pancakes

MicrostockGroup Sponsors


Author Topic: Email from SS - Regarding Repeated Words and Phrases in Image Titles  (Read 51944 times)

0 Members and 1 Guest are viewing this topic.

« Reply #275 on: July 02, 2016, 21:27 »
+4
http://www.shutterstock.com/de/pic-385305055/stock-vector-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart.html


What a sad portfolio - perhaps 200 actual images and endless repetitions of similars - glasses, skulls, easter eggs and so on

http://www.shutterstock.com/pic-318128897/stock-vector-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses.html

The titles are spammed, the keywords are spammed and visually, 9/10 of the portfolio is spam. SS must really want to bulk up the library to permit this sort of pollution.

If you change the title after approval, do they change the image page URL? If not, these URLs indicate the image was approved this way, not changed afterwards.


« Reply #276 on: July 02, 2016, 23:22 »
+2
The e-mail obviously wasn't a mistake. The mistake was in programming the bot that flagged up the "spammed" results, otherwise the initial confirmation that the warning was real wouldn't have been made. They must have thought a few dozen or few hundred serious spammers would be pulled up and receive the automated mail, but the bot made vastly more connections than they expected. That's how I read it, anyway.

But that's where one of several huge mistakes were made. The program that scoured the images looking for spam and the program that sent the email should have been separate and independent. Run the first program, evaluate the results, and only then, let the bot send the email. If they had done that, it would have been obvious that the first part of the program was buggy or the parameters were incorrect.
There was a famous chess player who said that when his opponent made his first mistake he usually ignored it because it was invariably preparing for an even worse mistake. I guess the same principle works here, too.

« Reply #277 on: July 03, 2016, 00:12 »
0
The e-mail obviously wasn't a mistake. The mistake was in programming the bot that flagged up the "spammed" results, otherwise the initial confirmation that the warning was real wouldn't have been made. They must have thought a few dozen or few hundred serious spammers would be pulled up and receive the automated mail, but the bot made vastly more connections than they expected. That's how I read it, anyway.

But that's where one of several huge mistakes were made. The program that scoured the images looking for spam and the program that sent the email should have been separate and independent. Run the first program, evaluate the results, and only then, let the bot send the email. If they had done that, it would have been obvious that the first part of the program was buggy or the parameters were incorrect.
There was a famous chess player who said that when his opponent made his first mistake he usually ignored it because it was invariably preparing for an even worse mistake. I guess the same principle works here, too.

consider the issues that have been occuring for the past 2 years:
- review of images , out of focus.. when many complained they were not OOF
- mass rejection of images by experienced contributors (poor composition, WB,etc as if
they all suddenly contracted alzheimer and forgot composition, wb, focusing,etc)
- email sent out warning many, correction... a majority.. that their accounts will be suspended
for spamming.words and phrases.

i suspect all these were done by bots.
it was only when Oringer came in to his corporation to find most of the work have not been
done by the humans he hired and have been paying them wages while they
played ping pong or use the new building facilities during work hours
while their workstation were all being manned by bots.

if it were not for this latest incident, a worse tragedy could have happened
ie. the king bot would be taken over the whole company
and bought over all the shares of ss,
and we would have a new AI takeover movie in real life.

and we would all be extras in the movie, without being paid  8)

Justanotherphotographer

« Reply #278 on: July 03, 2016, 00:53 »
+1
http://www.shutterstock.com/de/pic-385305055/stock-vector-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart.html


What a sad portfolio - perhaps 200 actual images and endless repetitions of similars - glasses, skulls, easter eggs and so on

http://www.shutterstock.com/pic-318128897/stock-vector-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses.html

The titles are spammed, the keywords are spammed and visually, 9/10 of the portfolio is spam. SS must really want to bulk up the library to permit this sort of pollution.

If you change the title after approval, do they change the image page URL? If not, these URLs indicate the image was approved this way, not changed afterwards.

