I uninstalled TikTok a week ago because it was showing me too much of the same stuff, but I may eventually return after I’ve had a break from using it for a while. Not yet but maybe later. Anyway, I agree a lot especially about getting a lot of views. Though in my case not in the millions and not even in the hundreds of thousands. But one of the first videos I posted to TikTok was seen 50k+ times. And it wasn’t even a good video, just a little bit of nonsense fun. Other videos I’ve posted since then, on a different account, have only gotten a few hundred views each but to me that is a lot still compared for example to YouTube where my experience has been that the videos I’ve posted to YouTube get a handful of views.
And the most interesting part is that the videos I’ve posted to TikTok don’t get a lot of likes. And to me this says that if I am able to make better content (not an easy thing to do btw), then TikTok is very much ripe for getting a lot of views.
And this is encouraging to me, because I’ve always liked the Reddit and HN way of using upvotes and downvotes, and I feel like TikTok does the same for video that Reddit and HN do for text and links, except that with the video format they have they can also use number of loops a video is watched by each person seeing it, in addition to hearts and comments. And the most encouraging thing is seeing the videos getting those hundreds of views because it shows that TikTok really is using this mechanism and that they are showing your video to people to determine if it’s worth watching.
This is interesting both from a software development point of view, and also from a video creator point of view.
Even before TikTok I was thinking about how Reddit/HN style ranking of content could be applied to video. TikTok has the additional strength you mentioned of discovering your preferences. I think the preference discovery is difficult to replicate without a bigger team of people and probably some knowledge about Machine Learning. But I think there might be room for more video services that use the HN/Reddit ranking method like TikTok also does, while being targeted at a specific audience, like individual subreddits are and like HN is. I worked a while on this with some other people, but limited budget made it very difficult to get anywhere and I have had to pursue other paths instead for a while. But we may continue down that line again at a later point.
And the most interesting part is that the videos I’ve posted to TikTok don’t get a lot of likes. And to me this says that if I am able to make better content (not an easy thing to do btw), then TikTok is very much ripe for getting a lot of views.
And this is encouraging to me, because I’ve always liked the Reddit and HN way of using upvotes and downvotes, and I feel like TikTok does the same for video that Reddit and HN do for text and links, except that with the video format they have they can also use number of loops a video is watched by each person seeing it, in addition to hearts and comments. And the most encouraging thing is seeing the videos getting those hundreds of views because it shows that TikTok really is using this mechanism and that they are showing your video to people to determine if it’s worth watching.
This is interesting both from a software development point of view, and also from a video creator point of view.
Even before TikTok I was thinking about how Reddit/HN style ranking of content could be applied to video. TikTok has the additional strength you mentioned of discovering your preferences. I think the preference discovery is difficult to replicate without a bigger team of people and probably some knowledge about Machine Learning. But I think there might be room for more video services that use the HN/Reddit ranking method like TikTok also does, while being targeted at a specific audience, like individual subreddits are and like HN is. I worked a while on this with some other people, but limited budget made it very difficult to get anywhere and I have had to pursue other paths instead for a while. But we may continue down that line again at a later point.