Thank you 🙏🙏🙏🙏🙏

[MelissaBaby]

Emily's point is that it cleared a huge number of votes not cleaned by the normal sweeps. Maybe this is a new form of sweep they are testing. It does seem to have been limited in scope, so I tend to believe they did look specifically at some people's stories, including Emily's.

Laurel started sweeping my two most recent stories on Sunday. It was still ongoing until yesterday. I don't know any details of how sweeps work internally, but it would not have taken that long to sweep two stores. Clearly she was making a much bigger sweep, that may of may not have included new methodology, and as PSG pointed out, lots of people are reporting they've had stories swept.

 

[lovecraft68]

I've never asked for a sweep because begging for one is past sad to me.

 

[MillieDynamite]

Maybe they've swept the whole thing under the carpet.

 

[old_prof]

It's not limited in scope. This was a massive site wide sweep.

I didn't understand that it was. I have heard more people say that they were unimpacted by it then said they were impacted by it.

 

[SimonDoom]

Emily can certainly say this: she was right. The massive deletion of one votes proves she was right that something improper was going on, and this wasn't just random, normal downvoting.

So, even if the relief is temporary, I'm glad it came.

 

[used2bjustj]

The problems I have been experiencing with industrial carpet bombing on my latest works have been resolved (at least for now, I’m not that naive!)

Since some time yesterday afternoon, I’ve had 160 - one hundred and sixty FFS - malicious votes removed from my most recent seven stories. As best I can figure (and I can figure pretty well): 143 x 1️ and 17 x 2️. That’s 21% of the votes previously cast.

ONE IN FIVE VOTES WAS MALICIOUS!

That’s even higher than I thought.

I had contacted the site and - with her permission - copied @MelissaBaby. Maybe she was my lucky charm as this note led to very extensive action.

I’d like to thank Melissa, I’d like to thank @Cacatua_Galerita who came up with a data visualization that seemed to explain my issue really well. I’d even like to thank my troll for making their activities so blatantly obvious in the last two days (17 malicious votes FFS).

But I’d mostly like to thank @Laurel and @Manu for acknowledging this problem and acting.

I have spoken to some authors who saw mild changes in their portfolio as well. Perhaps that is total coincidence, but my hope is the that site was trialling a more robust approach to sweeps; one that may catch more of the modern trolling. I hope so.

But anyway, I’m really pleased (understatement) to have had this response. Putting a year of effort into something - and pouring your soul into it - only to see it trashed is an appalling feeling. I wouldn’t wish it on anyone.

Happy girl today .

If I remember right I read in her profile that author Sandy_Paris also had a similar problem. Her 4.9 story was now a low 4.26. It's really a shame that some bot or person is doing this to some of the best authors here. I'm glad you were able to get the problem fixed. For now.

 

[SkyBubble]

The problems I have been experiencing with industrial carpet bombing on my latest works have been resolved (at least for now, I’m not that naive!)

Since some time yesterday afternoon, I’ve had 160 - one hundred and sixty FFS - malicious votes removed from my most recent seven stories. As best I can figure (and I can figure pretty well): 143 x 1️ and 17 x 2️. That’s 21% of the votes previously cast.

ONE IN FIVE VOTES WAS MALICIOUS!

That’s even higher than I thought.

I had contacted the site and - with her permission - copied @MelissaBaby. Maybe she was my lucky charm as this note led to very extensive action.

I’d like to thank Melissa, I’d like to thank @Cacatua_Galerita who came up with a data visualization that seemed to explain my issue really well. I’d even like to thank my troll for making their activities so blatantly obvious in the last two days (17 malicious votes FFS).

But I’d mostly like to thank @Laurel and @Manu for acknowledging this problem and acting.

I have spoken to some authors who saw mild changes in their portfolio as well. Perhaps that is total coincidence, but my hope is the that site was trialling a more robust approach to sweeps; one that may catch more of the modern trolling. I hope so.

But anyway, I’m really pleased (understatement) to have had this response. Putting a year of effort into something - and pouring your soul into it - only to see it trashed is an appalling feeling. I wouldn’t wish it on anyone.

Happy girl today .

Glad you got it taken care of.

