3 ms·
It will be nice to see the long term data. But I don't think being in the news has anything to do with it. Nobody who wasn't using Twitter before suddenly hear
by dvanduzer 11y ago
It will be nice to see the long term data.
But I don't think being in the news has anything to do with it. Nobody who wasn't using Twitter before suddenly heard the announcement and went "Hearts? Finally! I fucking hate stars!"
Twitter's a global media platform, so it's hard to tell when the tail is wagging the dog. Unfortunately, the article doesn't explain that it's a 6% increase in this specific activity relative to all the other activity they track.
¯\_(ツ)_/¯
- Zancarius 11y agoSpeaking as one data point, for myself, "favoriting" something suggests I want to bookmark it for later use (which I often don't), but I also know plenty of people who utilize the star as a pseudo-like. I've tried that myself for a while, but it just felt wrong. Apparently most people don't share the same cognitive block. I agree with tyre's assessment. "Liking" something is substantively different from "favoriting," at least for those of us who overthink its application or feel that marking something as a favorite has connotations beyond simply giving a nod to the person who posted. What this means is anyone's guess or even if it's going to matter in the long run. Insignificant observation? Likely! But it's one more data point that agrees with a few others here.
- dvanduzer 11y agoI'm in the vocal minority along with John Gruber who is upset about the fact that 9 years of stars are now hearts: http://daringfireball.net/linked/2015/11/04/red-heart-problem http://daringfireball.net/linked/2015/11/04/red-heart-proble... But I will get over that, because Twitter is a very useful medium. This has nothing to with the success of Twitter as a business. What I am interested in, is whether this is detectable from background noise of people using Twitter more because it is in the news. That is not the sort of thing that a TechCrunch puff piece is going to care about, and/or Twitter may have good reasons for not bothering to share the data. But I would like to know!
- Zancarius 11y ago> What I am interested in, is whether this is detectable from background noise of people using Twitter more because it is in the news. Probably not, or at least I'm sure there's some percentage of users who were oblivious to the fact they could "favorite" something. Now they know, and now they're using it (or were reminded?). That said, I'm sure there's also a non-trivial number of users who feel as I do. I'm sure their UX folks entertained the idea of having both (liking via hearts + bookmarking via favorites) and decided it would be too disruptive to change at this point. I can't really blame them, but I do have to wonder: Are those use cases substantially different enough to warrant having both, or is the advantage of having a single "like" feature (fairly well understood, even across platforms--think Facebook) a more overriding concern with greater utility while maintaining simplicity? It definitely sucks for people who were using favorites for what they were (or thought they were) intended only to have it changed up.
- pgeorgi 11y ago> "Hearts? Finally! I fucking hate stars!" I can't quite make sense of https://twitter.com/mjg59/status/661781723038912512 https://twitter.com/mjg59/status/661781723038912512 but it's quite clearly related to the (soviet-)star->heart transition over at twitter.