Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Testing a few new designs that don't show the public dislike count

twitter.com

Post reply on HN
Print view View original

451–460 of 508 posts

Re: Testing a few new designs that don't show the public dislike count

#451

Post by lawnchair_larry »

markdown wrote: post #413
EricE wrote: post #306

Earlier quoted context omitted.

There is absolutely a strategy - there is a particular point of view they favor, and anything counter to it is removed. Dislikes are embarrassing to, for lack of better terminology, their pet causes. Simply disabling them on certain videos or channels calls even more attention to them - so why not just remove them entirely. The only thing surprising to me is that it took them this long.

What a silly conspiracy theory. If the wanted to hide embarrassing downvotes on certain videos, they'd just remove those downvotes, not the entire feature.

They do that too

Re: Testing a few new designs that don't show the public dislike count

#452

Post by shuntress »

symlinkk wrote: post #440
shuntress wrote: post #351

What is going on in the comments to this article? Why has this relatively minor youtube feature change dredged up such significant vitriol and conspiracy-adjacent rhetoric in the hacker news comments? Skimming all the comments here I'm seeing lots of flaming and hand-wringing. I don't have any actual measures, but to me it feels like significantly more than normal even for an article about a high-profile site/company…

Youtube is one of the most culturally significant websites on earth, and the like / dislike ratio on videos are a huge part of the experience there. I can’t believe you think it’s a “relatively minor” thing, do you never look up videos online?

A quick glance through my phone's screen time log shows that I average roughly 45 minutes per day on YouTube.

Maybe I'm an atypical user but I would not call the like/dislike ratio "a huge part of the experience". I personally never look at it.

Re: Testing a few new designs that don't show the public dislike count

#453

Post by yjftsjthsd-h »

jasonfarnon wrote: post #405
EricE wrote: post #306

Earlier quoted context omitted.

There is absolutely a strategy - there is a particular point of view they favor, and anything counter to it is removed. Dislikes are embarrassing to, for lack of better terminology, their pet causes. Simply disabling them on certain videos or channels calls even more attention to them - so why not just remove them entirely. The only thing surprising to me is that it took them this long.

It doesn't need to be that dislikes are embarrassing to their pet causes. There were a few instances where business partners would, e.g., post an ad campaign that got so many dislikes the dislike ratio entered the news cycle. I'm sure google gets pressure from their business/ad partners when that goes down.

Wouldn't business partners be a trivial subset of their pet causes? It's not just a political set.

Re: Testing a few new designs that don't show the public dislike count

#454

Post by 81m »

frongpik wrote: post #196
argvargc wrote: post #24

Earlier quoted context omitted.

There's a site tracking these, some of the stats on older videos with high views are crazy (down the page): https://81m.org

Looks like trolling to me. There's 95% dislike rate even for a mundane "paycheck protection bill".

It is not trolling, and you can verify that fact.

I made the site in question (81m .org). My inspiration was the work of Zoe Phin (whose name is phzoe on HN), who noticed that YouTube was deleting dislikes and decided to write some scripts and build some charts.

I replicated her approach and automated it further. The result is 81m .org.

If you want to validate my data, it is easy to do:

- Trivial way: For some video, check the official YouTube likes/dislikes/views numbers from time to time, and compare them to the numbers I display. You will see that they line up.

- Harder way: Get a YouTube API key, query `http s:// youtube.googleapis .com/youtube/v3/videos?part=statistics&id=&key=`, and log the likes/dislikes/views for one or more videos. Compare to my data, and you will see that the two datasets line up.

You can find more info at 81m .org/about

(Sorry for the " .org" and " .com" and other whitespace insertions above. I am worried HN will shadowban me if I post links.)

Re: Testing a few new designs that don't show the public dislike count

#455

Post by CapricornNoble »

davesque wrote: post #108
liaukovv wrote: post #51

Earlier quoted context omitted.

Synthetic brigading is what exactly? Bands of synthetics downvoting videos to bring about downfall of humanity? Or do people's opinions count less if they are organized?

Not sure if that was hyperbole, but since you ask, yeah I think it could actually bring about the downfall of humanity. For example, what if climate science videos were being significantly brigaded by bots controlled by oil companies? Might not be that far-fetched. If we fail to communicate effectively about issues like that, we could very well see the downfall of humanity. Even "organic" brigading by the anti-vaxx c…

>>>For example, what if climate science videos were being significantly brigaded by bots controlled by oil companies?

