Imagine Stack Overflow only having the option to upvote Questions and Answers. The ratio of likes vs dislikes of a TED talk does help me to decide if I should stop watching a video which just isn't getting better.
While if it's a video of a political debate I would expect there to be a ton of dislikes by default, so I suppose it wouldn't matter much in that scenario.
And in howto/explanation-videos, if dislikes aren't shown, there's no signal for people telling them whether or not it's a good explanation.
I suppose you could devise a dislike count yourself by calculating the likes/views ratio. But why... people are deliberately putting their content up publicly for display and as such are opening themselves up for scrutiny/criticism/dislikes. It can't be all praise and hugs and kisses. That's how we treat children.
I sometimes look for obscure how-to's, and I get garbage videos not even answering the question. If YouTube can get rid of all that garbage, then not having a dislike button is fine.
The problem with Youtube as I see it is that these comments never rise to the top. It's always compliments and ass-kissing and high-fives that make it there.
Twitter tends to gravitate towards the opposite.
It's all fascinating though.
That's because they have engineered their comment system in a way that comments the AI perceives as negative are shadow-banned or buried deep down so people don't get a chance to read them.
In the 90's there was an add-on that allowed anyone to comment on any webpage: http://alumni.media.mit.edu/~orit/utok.html
Although imagine having such a tool nowadays and opening cnn.com, it'd be like walking into a MAGA rally.
Steam (a video game store) does this already. They categorize reviews as either being recent (made in the last 30 days) or all-time. If a game has excellent all-time reviews, but mixed recent reviews, then you can suspect something has recently upset the playerbase.
It is low signal all the way through, and relegating likes and dislikes to a signal for recommendations is wise.
Long-time Netflix subscribers might be having déjà vu right now. Old Netflix had a ratings system, and users could filter content by average rating and see the top-rated movies and shows. Netflix later replaced that system with a recommendation engine. Not only does the recommendation engine not get my taste right, but it sure seems to recommend to me a lot of Netflix Originals.
The steam review graph is a fantastically useful tool for users with a huge SNR and I don't understand why you think otherwise.
https://www.eurogamer.net/articles/2018-09-25-total-war-rome...
"Total War: Rome 2 has suffered a Steam review bombing run over women characters and a recent update - but it turns out the game is working as intended.
"Creative Assembly's PC strategy game, which came out in 2013, saw hundreds of new negative Steam reviews this week over the frequency with which women generals show up in the game and related claims about historical accuracy."
and another one:
https://www.techspot.com/news/91212-life-strange-true-colors...
"The reason behind the Chinese users' anger is the inclusion of a symbol banned in their country. On the main street of the game's Haven Springs setting is a shop called Treasures of Tibet, which displays a Tibetan flag prominently outside the store."
Valve has come out on more than one occasion to let customers and studios know that they are constantly working to mitigate brigading.
EDIT: While not the norm, as you were alluding, gamer toxicity and brigading can also be relatively mainstream. A popular website among such circles is
https://www.oneangrygamer.net/
Reading the posts and comments can be quite revolting.
Really though, my main point I actually care about making is that the steam review system is far superior to a simple like/dislike system which YT implements.
At the bottom of your first article:
> Meanwhile, Total War: Rome 2 has an "overwhelmingly negative" recent reviews rating, with a "mostly positive" overall rating.
Ahh, so it seems the steam rating system is working! Even though these games got "brigaded", users are still able to see the overall and historical ratings to get a better picture of what's going on.
This actually lets you see the rating before the brigade started. And if the game is actually good, the brigaded ratings will amount to a blip amongst the other reviews. Case in point, here is the current steam page for R:TW2, with "very positive " overall rating: https://store.steampowered.com/app/214950/Total_War_ROME_II_...
-- Anecdotal evidence doesn't mean that it didn't happen. My personal experience of using steam since its inception (HL2 launch on Steam, 2004), so roughly 17 years, is that I've come across this on a more than a few occasions. I tend to collect digital goods in a hoarding method, so I'm up to over 3700+ games in my library and I check a lot of them out when I can, even if I have no time to really play them anymore. The firs thing I read?
REVIEWS!