What a shame . That person has some talent and individual style too. The spamming must be an act of desperation.

« Reply #279 on: July 03, 2016, 00:55 »
0
I got it and then the next morning got a we apologize you shouldn't have gotten that letter.  Something about to wide of an audience.


Sent from my iPad using Tapatalk

« Reply #280 on: July 03, 2016, 01:06 »
+5
http://www.shutterstock.com/de/pic-385305055/stock-vector-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart.html


What a sad portfolio - perhaps 200 actual images and endless repetitions of similars - glasses, skulls, easter eggs and so on

http://www.shutterstock.com/pic-318128897/stock-vector-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses.html

The titles are spammed, the keywords are spammed and visually, 9/10 of the portfolio is spam. SS must really want to bulk up the library to permit this sort of pollution.

If you change the title after approval, do they change the image page URL? If not, these URLs indicate the image was approved this way, not changed afterwards.


How can this get past a reviewer? And here the rest of us have to put up with seemingly random rejection reasons. Something is deeply wrong there.

« Reply #281 on: July 03, 2016, 01:47 »
+2
http://www.shutterstock.com/de/pic-385305055/stock-vector-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart.html


What a sad portfolio - perhaps 200 actual images and endless repetitions of similars - glasses, skulls, easter eggs and so on

http://www.shutterstock.com/pic-318128897/stock-vector-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses.html

The titles are spammed, the keywords are spammed and visually, 9/10 of the portfolio is spam. SS must really want to bulk up the library to permit this sort of pollution.

If you change the title after approval, do they change the image page URL? If not, these URLs indicate the image was approved this way, not changed afterwards.


How can this get past a reviewer? And here the rest of us have to put up with seemingly random rejection reasons. Something is deeply wrong there.


I have the same question, what was the reviewer doing when he let these vectors pass with such a title:
Emoticon emoji set. Emoticon emoji icon. Emoticon emoji design. Emoticon emoji flat. Emoticon emoji art. Emoticon emoji image. Emoticon emoji illustration. Emoticon emoji vector. Emoticon emoji eps 10

http://image.shutterstock.com/display_pic_with_logo/2271365/429629455/stock-vector-emoticon-emoji-set-emoticon-emoji-icon-emoticon-emoji-design-emoticon-emoji-flat-emoticon-emoji-429629455.jpg

Even now these spammers don't care to correct these titles, obviously they got no email, or just don't care.
These spammy titles are a new trend and obviously they work wonders in search and
rank.

« Reply #282 on: July 03, 2016, 02:17 »
+4
Generally all this is not about a technical mistake, but continuing discussion about changes of treament of contributors in ss started several years ago. Workflow, e-mail wording, dependance of sales from outside activities of contributor, manner of communication, canned answers, etc etc etc. Attempt to remove spammers served other purposes.
Here are a lot of programmers on msg. Who will think that creation of a script which will cycle through database searching for spam by selected criteria, is so difficult that could lead to such a mistake? How to protect the data, resist attacks, data manipulation with such level of work? Shame really.

« Reply #283 on: July 03, 2016, 03:25 »
+5
I got it and then the next morning got a we apologize you shouldn't have gotten that letter.  Something about to wide of an audience.


Sent from my iPad using Tapatalk

there was no apology

Shelma1

  • stockcoalition.org
« Reply #284 on: July 03, 2016, 05:15 »
+5
http://www.shutterstock.com/de/pic-385305055/stock-vector-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart-heart.html


What a sad portfolio - perhaps 200 actual images and endless repetitions of similars - glasses, skulls, easter eggs and so on

http://www.shutterstock.com/pic-318128897/stock-vector-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses-glasses.html

The titles are spammed, the keywords are spammed and visually, 9/10 of the portfolio is spam. SS must really want to bulk up the library to permit this sort of pollution.

