Bring Back Youtube Dislike Counts
The debate is finished. The distribution of the voting points and the winner are presented below.
After 2 votes and with 2 points ahead, the winner is...
- Publication date
- Last updated date
- Type
- Standard
- Number of rounds
- 4
- Time for argument
- One week
- Max argument characters
- 20,000
- Voting period
- One month
- Point system
- Multiple criterions
- Voting system
- Open
Resolved: On balance, Youtube should bring back its Dislike Count display feature (showing how many people disliked a video).
What does this mean?
Please read the November 10, 2021 blog update to youtube dislike count [https://blog.youtube/news-and-events/update-to-youtube/].
If for some reason during the debate, Youtube has indeed brought back dislike counts, the resolution shall be interpreted as Maintaining the dislike count display feature. Either way, PRO is advocating for showing the number of people the clicked dislike on a video, while CON is advocating for Dislike count to be hidden.
BoP is shared.
- If you dislike a video (as in you actually don't like what is inside the video) it's your freedom to have that opinion without any influence or subtle pressure to conform to others disliking it or not disliking it.
- If you dislike a video and it's force-displayed unless all ratings are hidden, there is shaming likely to happen to the channel owner, which both financially and in terms of creativity limits the variety on YT. If enough dislike a new uploader's early video, it doesn't get recommended to others but what's likely to happen is if 2 or more disliked it, others will join the bandwagon if they even slightly find a reason to dislike it without watching the rest of it, assuming the first 1 or 2 were 'right to dislike it'. This is why smaller content creators are affected badly, since very few need to dislike it before it's barely recommended to anybody. This then demotivates them to try and to get advertised on YT, let alone be a creative contributor to it as opposed to a conformist.
- There is no uploader who wants to flex the amount of dislikes they have, generally speaking. It's true though that higher like vs dislike ratios fulfil some but see point 2 about the issue of it (point 2 isn't the same as point 3, I'm talking about displaying it here, not just it pressuring others to dislike).
- The counter isn't deleted entirely, content creators get to analyse and properly assess whether or not they care about the dislikes and what type of content people dislike more, this adds genuine credibility to the dislikes as it couldn't have been motivated to shame them or sheepishly mimic others that disliked.
- If you dislike a video (as in you actually don't like what is inside the video) it's your freedom to have that opinion without any influence or subtle pressure to conform to others disliking it or not disliking it
- Con asserts that your "freedom" not to conform to others is important, yet this already exists there. There is no reason why allowing you to see how many people formed a negative opinion stops you from expressing your opinion. By contrast, Without the dislike counter visible, you *prevent* that opinion. Remember how I stated what if Amazon hid all one star reviews. Are you *expressing* your opinion if it is hidden? Of course not. There does not seem to be a peer pressure for more people to deliver one star reviews -- they are merely showing the product low quality. Unless Con can show a powerful negative effect from showing the dislike button, this is a poor assertion with no sources for support.
- If you dislike a video and it's force-displayed unless all ratings are hidden, there is shaming likely to happen to the channel owner, which both financially and in terms of creativity limits the variety on YT. If enough dislike a new uploader's early video, it doesn't get recommended to others but what's likely to happen is if 2 or more disliked it, others will join the bandwagon if they even slightly find a reason to dislike it without watching the rest of it, assuming the first 1 or 2 were 'right to dislike it'. This is why smaller content creators are affected badly, since very few need to dislike it before it's barely recommended to anybody. This then demotivates them to try and to get advertised on YT, let alone be a creative contributor to it as opposed to a conformist.
