at the the show, people clap. at the end of a post, they like. at the end of a book, people sigh. how do reactions matter? and can we react while we read?
Yes and no.
Yes because writers need validation in some form.
No because if we can’t determine a way for that validation to reasonably and transparently reflect writer’s writing and contribution to HackerNoon, it just becomes a vain metric.
I’m not a big fan of 50 claps because like the early Youtube adopters, I either give a story 0 clap or 50 claps.
Not only are claps too binary to give writers meaningful feedback, they also put readers in a congratulatory mindset. I’m really excited about exploring emoji reactions and ways to encourage better follow-up discussions.
The idea of clapping for an article fell flat on me. It is basically a spin on likes that allows you to like 50x. Just feels lame. With that said, I do like the functionality of a clap. I would suggest re-skinning it to something Hackernoonesque. I also like Danes idea of emoji’s. People are inherently lazy to follow through with these behaviors. The easier and more relateable the better. Crazy talk random idea…come up with a hackernoon robot or rocket ship with jet propulsion. People click on it similar to claps when they like. The more clicks and the rocket ship or robot changes and the fire propulsion changes colors. For example a super bad ass article that got a ton of clicks the rocket or robot has hot red fire coming out of the feet or bottom of the rocket. This type of stuff make it unique to hackernoon and nerdy fun for the readers.
This pretty much sums up how I feel about it.
Having a wider range of emojis to choose from, and encouraging discussion, is much more valuable to me than claps. It’ll also give us more insight into how people reacted to a post, not just the fact that they reacted.
Does a given story make people happy, angry, argumentative, empathetic? These are things I’d really like to know, and claps alone can’t give us that.
I’d love to be able to do that! We could have an emoji stream that shows up in the margin, showing how people reacted to a given paragraph…I’m thinking something like Soundcloud’s audio player, if that helps. You should be able to highlight a section of a story and react to that too.
If we do go that route, it would be a distraction from people who just want to read the content. So I think we’d want a
focus mode that removes distractions. But my gut tells me we should build these features in a non-distracting way so a focus mode isn’t necessary. Always tricky to get these things right.
We want to make the thing engaging but not distracting.
Whatever route you decide to go. Remember people are super competitive. Despite my personal belief that claps are lame, their are people who measure their success by the claps they get. This is why incorporating a unique “like” or “clap” like feature is good. Digg used the thumbs up. Other websites use the fire symbol. Emojis are cool as an add on but there needs to be an intuitive simple way to react positively to an article. People are vain. Tapping into that vanity will drive people to work harder for positive responses.
I don’t understand the point of claps in Medium. They should be measuring read time and showing aggregate read time… for the entire article, not just a page load.
The biggest you can give an article is to read the entire thing to the end.
(No, I don’t mean “rapidly scroll to the end”, either.)
We do track this in Discourse, per user and per post even, because reading is fundamental.
I’ve always like Github PR reacts. Use emoji to express how you feel about a comment/PR, but limit the scope of acceptable reactions to present (e.g. no reacts!)
This is the stat we plan to put atop all stories in Hacker Noon 2.0: TIME READ.
As a writer for a few medium publications pre and post clap era, I’d have to say that claps give a very warped view of how popular an article is. For example, even if a post has 400 something claps on it, it can easily just be a writer’s 8 close friends artificially boosting that score with 50 claps each. While a heart/like is binary and doesn’t show a spectrum of “likeability”, at least it does give a viewer a realistic idea of article quality + popularity (without having to see how many people actually clapped for the article).
Perhaps some middle ground could be achieved? I.e. Show “400 claps by 8 people” instead of just the number 400.
If claps were a limited source it may be more valuable. If readers only had 50 claps to give a day and they expire then maybe there would be more incentive to give them and not just give the first article they read all 50.
Never understood medium claps - what’s it really for? And this awful feature when you can give 50 claps just make things messy.
I personally love reddit upvotes and downvotes - it’s the best to understand user’s reaction - especially downvotes.
For hackernoon 2.0 I suggest to make UPVOTES and DOWNVOTES. But downvotes visible only for an author, it will help to limit crowd hate - when you read a story and see a lot of dovnvotes it changes your opinion for sure, that’s why it should be visible only for author.
But for sure you need some type of likes or upvotes - just to understand how users react to the content.
But the killer feature for hackernoon 2.0 could be in-depth (ga style) analytics for each topic. Medium’s analytics is awful.
