Wednesday, February 13, 2019

So It's Great Work Now


The review of the citation this blog received from Google for containing allegedly dangerous or derogatory content is copied below. The specific name of the program policy is blotted out.  

Dear Publisher, 
This Google Publisher Policy Report gives you an overview of recent activity related to violations found on specific pages of your websites. As enforcement statuses may change over time, please refer to the "Page-level enforcements" section of the xxxxx Policy Center for the current list of active violations.  
(The Policy Center in blue text above was a clickable link, which took me to a page that showed the following screen-captured information.)  



The email continued, with attention to the bold text:

Please note this report doesn’t cover violations that may happen on an overall site or account level. You may be notified by a separate email if site or account level violations are found. Ads will continue to serve where no policy violations have been found, either at the page- or site-level. 
In the last 24 hours: 
  • 1 page was reviewed at your request and no policy violations were found on the page at the time of the review. Ad serving will be restored on this page and your monthly review limit will be credited.

Further details on enforcements can be found in the xxxxx Help Center. To learn more about our program policies, please view the xxxxx Program Policies.  

Kind regards,
Google Publisher Policy

The above form email was sent from a "no-reply" email address.  So I could not contact Google xxxxx to request an explanation for this insult.  All the anguish this caused me was for nothing except to disable ad serving (paying for clicks on page ads).  This incident was an egregious error on their part which they failed to admit and offer an appropriate apology, which means it could happen again.  

I went to the xxxxx Help area and sent the following query:
Re:  https://blackdollcollecting.blogspot.com/
After receiving an erroneous policy violation notification for "dangerous or derogatory" content, without fixing anything (because there was nothing to fix), I requested a review.  The result of the review was that my page was not in violation.  

I'd like to know how these violations are determined.  Are pages reported by individuals?  Is an algorithm used to detect keywords that might be deemed inappropriate?  Randomly suspending ads on pages that are not in violation and providing a review that finds no violations of any policy is an insult to the publisher and a waste of time to all parties involved.  This procedure needs to be revised to include diverse humans who can make valid decisions as to what is in violation and what is not.

Wouldn't it be more considerate to provide the publisher with specific text or images that are in violation instead of the vague details that a page is in violation that needs to be "fixed" and automatically suspend ads before a review can be requested?

The Reply:
I notice you have written two blog posts about this on your site. Writing about xxxxxx is not a good idea. It draws attention to your ads, and could result in invalid clicks which could harm your account. I'd remove them if I were you.

Violations can be detected by bots (looking for stop words or scanning images). Or they might be the result of a human review following a report. In your case it seems like a bot detection. The bots aren't intelligent enough to know the true context.

It wouldn't be cost-effective for all bot detections to be reviewed by humans as well, as most of the time the bots get it right.

Your review request was successful.

My opinion is, the system is working as designed, even though it was inconvenient.
--------
My Reply to the Reply: Thank you for the explanation.  I agree; it is very inconvenient when there is no there there.  
--------
(The name of the specific program is blotted out in the previous post as well.  My opinion is the bot system failed miserably in this case.)  


dbg
There is always something to collect and write about. Black dolls chose me.
__________

Thank you for following, commenting, and sharing using the share button below.

Check out what I am selling here
Check out my eBay listings here.
Please follow my sister blog Ebony-Essence of Dolls in Black.
Donate here to support this blog. Thank you!

6 comments:

  1. Debbie,

    I am glad that the situation was resolved but I don't get the whole tone of their message to you. It sounds like what my mother would call a "back-handed" apology. I think they should look at your suggestion to provide recipients of their violation messages with a snap shot of what is wrong. I am sorry that you had this experience.

    ReplyDelete
    Replies
    1. Thank you, Pam.

      I agree that some form of evidence should be provided with these violation warnings. A screenshot or the something visual would help, but that probably wouldn't be "cost effective."

      I don't understand the reason this person "can understand why they flagged it." What about my page would cause a bot concern?

      dbg

      Delete
  2. Ah, subtle intimidation ("not a good idea" to write about specific policy and "I'd remove them if I were you") plus a non-apologetic apology: Keep up the great work, Google!

    The "I can understand why they (bots) flagged it" statement is DEEPLY weird.

    ReplyDelete
    Replies
    1. The subtle intimidation almost worked, Gini. I was about to delete, but decided to just remove the specific policy name. These posts, this one in particular, needed to remain published for readers to see the results of the review. I'll remove or place in draft mode at my discretion within a couple of days.

      dbg

      Delete
  3. I’m so glad that they realized what we all already knew. So glad for you and your followers.

    ReplyDelete
    Replies
    1. Thank you, Brenova! I am glad it's over and hopefully won't happen again. If it does, I'll know it's probably their imperfect bot patrol.

      dbg

      Delete

Your comments are appreciated. To eliminate spam, all comments are being moderated and will be published upon approval. Thank you!