I think we need to separate the bot problem from the organic humans. BOTS should be the issue that is countered, regardless of whether they are upvoting or downvoting content. Humans expressing themselves is just grass-roots organizing.

Re: Testing a few new designs that don't show the public dislike count

#456

Post by CapricornNoble »

Mountain_Skies wrote: post #207
swagatkonchada wrote: post #93

well, time for someone to build a third party dislike count and release browser extensions

Or for tiktok to start a competing video streaming site. Sure, if you go against the CCP, your video will be deleted and your account nuked, but you can still post those on YouTube. For everything else, tttube won't care who it offends in the US.

I would expect a very different type of video content from a website that is phonetically similar to "Titty Tube"...

Re: Testing a few new designs that don't show the public dislike count

#457

Post by notriddle »

adventured wrote: post #426
water8 wrote: post #417

Earlier quoted context omitted.

Let’s cut the BS. It’s about not looking bad. Its that simple

That's absolutely what it's about. There's nothing to debate about that. It's the same reason Trump wanted to be able to block people on Twitter. For the same reason he wasn't allowed to do that, the White House shouldn't be allowed to disable YouTube commenting. But hey, it's not like there's a double standard in favor of the Democrats and Biden.

I’m curious if Twitter would’ve allowed him to blanket disable comments. Not blocking specific individuals, but just not having any comments at all.

They have an option for that (technically, “only people you @mention,” but as long as you never @mention anyone, it won’t let them reply), but I don’t remember hearing anything one way or the other about him trying to do that.

Re: Testing a few new designs that don't show the public dislike count

#458

Post by qiqitori »

hn_throwaway_99 wrote: post #245
causality0 wrote: post #83

Earlier quoted context omitted.

Companies are only interested in user behavior that increases engagement. Everything else is on the chopping block.

Totally agree. I notice I find comment sections of sites that allow downvotes (e.g. HN, Reddit) much easier to navigate and filter than sites like Facebook and Twitter that only allow upvotes. Only allowing upvoting gives you similar dynamics to partisan primaries in the US - the most polarizing content filters to the top instead of sinking to the bottom.

HN doesn't really allow downvotes though. At least not unless you've reached a million karma points (from my understanding).

Re: Testing a few new designs that don't show the public dislike count

#459

Post by notriddle »

EricE wrote: post #66
Hammershaft wrote: post #48

Earlier quoted context omitted.

You think Youtube is removing dislikes because of white house vids that average out less than ~25000 views each?

Not just the white house. There's wholesale rejection that is embarrassing and counter to the messaging being pushed my the media and cultural elites. Disney with the Gina Corano firing has been a huge source of continued embarrassment. Widespread rejection of woke takeovers of popular culture in general are problematic so no, I'm not surprised downvotes are getting remove universally. Much easier to hide reality whe…

It would be even easier to hide reality by... actually hiding reality. YouTube is closed-source, and they already use fuzzy numbering for dislike counts. They could’ve rigged it so that, instead of actually displaying the number of dislikes on the WH videos, they instead displayed the logarithm. How would you know?

The fact that they are removing the dislike count entirely instead shows me that the problem is not limited to a certain channel, or even a certain category. They are hiding that number on the entire website, because they are trying to solve a perceived problem that affects the entire website.

Re: Testing a few new designs that don't show the public dislike count

#460

Post by onion2k »

retsibsi wrote: post #9

This seems rather user-hostile. The downvote count is useful for identifying videos that aren't what they claim to be. And how about DIY videos that give terrible advice? A big downvote ratio will get most people's attention in a way that critical comments may not.

The downvote count is useful for identifying videos that aren't what they claim to be.

Down votes only tell you someone decided to give a negative reaction. They don't tell you anything about why they did that. Using them as proxy evidence for your assumptions is dubious, especially on videos with small numbers of votes.

Often the reason why public dislike numbers are hidden is because people copy reactions - a visible count of down votes attracts more down votes. That doesn't tell you everyone who watched the video had rationally decided that the content is actually bad.

Post reply on HN