to OpenDNS?
How do I tell OpenDNS about a mistakenly-blocked site?

How does OpenDNS decide if a site is a phishing site?
How do I report a phishing site to OpenDNS?
How do I tell OpenDNS about a mistakenly-blocked site?

The “filtering” capabilities of openDNS is actually one of the potential downsides for me. As with any filtering system, the question is who controls the list, and what happens when something gets put on the list that you actually want to go to or see.