Show HN: Hide Likes Everywhere

I used a similar extension a few years ago which just targeted facebook; and it made a big difference for me in how I experienced the site.

Before I used it I would find myself checking FB sometimes throughout the day after posting something, because some part of me wanted to know how popular it was. The metric being front and center made part of my brain want to optimize for it. After installing the extension I couldn't really know how many people saw what I wrote except vaguely by looking at who commented or something. And as a result I found myself being less self conscious when I posted things - since there was no negative "not many likes" signal I found myself writing more what was actually on my mind. (Rather than what I thought my friends wanted to read).

I hear what you're saying about being at an information disadvantage - but that only matters if "number of likes" is something you're trying to optimize for. I think thats a pretty shallow way to live your life. There's lots of other reasons to speak and listen than how "liked" those words are in the moment. If we rated food the same way we'd only eat junk food. If we raised children optimizing for likes they'd grow up stunted. If we rated academic papers that way we wouldn't even know about the replication crisis.

Ironically when I couldn't see the number of likes on FB, I think what I posted was more interesting as a result anyway. It was certainly more authentic, since while authoring I spent more brain cells thinking about what I wanted to say rather than obsessing about social perception. Every person who's opinion you're worried about when you write makes your writing about 1% worse - which isn't much, but it adds up.

The only real downside was that the extension didn't work on my phone, so I'd end up seeing the numbers of FB likes there anyway. I ended up deleting FB year or two later because it was still too addictive, but I think this extension is a great idea.

It doesn't cost anything to try it out. If you're not sure, run the experiment and see if you have the same experience I did.

Likes themselves (and ranking content using them) aren't the problem — it's their public nature that makes it feel as if our social value is determined by them.

If likes, followers, etc. were private (that is, you could see who liked your post but no one else could), you wouldn't worry about what others would think of you based on those numbers. But since they're public and prominently displayed, we equate those numbers to our social value and end up with undesirable effects like comparing ourselves to others and focusing on what would get approval rather than what's authentic.

Your concerns are valid if you are the type of person/organisation who worries about how many likes they get. I post to Instagram purely for my own amusement, and only use the feed to catch up with family and friends - I really don't care how many likes I get or any post I view gets.

I'm not sure if that's true anymore; I read the other day that e.g. Youtube's algorithm works with engagement, that is, the sum of up- and downvotes. Because downvotes don't mean it's a bad video anymore, it means that people disagree with it - but still would watch it. The algorithm probably also takes views, time watched (until voted), total time watched etc into consideration though. And some magic sauce depending on how advertiser friendly it is, which likely explains why the default suggestions (in anonymous browsing mode) are so weird.

I already adblock the Like elements from the sites I browse. It's less clutter on screen and if you never vote it serves no purpose. A lot of that stuff is 3rd party junk anyway.

It would probably be simpler to just release an adblock filter instead of having another redudant addon, but I'm okay with its purpose.

Instagram has been testing hiding likes for over a year now. They've recently begun testing in the US [1], so there's a good chance they'll hide them for real worldwide. I don't know if Facebook will follow through, but IG will definitely set a precedent.

1. https://later.com/blog/hidden-likes-instagram/

Like stated on the page it probably helps to stop comparing yourself to others. You call it "information disadvantage", but exactly this is the feature. Social media is distorting our views. I see your point though, this doesn't defeat any biased algorhitms. But I don't think that is the point of this addon either.

Set your platform to chronological sort if you can. If you can't, then this plugin is still good and complements a hypothetical plugin that forces chronological sorts.

Someone could rewrite Facebook with some minor tweaks and become very successful!

So would you rather HN display points on comments?

Yup! Planning to get that done in the next several days.

I created a form if you'd like an email when the Firefox build goes live: https://forms.gle/aiEqoSVk6XMb5DaL8

Chiming in as another firefox user to say that I'd try this if my browser were supported.

Sites with likes/voting rank things for you based on what they think is worth looking at. So as long as you're not on HN/reddit's "new" sections, you're not randomly clicking in the dark. You're seeing the "top"/"hottest" things on those platforms right now.

Like/vote counts add bias, induce jealousy, and make it harder for people without reputations to get visibility. For a lot of platforms, I'd guess that inducing jealousy helps drive engagement. Which may be why Facebook is still only "testing" removing like counts - if removing like counts makes people who use Facebook happier, but reduces engagement, then I doubt they'll do it.

At the same time, I agree that there is value in like/vote counts. If something on HN has over 1,000 votes, it's very likely that I'll get value out of reading it.

With voting systems, there are lots of tradeoffs. Most of the time, unfortunately, platforms seem to be optimizing for engagement rather than happiness. I wonder what would happen if they optimized for happiness instead.

You may be able to do this with a content blocker that hides elements matching the selector a[id^='down_'].

I used to read HN comments first to determine if the article is worth reading. However, if I read the article first, I have the same experience as you where I'm very surprised how other people interpreted the article. Every once in a while there is some enlightenment.

HN itself is a good example of a site with up/down voting with no counters (unlike reddit). It would be interesting to see some research on how that affects user behaviour.

As a counterpoint, if you recognize it could be problematic in the future, the extension could be used as a preventive measure.

Thanks for the kinds words! Planning on releasing on Firefox in the next several days and I created a quick form if you'd like an email when it's ready: https://forms.gle/aiEqoSVk6XMb5DaL8

Hey, I built it with the intention of eventually allowing anyone to modify any site they wish (thus the request for all sites), but I think I'll just make that into a separate extension and tighten up the permissions for this one.

Thanks for the feedback!

agreed that they should open the source. I imagine it's a pretty simple app. a mapping of css selectors with a single line that runs on page load, something like this:

    document.querySelectorAll(css_selector).remove()

the hard part would be maintaining the css selectors. This is probably and arms race situation, where websites will obfuscate their elements in such a way where identifying the elements you want is nigh impossible, but that would probably take more engineering than it's worth. What's really interesting is that facebook does not scrub their `data-testid` attributes, so you can basically key on that since they probably test that part of the application extensively, and probably run smoke tests/chaos engineering on their live app so removing them might be impossible in their current implementation.

right now on facebook the appropriate js code would be:

    document.querySelectorAll('*[data-testid="fbFeedStoryUFI/feedbackSummary"]').forEach(element => element.remove())

If you have an extension like Stylus installed, you can quickly write a new rule. This might also change the saturation of other elements, but I'd be surprised if you wouldn't consider that a benefit.

The desaturation is just a collection of rules like the following:

    .c88, .c88 a:link, .c88 a:visited {
        color: #888888;
    }

If you want to make it darker you could just use the following in a Stylus rule for news.ycombinator.com:

    .c5a, .c5a a:link, .c5a a:visited,
    .c73, .c73 a:link, .c73 a:visited,
    .c82, .c82 a:link, .c82 a:visited,
    .c88, .c88 a:link, .c88 a:visited,
    .c9c, .c9c a:link, .c9c a:visited,
    .cae, .cae a:link, .cae a:visited,
    .cbe, .cbe a:link, .cbe a:visited,
    .cce, .cce a:link, .cce a:visited,
    .cdd, .cdd a:link, .cdd a:visited {
        color: #000000 !important;
    }

On which platform?

Works well on Firefox on Android and GNU/Linux.