As a reader, I like the idea of claps because it allowes me to express my appreciation to the author in a more nuanced way than binary “like”. I did give claps in between of 0 and 50. In the paid content world, my claps would be flattr credits.
That said, as a reader I don’t necessarily trust claps (even paid) as a metric to decide whether I should read an article. It’s clear that “viral” article on a hot topic get a lot of claps simply because of the topic or opinion expressed, so it is also a measure of affirmations from people who agree with the author’s position which may not reflect the quality, truthfullnes, objectivity of the article at all, or even the fact that people actually read the thing.
To continue an analogy with a concert claps - I don’t look for most popular mainstream concerts to attend. As a consumer, I am looking for content matching my “bubble”, my current needs and maybe pushing the boundaries a bit so that I can learn new things and opinions I don’t understand or agree with. Quality bar must be met at all times.
So, as a reader I need content to be curated to me personally given criteria above, and within that curated content I need an indicator of the quality of the content. Perhaps within my curated “bubble”, it’s ok to assume quality = popularity among the people from the same “bubble”.
I also like comments, especially when they complement the article with more insights, corrections or alternative points. I am curuous to see how recent comment types at ArsTechnica play out - upvote tags “interesting”, “funny”, “adds to the story” etc. The presense of comments or responses is usually a better signal to me as a reader than likes or claps (same caveats about viral topic apply).
So, from perspective of choosing what to read next, claps are far from ideal. They just an incrementally better thing over likes, but even in this form the clap value of the article should be presented not as an absolute total but some logarithmic score, given the nature of how it is given.
It would be cool if Hackernoon could innovate here. I am actually wary of bragging how many new authors are going to publish, growth metrics etc, because for me as a reader it says “watch out, a lot of mediocre crap is coming soon for you to sift through”… Whatever tools you develop for authors are great, just make sure they promote quality content and not aid clickbait titles and viral content. Unless authors start paying you for the privilege to publish, the readers are your primary customers. Which brings me to the ultimate question - viral content = more visitors = more ad money. Is Hackernoon’s goal to maximize money or generate quality content with “ok money”? Because there is a conflict here.
I think the clap metrics were orignally implemented for the monetisation purposes. I’m not sure on how the “premium article” rewards work.
Personally, all that matters to me if the blog post gets good reception.
A good signal is that sometimes @David would love a personal note “Nice!” or so and these would really be a confidence booster for me personally.
PS: I have never even checked the number of claps on my stories
first off, thanks for the great insight + detailed response.
Our goal is to maximize for writer’s distribution. We will always value expertise over breaking news, so virality looks quite different in Hacker Noon. People come to Hacker Noon first and foremost to read in-depth, long-form, sometimes deeply technical article. But it’s also a number game. The more stories we publish, the more readers there will be. Hence, more ad revenue from sponsors :). And that means increasing rates of publishing, ramp up our current editorial flow with more editors and in 2.0 more community editors. We want the community to tells us what stories are worth pursuing/publishing, to make Hacker Noon truly reflect our little “bubble” corner of the internet, like you said
What is “writer’s distribution”? If it is finding the “long tail” of audience for each writer, then it is aligned with reader goals. But that would need highly personalized curation for readers. If Hackernoon is a single “bubble” with no customization for readers, how can you maximize each writer’s distribution given finite eyeballs?
Reactions & claps only have meaning if they’re scarce. I rarely clap for articles on Medium for this reason, they don’t seem to mean much - at least with regard to how much I enjoyed the article or whether or not I enjoyed a particular thought, idea, or the article overall.
In addition to the standard reaction metrics (likes, etc.), having an additional set of limited reactions (“premium claps”, little Hackernoon watch icons, etc.) per day would very likely give me more reason to react to articles I enjoy. It would actually mean something: “I only have so many of these things to give out today, and I’m giving some to you” means a lot more than “I’m going to hold this button down for several seconds because I agree”. I could also see myself engaging and reading more articles if the number of daily “premium reactions” I got scaled with the amount of time I spent reading articles on average.
I think reacting while you read would be more powerful than just reacting to the articles themselves. Reacting to a particular thought or idea is more engaging to the reader and gives contributors valuable insight on what resonates with readers. Being able to throw a few claps at a particular passage would be interesting to me.
I really like the idea of having a limited number of reactions per day, to make them more meaningful. As others have pointed out in this thread, “50 claps” doesn’t mean much if I can do it on every article.
Maybe as you contribute to HN, in the form of submitting stories and comments, you earn more emojis? We still have to think about it as a team, but it’s an interesting concept for sure.