- Con tells us that people would cause low viewed channels to result in disliked easily and affected with less recommendation. Yet he shows us zero examples of this. On the other hand, there are countless bad tutorial videos that now can easily slip by with no ability to give feed back. For example, The youtube video telling us how to switch from Static to dynamic IP Address has 600 likes with 250,000 views. [https://www.youtube.com/watch?v=vGAW9ErlarU&ab_channel=GoodFightConnection] All the comments are quite positive, and on the surface this looks pretty decent. However, these comments were all 5 years ago, far before the dislikes were hidden. If we take a more careful look, Reddit tells us that all these comments are sarcastic and that this is a horrible tutorial [https://www.reddit.com/r/videos/comments/56au88/i_may_have_just_found_the_worst_video_on_youtube/]. The man only showed his own IP address and did almost nothing. He spent a lot of time accessing his home router without explaining the title at all. As you can see I showed a clear example of a bad tutorial example where you could potentially waste 6 minutes. Archon with 355 upvotes lists:
- Long intro, displaying the work "hacking". Good start.
- Does things in a roundabout way, possibly because of lack of planning ahead of time.
- Constant panning and zooming makes it increasingly difficult to read what is typed or clicked. Keeps you guessing what is happening.
- Misinformation.
- Task in the video title is not completed. To be honest the video is almost completely irrelevant to the task.
- Funky song.
- There is no uploader who wants to flex the amount of dislikes they have, generally speaking. It's true though that higher like vs dislike ratios fulfil some but see point 2 about the issue of it (point 2 isn't the same as point 3, I'm talking about displaying it here, not just it pressuring others to dislike).
- The counter isn't deleted entirely, content creators get to analyse and properly assess whether or not they care about the dislikes and what type of content people dislike more, this adds genuine credibility to the dislikes as it couldn't have been motivated to shame them or sheepishly mimic others that disliked.
- The dislike-counter never displays/displayed on the screen before you had opened the video and given it a view anyway meaning that in terms of what you see (views and like-count) nothing has actually decreased in one's ability to judge a video pre-opening it.
- The fact that when you open a video, the dislike counter showed, could bias you against or for the video if more vs less dislikes were there but art that irritates more people isn't inherently inferior, this is art we are talking about after all and the height of creativity offends and irritates as much as it pleases, in fact this is an inherent problem with 'dislike and like' counters in general in matters of art.
- By removing the dislike-counter displaying but maintaining it being shown to content creators as well as being used in the recommendation algorithm, the most superficial part of it has been removed, increasing the percentage of meaningful elements of disliking that matter. The urge to join a dislike-bandwagon to hurt the feelings of the creator are reduced and indeed as Pro says, people are more motivated to comment what it is about the content that they dislike.
- People want to be able to dislike videos
- People want to know 'bad' vs 'good' content as rapidly as possible into bad content, so seeing a discount liker that tells them people dislike the video to a certain proportion, securely tells them that they/he/she as an individual, will dislike the content and saves them time viewing it.
- It's dishonest because it's letting content that's 'bad' get away with being bad.
As part of this experiment, viewers could still see and use the dislike button. But because the count was not visible to them, we found that they were less likely to target a video’s dislike button to drive up the count. In short, our experiment data showed a reduction in dislike attacking behavior1. We also heard directly from smaller creators and those just getting started that they are unfairly targeted by this behavior — and our experiment confirmed that this does occur at a higher proportion on smaller channels.Based on what we learned, we're making the dislike counts private across YouTube, but the dislike button is not going away. This change will start gradually rolling out today.What’s changing for creators and viewers starting today
Creators will still be able to find their exact dislike counts in YouTube Studio, along with other existing metrics, if they would like to understand how their content is performing.We want to create an inclusive and respectful environment where creators have the opportunity to succeed and feel safe to express themselves.Viewers can still dislike videos to tune their recommendations and privately share feedback with creators.
We heard during the experiment that some of you have used the public dislike count to help decide whether or not to watch a video. We know that you might not agree with this decision, but we believe that this is the right thing to do for the platform.
We want to create an inclusive and respectful environment where creators have the opportunity to succeed and feel safe to express themselves. This is just one of many steps we are taking to continue to protect creators from harassment. Our work is not done, and we’ll continue to invest here.