As mentioned - this is not the norm, and I agree with your general premise, but please don't dismiss what actually happened and what I witnessed (no, I don't have screenshots because I wasn't documenting it happening in real time when it occurred). The examples I brought up were illustrative of some gamers' groupthink mentality (including oneangrygamer.net, which has in mid-2021 disabled comments because they were ridiculously toxic - misogynistic, sexist, homophobic and racist)
Also - I'm not OP, FYI. I was backing OP's point.
I can't be bothered to respond to the rest of your post, unfortunately, because it comes across as vapidly dismissive.
Look up GamerGate if you'd like more information on brigading and misogynistic groupthink among related sub-communities.
What you are thinking of is metacritic where anyone can post, it is very prone to brigading, but steam reviews isn't.
[1]: https://81m.org/
Archive: https://web.archive.org/web/20211023061808/https://81m.org/
Say for example I used a batch of 100 HN accounts to mass downvote your comment. After a while an hourly job runs that looks at all those votes in aggregate and determines they were coordinated (for simplicity sake they all came from the same IP) and removes them. You'd see a large shift in the net score of your comment all of the sudden. This isn't HN manipulating anything, it is them doing their job to prevent abuse on the platform.
I think that unintentionally removes down votes on obviously bad videos that people do not waste much time watching.
For instance, an hour-long video claiming to be what you're looking for, but actually consisting of a still image and a URL to a scam/spam site.
I've seen dislike attacks happen. And if it takes a non-trivial procedure to make them count, people will document that procedure. For instance, hypothetically:
"Alright everybody, here's the link. Remember, mute the tab right away but don't mute the video itself; wait 2 minutes (the timer that starts when you click the link will go green to let you know), then click the dislike button."
I've seen much more complex instructions offered as part of gaming a poll, as well as sites built to help semi-automate or simplify the process.
The key distinction that would be useful for YouTube to measure: did you encounter the video and then dislike it, or did you visit a video you were referred to for the sole purpose of disliking it? I don't think hiding dislike counts serves that purpose, though.
Dislikes on YouTube and similar platforms don't have any cost or barrier to entry.
The same is simply not true of YouTube or any other kind of social media upvote/downvote system - anyone can sign up for an account, and their votes immediately count and have equal weighting.
The same is not true of Stack Exchange or HN, where you have to put a lot of effort in with a community to be able to downvote something.
I don't think this is true for YouTube - you need a Google account, and you can't create a Google account without verifying your phone number. (Unless you can create numerous accounts with one phone number? But that seems like an easy loophole to close.)
I've argued against it in the past (since it's very important context to any reader) but to no avail. An upvote and a downvote rarely cancel each other out. They have different units and shouldn't be summed together. A score of 25 could mean the answer is perfect but has only seen moderate traffic, or incredibly dangerous but has seen high traffic.
Here's a good list of controversial answers on SO: https://data.stackexchange.com/stackoverflow/query/466/most-.... They would mostly look like great answers from the net score.
I agree, this "toxic positivity" can't be good for society in the long term. The continued drive towards stupidity and unquestioning, docile populations is going to slowly turn into a massive dystopia.
There is a red line between caring for the others, and what is nothing more than plain censorship via "toxic positivity".
Hiding dislikes is something Brands want.
That's the word. It isn't really "likes" in the sense that we normally think of it. It's ticks on the relationship of "engagement" to "dollars".
IIRC, Facebook was pretty upfront about this being why they didn't implement a "dislike" button; the information wasn't valuable to them.
I honestly don't know if it's a generational thing or overall cultural shift, but it feels like negativity (at least in the workplace is where I've seen this) is either less tolerated or less shared. There's a bias towards being positive about everything, even for things that aren't... great. It feels like self censorship, and this isn't even for things that are political, just subjective.
Keep the actual buttons, as they signal to YouTube what you do and do not want to see in future, but the actual counts only have use to the creator... and even then a simple ratio or percentage alongside the video in the Creator Studio is enough; there's no need for actuall numbers.
What's not been mentioned is that YouTube add a small % of fake clicks to the like/dislike buttons, and also hide the real counts for a while, if there's a rush on clicks (to dissuade click-bandwagoning).
It's part of a trend towards fake, cloying, passive-aggressive forced positivity across the whole industry. It's a California thing.
Time was, if you didn't like a bit of code, you could say so. Nowadays, you have to wrap criticism in right zillion levels of disingenuous praise before you can even start getting to the point -- and the you get fake "thank you for the excellent feedback!!!!1!!" typed with gritted teeth and fantasies of backstabbing later on.