 

[MrPixel]

I didn't understand that it was. I have heard more people say that they were unimpacted by it then said they were impacted by it.

It was. I don't keep track of scores in much of my standing library, but I did notice that all of a sudden I have a bigger list of 'H' stories. I'll take it, frankly.

I had protested previously about this nonsense - I believe I'm the one who coined "score vandalism" - and I'm glad that the powers-that-be are stepping up to the plate. Kudos to @Laurel and @Manu, and kudos to @EmilyMiller for her persistence and knowing analysis of the problem.

 

[EmilyMiller]

If I remember right I read in her profile that author Sandy_Paris also had a similar problem. Her 4.9 story was now a low 4.26. It's really a shame that some bot or person is doing this to some of the best authors here. I'm glad you were able to get the problem fixed. For now.

Thank you. It’s seems likely sadly that high-scoring stories are triggering for trolls.

I’ve used a climbing analogy before. I’ve met some of the world’s best climbers, I’ve had group and one on one coaching with two different ones. They are super keen to help beginners like me to enjoy the activity they love and amazingly encouraging.

The only climbers I’ve had any issues with are those who are kinda OK, and super pissed that new climbers are quickly approaching their proficiency. That’s when you get comments about it being unfair that I have small fingers (like being 5’1” is a big advantage climbing ), or problems (climbing speak for short climbs) are ‘girl problems.’ It’s only a small minority, but it could get nasty.

I’ve found it the same way with authors. The really good ones are also super friendly and supportive. A tiny minority of the middling ones who desperately want to be better than they are? Look out for them.

 

[RejectReality]

The troll probably just got overzealous and dumped too many bombs in too short of a time, in gluttonous response to the sweet, sweet tears they were getting fed — thus exposing the whole slate. That then spilled over into everyone else who was getting bombed from the same source.

There's no such thing as a sweep-proof or a sweep-resistant vote. It's whether the asshole's activity has fallen into scope of the examined dataset, and whether that dataset is large enough. That's all. Asking for your story/stories specifically to be swept is unlikely to result in anything beyond the most basic of dumb bombs/boosting getting removed, because the dataset is too small for patterns emerge unless the troll makes a stupid mistake. ( which they all almost always do eventually.)

 

[EmilyMiller]

It was. I don't keep track of scores in much of my standing library, but I did notice that all of a sudden I have a bigger list of 'H' stories. I'll take it, frankly.

I had protested previously about this nonsense - I believe I'm the one who coined "score vandalism" - and I'm glad that the powers-that-be are stepping up to the plate. Kudos to @Laurel and @Manu, and kudos to @EmilyMiller for her persistence and knowing analysis of the problem.

Thank you

 

[ElectricBlue]

I didn't understand that it was. I have heard more people say that they were unimpacted by it then said they were impacted by it.

Maybe all that means is, they hadn't been trolled.

My belief, reinforced by this and that over the years, is that every sweep is site wide, and goes back at least a decade, even longer. Part of the algorithm must be pattern and repeated behaviour matching (amongst other obvious things), and a larger dataset has to better for that than a small one.

 

[old_prof]

Maybe all that means is, they hadn't been trolled.

My belief, reinforced by this and that over the years, is that every sweep is site wide, and goes back at least a decade, even longer. Part of the algorithm must be pattern and repeated behaviour matching (amongst other obvious things), and a larger dataset has to better for that than a small one.

At the minimum, I was trolled. I received a few dozen 1's around midnight a few weeks ago. None of which were cleared in the sweeps. I had four storiesdrop a half point in rating or more in roughly an hour. Stories that had seen no votes at all in weeks. I did send Laurel and Manu a note about the suspicious voting. But nothing has happened to them as of yet.

 

[MelissaBaby]

This is what a crazy roller coaster this has been. My Valentine's Day story, which a couple of days ago had a rating of 4.44 is now #1 on the All Time Lesbian list.

 

[MelissaBaby]

Oh, and #5 all time for all stories,

 

[ElectricBlue]

At the minimum, I was trolled. I received a few dozen 1's around midnight a few weeks ago. None of which were cleared in the sweeps. I had four storiesdrop a half point in rating or more in roughly an hour. Stories that had seen no votes at all in weeks. I did send Laurel and Manu a note about the suspicious voting. But nothing has happened to them as of yet.