1 Analysis conducted July 2021
Purpose of Constructive FeedbackThe purpose of constructive feedback is to give feedback to an individual in a way that will lead to improvements or corrections. This is important, as it enhances personal and professional growth in individuals.For example, constructive feedback can:
- Improve employee morale
- Reduce confusion regarding expectations and current performance
- Provide a new perspective and give valuable insight to the person receiving feedback
- Positively impact an individual’s behavior
okay
Unlike the other big dogs, I have experienced the pain of what I feel is undeserved defeat over and over. This is but part of the battle-hardening necessary for a champion who isn't planning to run 98-1.
As for the job interview, I don't believe in wishing good luck if I don't need to. For all I know it's a job that isn't the one you end up enjoying vs other potential timelines. So, I wish you to do your best and endure defeat like I am here or win like a champion.
I have a job interview I am prepping for today, so I will not have time to vote.
gugigor states that Con's argument falls apart because he has not given one single example, study, expert, news, about people banding together to dislike just because they dislike. Actually, he got Ragnar to remove my vote to show people banding together to like just because they like.
Thank you for voting
This is not a huge issue and I did not take it into consideration in my RFD, but you should sign post to make the debate a little easier to read and evaluate.
**************************************************
>Reported Vote: FLRW // Mod action: Removed
>Voting Policy: info.debateart.com/terms-of-service/voting-policy
>Points Awarded: 6 to con
>Reason for Decision: See Comments Tab.
>Reason for Mod Action:
Basically you're crossing the line into vote bomb territory, as if trying to outweigh another vote without giving reasons for the assignments (doubly so for conduct, as the person you indicated had vastly superior conduct forfeited a round). Further, a vote really should be more than just quoting one of the debaters.
To cast a sufficient vote, for each category awarded, a voter must explicitly perform the following tasks:
(1) Provide specific references to each side’s utilization within the said category.
(2) Weigh the impacts against each other, including if any precluded others.
(3) Explain the decision within the greater context of the debate.
https://info.debateart.com/terms-of-service/voting-policy#casting-votes
**************************************************
FLRW
Added: 2 hours ago
#2
Criterion
Pro
Tie
Con
Points
Better arguments
3 point(s)
Better sources
2 point(s)
Better spelling and grammar
1 point(s)
Better conduct
1 point(s)
Reason:
Pro says Con's argument is growing weaker by each round and it is very illogical. Con actually counters this by a number of statements including this one: I countered pillar 2 with the idea that not only has the recommendation algorithm remained in place (and it takes dislikes into account) but that it's often 'smarter' about an individual's taste than they expect. In other words, to put my previous 2 Rounds together, I am asserting that there is such a thing as content that people will like that many disliked, which they cannot know they'd dislike for sure without viewing it. The only real thing influencing them seeing it faster and more expectant of high content should be either that it had many views and likes or that they got recommended it via the YouTube algorithm.
In case I come back to finish voting on this, here is the start to a review (read into R3).
1. Audience review -- Video quality
Pro argues that the ratio serves as an effective review, which the absence can be harmful for tutorial videos.
Con adds that the dislikes were already hidden before opening the video, so very disliked videos still get viewed; and further that videos are art, so we shouldn't base our opinion on it on a bandwagon. And finally that the increased engagement of negative comments (which unless I'm mistaken can be upvoted) toward bad videos are encouraged by hiding the main dislike counter (pro has pre-defended this by pointing out that creators can disable comments).
Pro asserts that the dislike ratio defines if a video is good or not.
2. Encouragement of Bad Content
Pro seems to propose that any video with too many dislikes should be automatically deleted, to save YouTube employees from having to manually review them.
3. Youtube is a hypocrite
This gets immediately muddled, as pro accuses YouTube of removing it to hide their own shame at a disliked video, and further that the existence of dislike attacks is a myth. Further they apparently already brought it back (which would invalidate the resolution, since they can't bring back what they already brought back).
