lmao ait's all fucked
reblogs are fucked
the way this works is, someone up the chain gets an idea in their head that "tumblr is too hard to use for new users" (which is true, more on that later) and the easy scapegoat is "why are there THREE different ways to reply to a post: replies, reblogs, and tags???" they bounce a few brain cells together and come up with a plan, albeit not a good one. they usually try to copy what other platforms are doing because those are known, working methods for building a feed with likes and re-(verbs). in this update's case they finally pulled the trigger on the idea that reblog chains should split off from the main post because they're effectively new posts (in the tumblr DB, every reblog is its own post but that's not important) i'm gonna be honest: when I was working at tumblr, I had this same exact thought. I thought "why don't I get activity on my reblog comments that get their own likes/reblogs/etc". Then I thought about it for more than 10 minutes and realized it's a bad idea. I was also an engineering grunt and not a product decider so I had no real say anyways.
Anyways, plan from above gets executed with "A/B" testing, sometimes. In this case I'm not sure if it was but, in my experience, even when Tumblr had more competent knowledgeable people working on it, they never did a real A/B test the way you should. They got close, but in the end things just got rolled out if they didn't disturb things "too much". They also look for signals that it's working if people post more. do they separate out "people posting about this saying this is a bad idea" vs "posts in general"? No. I was there for the "no more replies" then "replies are back!" fiasco. The justification for ditching support for a feature (any feature, not just replies) is that anything that's an extra thingamajig upon the Tumblr codebase requires you to maintain it and fix it if it breaks. Less unique features = less stuff to fix. In this case, they saw replies were barely used vs other methods of communicating, so they figured they could get rid of replies when they introduced Messaging! Because now you can just DM the person rather than leaving a reply!
obviously that backfired. Tumblr once again touched the hot stove of "small but vocal group of users use this, actually" and they had to revert it. What they did do, at least, is fix a bunch of stuff with replies before bringing them back. I can't remember off the top of my head what they fixed but it was something like making them easier to use and read.
back to "tumblr is too hard to use for new users". it is because not many people use tumblr vs other places. so its patterns are not familiar. but that's not a good enough excuse for companies that own tumblr because infinite growth is the goal. most posts = more users therefore friction between users and making new posts must be eliminated. however this conflicts with what makes tumblr unique, which is why people stick around to use it. catch 22.
this is going to keep happening as long as people, at any company, in charge of "product decisions" do not actually use the product they're making decisions for, or don't bother to use it in a way that people who really like it use it.