Cross fingers and hope it's a progressive thing then. I should clarify my comment - over the years I've seen sweeps go site wide, but "as yet" might be the key phrase.

A really deep clean several years ago took several passes over a number of days. I can't recall if it was associated with a Contest or not, or "fixing" a site fix that didn't go quite right.

The moans when that one went through were mostly, "How come all my brilliant fives disappeared?" Some folk tried to point out, "Yep, along with all those dastardly ones," but irony goes missing when you're aggrieved.

 

[Voboy]

Subscribed, mostly because I'm curious how all this works; I am no expert. FWIW, I don't see any changes to any of my stories, but I stop keeping careful track of new ones after a few weeks.

Whether this was the result of Emily's entreaties or whether it was already going to happen anyway, it's good that we're being told about it by someone invested. Since the site will never let us know.

I'd be very interested in seeing monthly sweeps come back, like they used to for the contest. That seemed to keep things more or less stable. For the life of me, I can't imagine why they don't keep running the sweeps even if they decided to end the contest.

 

[MelissaBaby]

Subscribed, mostly because I'm curious how all this works; I am no expert. FWIW, I don't see any changes to any of my stories, but I stop keeping careful track of new ones after a few weeks.

Whether this was the result of Emily's entreaties or whether it was already going to happen anyway, it's good that we're being told about it by someone invested. Since the site will never let us know.

I'd be very interested in seeing monthly sweeps come back, like they used to for the contest. That seemed to keep things more or less stable. For the life of me, I can't imagine why they don't keep running the sweeps even if they decided to end the contest.

I think that what happened in this case was that I asked Laurel for a sweep on Candy Season, because it had dropped from 485 to 4.44 in just a few days. I started to notice changes in the score on that, and on my other new story on Sunday, just a few at a time at first. It took most of 4 days before the sweep ended on just those two stories

When Emily asked for a bigger sweep on her stories, and copied me as sort of a reference that this was not a fluke, but a real, ongoing issue, I suspect Laurel realized we need a site wide sweep.I can't even imagine how long that might take, but I would expect there are a lot of people who haven't seen any changes because they just haven't come up in the queue yet.

 

[SimonDoom]

I saw a small but noticeable uptick in the mean score for all my stories, so I assume the Site did a comprehensive sweep.

 

[RejectReality]

Near as I can tell, I had one lose one vote and go up an additional .02 that had recently lost 2 votes and went up .04. If anything else got swept, it was lost in the noise.

 

[EmilyMiller]

This is what a crazy roller coaster this has been. My Valentine's Day story, which a couple of days ago had a rating of 4.44 is now #1 on the All Time Lesbian list.

View attachment 2601173

Congratulations - I think we both know what happens next

 

[EmilyMiller]

FWIW - I’d like to see a monthly sweep including whatever they did differently to catch the 160 malicious votes on my stories which were caught by neither the Winter Holdays nor Valentine’s Dat sweeps.

IIRC the Halloween Sweep was a total non-event (I was running the official thread for that and Valentine’s).

 

[MelissaBaby]

Congratulations - I think we both know what happens next

That's why I always get a screenshot to memorialize the occasion. Queen of the Roller Derby hit the 100 votes mark with a 4.99 score, and was #1 all site, all time. I took a screenshot, went to bed and in the morning, the score was 4.86. But I got my screenshot for a nice little trophy.

 

[MelissaBaby]

FWIW - I’d like to see a monthly sweep including whatever they did differently to catch the 160 malicious votes on my stories which were caught by neither the Winter Holdays nor Valentine’s Dat sweeps.

IIRC the Halloween Sweep was a total non-event (I was running the official thread for that and Valentine’s).

I had no change in my vote totals or score leading up to the Halloween deadline. I mean, no votes at all, no sweep at all.

 

[ShelbyDawn57]

FWIW - I’d like to see a monthly sweep including whatever they did differently to catch the 160 malicious votes on my stories which were caught by neither the Winter Holdays nor Valentine’s Dat sweeps.

IIRC the Halloween Sweep was a total non-event (I was running the official thread for that and Valentine’s).

I’d like to see the logic for the sweep as a proactive filter.