I've always found the in-your-face New York style more genuine, more honest, more efficient, and ultimately, kinder and more respectful.
You as a user should watch the video no matter what and then rate the creator, not skip it if you see a large amount of dislikes.
A good intermediate solution would be to only record downvotes that come with a comment.
Typically administrators respond by making it harder for the contributors rather than taking risk of poorer quality for the users, to contribute you have to become effectively professional and abide by institutional rules you may not agree with. Distributed editing tools only solves part of the problem and cannot not solve how to keep generating professional quality content without paying.
Strong moderation is what differentiates SO from Quora, I rather live with SO style rules as a contributor than suffer poor Quora type content quality as a user.
[1] I am cognizant of outdated version problems you raised, my anecdotal experience is in general the quality of answers that are present for a reader has not dropped massively in last 10 years.
My code throughput have reduced a lot in recent years with changing roles and I prefer reading the actual documentation, I am no longer solving a immediate/specific problem. Even then I still click search hits on SO few times a week, and most times after clicking through a few questions I can find what I need.
Your(others) experience will vary from my mine of course, but I don't see anyone attempting a better model or better content short of source documentation .
As the mix of us who knew early versions of Google and Stack overflow gets diluted year by year by new users who don't know how well these systems used to work this will become less and less of a problem for them so they can sleep well at work.
Until someone disrupts them. I'm happy to say it feels we are getting close to that moment now ;-)
At some point I wanted to have a high score to see if it would affect my daily rate (it didn't) or if I could use it somehow (same as having tons of stars on GitHub - another useless errand).
I started replying to a bunch of questions and the sheer stupidity of most of them made me realise I wasn't ready for that. Most of the content on SO is not great. There are definitely few gems in there and I certainly enjoyed "the <center> cannot hold", but most of the times the docs are better.
On a side note. I never understood people that copy code from Stack Overflow. I can understand people who are not familiar with a programming language - but surely if you're a minimum proficient it would be faster to just write what you need to do or look for a library to do it?
Encouraging them to write their own solution or even a critique/comment is not necessarily better.
You can also down vote questions not only answers, a question could be poor but it does not mean I can come up with a better one.
Common issues I come across are very vague statements without an actual question, very broad advisory /opinionated questions, very specific solve this bug in my day job/code , no examples/illustrations added, or a big log or code dump - i.e. no effort to separate out the actual problem statement, do my homework kind of questions, typically there will already be comments asking for improvements without much impact.
I wouldn't down-vote a question that I don't fully understand ( I guess that is true for most users too), at the same time I want to make sure bad content is not encouraged. It is just internet points for both users, nothing to fret about.
and way more spam, no thanks.
I constantly got heavily downvoted slideshow + text-to-speech autogenerated garbage videos that were ranked above good videos on search results
If they actually believed in the rationale behind this change, they would get rid performance reviews for engineers at Youtube too.
That’s literally not the case anywhere though. Maybe China’s social credit system, maybe, but even then I think you start with a presumed “innocence”.
I can’t think of any system of rights that are afforded after evidence merit and not birth. That’s not what rights are.
Not going to ask you what full rights means, as it implies you consider rights a gift from a government and not restrictions on a government.
Privileges are features that may be taken for granted as rights but are not enumerated protections. You have no right to drive, it’s a privilege that we all agree is handy.
I don’t think you are playing with semantics, but this is not a difficult concept.
GP makes a completely baseless claim that you don’t have “full rights” until someone says you do. They are talking about privledges.
Same. It feels like a passive-aggressive elitism, and not simply an anti-abuse feature.
Given the site we're on, I don't have a problem as much with the elitism (I'm not the target demographic, after all -I'm older and not working in tech) as much as with the "passive-aggressive" part of the rhetoric behind withholding downvote capability.
Stories can only be flagged, which as I understand can lead to bans/losing the privilege if it's abused.
A fake trailer can for example be spotted by the number of negative votes. If the negative votes concept is gone, using a low number of positive votes as indicator is less effective.
When you put dislike button, you are adding a degree of freedom in expressing an intent for the user. It is upto you, how you use this information. Most systems fail in this task and then they just remove the degree of freedom as their solution.