Pro repeats his request for proof.
4. Peer Pressure
Con asserts we shouldn't be pressured into disliking a video.
Pro compares watching YouTube videos to purchases on Amazon (and claims to have done so in R1, which I am not spotting).
5. Recommended videos
Con asserts a band wagon effect of dislikes, which unfairly prevents videos from being suggested to other potential watchers.
Pro repeats his request for proof, and tying to point 1 shows a bad tutorial video.
6. Uploaders don't want to flex dislikes
Enough said.
7. Uploaders can still see the count
Con asserts that it is made more accurate due to not having the bandwagon effect, which can in turn still motivate them to make better videos.
Pro asserts that a more powerful motivator would be the public shame.
that vote will never hold up but thanks... should just award arguments next time if you meant the vote.
YT's new policy comes after Google’s video service launched an experiment earlier this year to see if removing the count would protect content creators from harassment and campaigns that purposely drove up dislikes on a clip.
“In short, our experiment data showed a reduction in dislike attacking behavior,” the company said.
Too strong, lol. I see.
Also, about the rating, chill out. Theres more to life than debates
Not exactly, It's just that the Amazon analogy was too strong of a point to go unaddressed, so thats one. Im not saying it was the sole point switch on arguments. If you really want I can chnage arguments to TIE, but it just doesn't seem right.
I am curious, if I had pointed out how the Amazon thing was irrelevant, in Round 4, would you suddenly say I won the debate?
it is bad conduct for me to bring up brand new points about the difference between amazon rating system in a round so late into the debate.
I didn't want to dig into an entirely irrelevant point but sure, you got the grudge out of your system. Just know that the lower that my rating is when you lose against me, the bigger the hit to your rating.
Round 3:
PRO goes rapid-fire weaving through CON's round two.
He counters by saying the audience can tell if a video is good or bad with the dislikes, once again points out that CON has not given any examples of targetted dislike campaigns or dislike trends on lower creators (CON seriously should provide this evidence because it ha been mentioned multiple times and the readers pick up on this) points out that CON failed to address his bad content argument. He poses a question: Should Amazon remove showing how many people rated the product 1 star.
CON then makes a sort of strawman of PRO's argument by saying "Pro's case is based on the following pillars: people want to be able to dislike videos".
He argues that the recommendation algorithm remains, and takes dislikes into account, so they still affect the video.
CON then posts a youtube blog source that shows YouTubes' experiment that concluded with dislikes hidden, people were less likely to dislike the video. CON says if you don't trust the source, it's a "conspiracy theory", and you must justify it. CON makes the case that constructive feedback in the comments is incentivized and more helpful/beneficial than dislikes
CON does not answer the question of whether Amazon should remove one-star ratings.
Round 4:
PRO rounds off. He points out CON's assessment of the rating system isn't exactly true and video with many dislikes can be relatively near the top base don views. He uses his specific example for this. He points out CON has not addressed the waste of time that one can experience watching a bad video unknowingly while visible dislikes make one more cautious and aware. He amusingly argues that CON says dislikes encourage bandwaggon negative feedback "just because" as CON still has provided no source that illustrates this. He points out that CON has not addressed, nor shown any reason why, given what he said, amazon should not remove one-star ratings. CON forfeits round 4.
Arguments: PRO
CON dropped the amazon analogy and failed to provide any evidence of a source that corroborates the idea of dislike bandwagoning. It's not just a one-time thing, CON was asked many times to the point where it kind of looked bad. PRO showed a specific example, so he has the upper hand in specifics. CON didn't really address youtube being unable to fully regulate bad content, and PRO provided specific evidence of this, but PRO does not exactly show how having visible dislikes somehow regulates bad content. Regardless, PRO had a counter for everything CON argued, yet CON dropped a few arguments that were pretty strong points against him.
Sources: I'll give it a tie. However, CON did not provide a source that showed the dislike bandwagon effect.
Grammar: Tied, no significant difference.
