SPAM detection - Maximum number of external links that can be added to a blog?

Hi javagoza ,

I don't want to publicly say how many links. I can add permitted website domains to be exempt from that link count if I'm made aware of them.

Out of the box, Verint's moderation is strict and unforgiving - their train of thought is probably "well that simply catches the content, a human decides whether or not to allow it in the end".

What is most terrible is the lack of ability to interact with your content, and lack of back and forth of conversation on the moderation queue - as a user you deal with a black box. Your messages go into the void, you don't know if someone sees it or if you can act on it.

Most importantly though, I believe it hurts because your time is valuable and the moderation system does not respect you or your time or your effort as a user.

I refer to Verint's Community as 'user hostile'. We've told them as such and explained why, other customers of Verint have done so, too.

I've applied to the dev' team to introduce more measures of anti-spam that would allow us to relieve the pressure of the strict moderation system:

- Akismet anti-spam identification

- StopForumSpam spam account verification

These two methods would deal with 'known bad' spam accounts and stop them before they even had a chance to post, unfortunately, I'm still waiting for these solutions to be resourced.

But what can we do about this now?

One good thing that came with Verint v12 is that we can now exclude 'roles' or 'groups' of members from moderation. We could make this an automatic process, but since we can barely trust Verint's moderation system at present, I don't trust the system to do that (what's to stop a spammer legitimately raising their site reputation, and then abusing it, for example?).

On a case-by-case basis (and Top Members are considered trusted in this regard) I monitor the content that goes into the moderation queue and then I add people to a 'trusted member' site role where appropriate - this role circumvents moderation on the site.

It's taken far too long to get anything like this, but suffice it to say, I'm sure you and others can be a 'trusted member'.

Hope this helps.

Hi javagoza ,

I don't want to publicly say how many links. I can add permitted website domains to be exempt from that link count if I'm made aware of them.

Out of the box, Verint's moderation is strict and unforgiving - their train of thought is probably "well that simply catches the content, a human decides whether or not to allow it in the end".

What is most terrible is the lack of ability to interact with your content, and lack of back and forth of conversation on the moderation queue - as a user you deal with a black box. Your messages go into the void, you don't know if someone sees it or if you can act on it.

Most importantly though, I believe it hurts because your time is valuable and the moderation system does not respect you or your time or your effort as a user.

I refer to Verint's Community as 'user hostile'. We've told them as such and explained why, other customers of Verint have done so, too.

I've applied to the dev' team to introduce more measures of anti-spam that would allow us to relieve the pressure of the strict moderation system:

- Akismet anti-spam identification

- StopForumSpam spam account verification

These two methods would deal with 'known bad' spam accounts and stop them before they even had a chance to post, unfortunately, I'm still waiting for these solutions to be resourced.

But what can we do about this now?

One good thing that came with Verint v12 is that we can now exclude 'roles' or 'groups' of members from moderation. We could make this an automatic process, but since we can barely trust Verint's moderation system at present, I don't trust the system to do that (what's to stop a spammer legitimately raising their site reputation, and then abusing it, for example?).

On a case-by-case basis (and Top Members are considered trusted in this regard) I monitor the content that goes into the moderation queue and then I add people to a 'trusted member' site role where appropriate - this role circumvents moderation on the site.

It's taken far too long to get anything like this, but suffice it to say, I'm sure you and others can be a 'trusted member'.

Hope this helps.

I back you 100% on this. So keep pushing as their approach needs changing. I also think that the approach you've taken to deal with the now situation seems reasonable enough to me. At least we now know the situation. Thanks for providing the feedback.

Following this, some communities that run "Verint Community" set the permitted link count to zero.

Why would they set it to zero when it obviously annoys members?

Because spammers do this:

Let's look closer:

And it's not even an automated bot:

Why indeed.

Thanks to those members that have a keen eye for this and report the account, it goes straight into the moderation queue and the anti-abuse process.

I think a member whitelist is required, constructed of members who have reached X points with at least Y posts and Z interactions (likes, vote up/downs, etc); anyone who spams before X points reached are banned; anyone on the whitelist who spams has to explain themselves before being banned.  That ought to be automatable.

cstanton said:

What is most terrible is the lack of ability to interact with your content, and lack of back and forth of conversation on the moderation queue - as a user you deal with a black box. Your messages go into the void, you don't know if someone sees it or if you can act on it.

Most importantly though, I believe it hurts because your time is valuable and the moderation system does not respect you or your time or your effort as a user.

I believe it's not so much that the rules are at fault but it is as cstanton  explains... your draft post is whipped away, because you've broken some spam rule, and then you are unable to rectify as it's in the moderation queue (a user black box) thereby spoiling the vital time you've allocated to writing up the blog.

That discussion thread is ripe for spam. I was trying to see what was weird about this comment there:

My keyboard doesn't have that character, but maybe other country ones do. I couldn't see any embedded URLs so far, but perhaps they may occur later (if that comment allows editing).