If you change the title after approval, do they change the image page URL? If not, these URLs indicate the image was approved this way, not changed afterwards.


How can this get past a reviewer? And here the rest of us have to put up with seemingly random rejection reasons. Something is deeply wrong there.


I have the same question, what was the reviewer doing when he let these vectors pass with such a title:
Emoticon emoji set. Emoticon emoji icon. Emoticon emoji design. Emoticon emoji flat. Emoticon emoji art. Emoticon emoji image. Emoticon emoji illustration. Emoticon emoji vector. Emoticon emoji eps 10

http://image.shutterstock.com/display_pic_with_logo/2271365/429629455/stock-vector-emoticon-emoji-set-emoticon-emoji-icon-emoticon-emoji-design-emoticon-emoji-flat-emoticon-emoji-429629455.jpg

Even now these spammers don't care to correct these titles, obviously they got no email, or just don't care.
These spammy titles are a new trend and obviously they work wonders in search and
rank.


I'll bet you dollars to donuts that everyone who received the first email received the second as well. I'm sure they didn't do any further programming to tease real spammers out of their original email list. So all the spammers now feel there's no action needed on their part.

Obviously this port passed through review with the spammy titles in place. All in all a complete fail for Shutterstock.

« Reply #285 on: July 03, 2016, 06:35 »
0
I'll bet you dollars to donuts that everyone who received the first email received the second as well. I'm sure they didn't do any further programming to tease real spammers out of their original email list. So all the spammers now feel there's no action needed on their part.

Obviously this port passed through review with the spammy titles in place. All in all a complete fail for Shutterstock.

I could understand if a few of these spammy images passed review but there are thousands of spammy vectors and entire ports. You just have to hover over the images on this page and see for yourself.
« Last Edit: July 03, 2016, 06:41 by Dodie »

« Reply #286 on: July 03, 2016, 08:23 »
+1
So, how do we know, that the titles weren't changed after files were accepted?

« Reply #287 on: July 03, 2016, 08:28 »
+3
So, how do we know, that the titles weren't changed after files were accepted?

I believe that it's because SS assumes that their reviewers would have picked up the spam and the artist "HAD TO HAVE CHANGED" them post acceptance.  Hence the accusatory message in their email. To me it seems like an awful lot of work to submit a legit batch then go into the editing system after acceptance with the strategy to spam your titles (and even keywords). So there has to be culpability on the part of inspectors.  It will be VERY INTERESTING over the next couple of months to track a few spammy ports to see if they actually take action on the offending artists.

« Reply #288 on: July 03, 2016, 09:04 »
0

Obviously this port passed through review with the spammy titles in place. All in all
I could understand if a few of these spammy images passed review but there are thousands of spammy vectors and entire ports. You just have to hover over the images on this page and see for yourself.



What gets me is that contributors catch the problems before the actual people who get paid to review do. Apparently they dont have paid people who continually monitor the database. But why should they...they have us. I am willing to bet it only came to their attention after a contributor noticed and complained.  ::)

Yes, it will be interesting to see what becomes of this whole snafu.

« Reply #289 on: July 03, 2016, 09:27 »
+1
So, how do we know, that the titles weren't changed after files were accepted?

I believe that it's because SS assumes that their reviewers would have picked up the spam and the artist "HAD TO HAVE CHANGED" them post acceptance.  Hence the accusatory message in their email. To me it seems like an awful lot of work to submit a legit batch then go into the editing system after acceptance with the strategy to spam your titles (and even keywords). So there has to be culpability on the part of inspectors.  It will be VERY INTERESTING over the next couple of months to track a few spammy ports to see if they actually take action on the offending artists.


That maybe a lot of work, but if it really helps search positions it might be worth it. After all, you also re-keyworded your old files, right?

But they just might let the files go through when they are reviewd. After all, they're also allowing all that spam to their database. Who knows.