Conduct: PRO gets conduct due to CON forfeit.
Round 1:
PRO makes basic arguments;
1) Removing dislikes affects people's ability to gauge th usefulness or reliability of a video.
2) Encourages bad content
3) Shows YouTubes hypocrisy by removing dislikes when they have the most disliked video on their own site
PRO doesn't respond just yet, RM saves that for the next round. Rather he makes his case, youtube is entitled to remove dislikes
1) People have the freedom to form opinions regardless of subtle pressure amount of dislikes
2) It negatively affects smaller creators
3) No one wants to flex their dislikes
4) the counter still exists but is simply hidden, and creators can see feedback and improve.
Round 2:
PRO attacks CON's case arguing that there isn't any reason as to why seeing others' negative opinions stops you from forming yours. PRO makes a strong point here "Are you *expressing* your opinion if it is hidden?". PRO uses an analogy to compare this to amazon hiding one-star reviews. If you express an opinion that their product is one star, and it is now shown, then is it really expressed?
PRO points out that CON provided no examples of lower channels being attacked with dislikes. On the contrary, CON provides a specific example of a video that could slip by this new policy and waste people's time, when they can't see previous disapproval.
PRO also counterargues that creators can still see dislikes, but that isn't as good motivation to improve as having them visible publically as if dislikes are hidden, creators will have no significant incentive to improve.
CON argues "Pro's case is partly based on the idea that currently, you cannot dislike a video" which CON has established is false, but PRO just previously acknowledged that the dislike counter still exists and is hidden, so this is technically false. Regardless, CON goes on.
CON argues that
1) you have given the video a view before even seeing the dislikes so nothing has decreased one's ability to judge a video BEFORE opening it. 2) CON argues that dislikes can induce bias
3) CON argues the incentive for dislike bandwagons is reduced, but again provides no sources or specific examples of these incidents as PRO pointed out previously.
CON argues that motivation to comment constructive criticism is a good thing.
pls vote
are you willing to vote on this?
If anyone needs votes, tell me
I just realized I can vote now, and I will indeed vote
vote please
vote pls?
Gugigor definitely won this debate, at least from a rhetorical standpoint that is. So because of this I have to give my vote to him. Also, doing an ad-hominem attack against your opponent doesn't work. Calling them a "conspiracy theorist" won't just discredit them. So be a bit more careful when doing such a thing. Oh wait, I can't vote lol.
voting time.
Source of the video being 7th place: http://prntscr.com/26hqlgw
I like looking at dislikes, because a lot of times it alerts you to a title that is just click bait.
Is CON arguing that not having dislike counts can still be coped with or is it superior to having dislike counts?
Done.
Well, in a sense, they never really even left with such a feature. I mean, you can still hack into the code(which is probably on the surface web) and see the count. Youtube just didn't actively show it. Passively showing it is still displaying it.
Increase the character count to 20k or thereabouts and I'll accept.
I will run an interesting Kritik as well as good rebuttals to what you bring forth.
While dislikes can be misused, a display of their number can be very informative. I defend YouTube's right to change the display, and it may work out for the betterment of their site, but short term it seems a step backward.
thoughts?
I agree wholeheartedly. Frankly, it seems like a means of silencing dissent. However, this change has led to a few things I find festinating. Since people who usually dislike no longer feel their opinions are heard in that respect, they more readily state them in the comments. Also, people will be less likely to watch channels with a high conflict personality since the dislike button often creates its dopamine-inducing feedback. This means that these abusive channels will no longer get the same amount of view and therefore fade into obscurity. However, these are just a few positives in a vast sea of negatives regarding the new update; therefore, I can't imagine taking you on. Thanks.
I agree. Personally, it always vexed me that individuals would dislike a video without so much as submitting a reason, but on the other hand, the like to dislike ratio has always served a purpose of identifying "fake" videos. I'm sure many a self-esteem can survive a few dislikes here and there.