And I think, that SS, allowing this, is digging their own grave. It's just a matter of time, when customers will be fed up with all that crap. How can they even find images, when searching for "new" content, when there is pages and pages of that crap.
« Last Edit: July 03, 2016, 09:52 by Dumc »

« Reply #290 on: July 03, 2016, 09:29 »
+1
If SS was to remove all the spam and similars (icons for example, or that pot-guy), their databse would shorten for a few million files...

Shelma1

  • stockcoalition.org
« Reply #291 on: July 03, 2016, 10:20 »
0
So, how do we know, that the titles weren't changed after files were accepted?

Well, I just changed one of my titles to see what would happen, and the URL stayed the same. So either this passed review or some people's work gets uploaded without review.


« Reply #292 on: July 03, 2016, 10:28 »
+1
Well, changes don't take effect immediately, so, maybe if you wait a day?

« Reply #293 on: July 03, 2016, 10:35 »
0
How do you get URL with the title anyway? I only get URL with image ID number.

« Reply #294 on: July 03, 2016, 10:56 »
+3
Here is a URL with title:
http://www.shutterstock.com/pic-363531434/stock-photo-wild-brown-hare-with-big-ears-sitting-in-a-grass.html

But you can simply replace that title part with anything:
http://www.shutterstock.com/pic-363531434/shutterstock is great.html

Image ID is the only important thing here...


Shelma1

  • stockcoalition.org
« Reply #295 on: July 03, 2016, 11:02 »
0
Well, changes don't take effect immediately, so, maybe if you wait a day?

I waited a few hours and refreshed the image. The title changed, but the URL stayed the same.

(You get the url with the title when you don't log in as a contributor. When you're logged in you see your image # in the url instead.)

« Reply #296 on: July 03, 2016, 11:10 »
0
Ok, thanks.

« Reply #297 on: July 03, 2016, 12:37 »
+1
I'm pretty sure that those ports were uploaded with the spammy titles and I'm sure because there are many new ports from 2016 among them. Which beginner contributor would risk to be suspended right at the start?

This is very sad, not just because it is unjust but because it encourages further spamming.

« Reply #298 on: July 03, 2016, 13:27 »
+1
I'll bet you dollars to donuts that everyone who received the first email received the second as well. I'm sure they didn't do any further programming to tease real spammers out of their original email list. So all the spammers now feel there's no action needed on their part.

Obviously this port passed through review with the spammy titles in place. All in all a complete fail for Shutterstock.

I could understand if a few of these spammy images passed review but there are thousands of spammy vectors and entire ports. You just have to hover over the images on this page and see for yourself.


so now we see that perharps there is no review going on with vectors
or that there is a bunch of lax reviewers for vectors
while there is an extreme ANALysis of reviewer(s) for photographs.

« Reply #299 on: July 03, 2016, 13:36 »
0
Here is a URL with title:
http://www.shutterstock.com/pic-363531434/stock-photo-wild-brown-hare-with-big-ears-sitting-in-a-grass.html

But you can simply replace that title part with anything:
http://www.shutterstock.com/pic-363531434/shutterstock is great.html

Image ID is the only important thing here...


right - the title desc is never used - you can bring up the image with just the pic-xxxx

so looks like this is SS trying to get google attention with a descriptive url that doesn't exist

(also, looking at the source, they're still using meta keywords which google abandoned years ago)


 

Related Topics

  Subject / Started by Replies Last post
3 Replies
3132 Views
Last post March 06, 2008, 13:33
by Waldo4
2 Replies
5006 Views
Last post April 03, 2010, 17:05
by louoates
11 Replies
5241 Views
Last post June 29, 2012, 21:29
by ComfortEagle2095
8 Replies
3155 Views
Last post January 01, 2013, 01:01
by tab62
1 Replies
2580 Views
Last post May 28, 2013, 12:44
by cascoly

Sponsors

Mega Bundle of 5,900+ Professional Lightroom Presets

Microstock Poll Results

Sponsors