Twitter is testing a new reply design

Interestingly, Facebook did the same (remember how bad Farmville was at first), but at a much later date, and they seem to have weathered the change fine.

Third-party Twitter client-makers could've offered users an experience Twitter itself seems unable to offer - such as much better anti-abuse systems, better spam protection, better identification of bots, and so on.

But no, Twitter had to ban them so it can "control the experience" (as in offer you whatever type of ads they wanted and not have to deal with client makers when pushing them to users).

Everything that triggers a notification increases your score, and certainly favouriting tweets does so--but you have to be extra annoying for you to get banned doing this.

You can also get shadowbanned, which means your interactions trigger no notifications and your tweets simply disappear from everywhere. So if you're spamming you probably want to check every once in a while if your account is shadowbanned (by triggering a notification directed to a "control" account and seeing if it goes through) but there's no way to remove a shadowban: not even disabling/reenabling your account will remove it. It expires after a few days though.

There are "hard" rate limits in place as well (amount of favs/minute), but those can be avoided by using the API key of an official app. (Don't be mistaken--you build up a "spam" score anyway.)

This is something you often do in game design: Rather than say "pah, that'll never work", you try it out and see what happens.

> It's almost as if the devs don't use Twitter...

They probably use it -- but possibly they don't participate in it, as a social network, very much.

I follow and RT a lot of funny shit, and occasionally ironic/subtweet stuff. One time, I remember a Twitter dev -- after the debacle about testing extending the character limit to 10k characters, among other things -- tweeted, posting about how "We aren't idiots, and we aren't going to ruin twitter, get a grip." That tweet he posted quickly gathered over +2,000 RTs. Most of them being ironic, highlighting massively bad decisions Twitter made in the past to the detriment of its users (such as killing almost all 3rd party clients) -- or simply boggled by the fact they can't understand "a 10k character limit fundamentally destroys what twitter is" (this one is pretty obvious TBH, although I do admit I don't know if they ever even tested the 10k limit thing). I'm pretty sure I RT'd it, myself, just for shits and giggles.

The same individual then later tweeted in the day "Wow, I've never had anything blow up like my last retweet, with the irony and subtweeting. I've never had my tweets even leave my own follower circle." The guy had ~900 followers, and, well, was a Twitter developer (for reference: I'm a nobody and have almost ~1,300).

Now I don't mean to rag on the dude endlessly -- Twitter as a social thing can be fickle and not everybody is super popular or whatever. It certainly has a "way of working", depending on what sphere you find yourself in. But the fact that a product developer seemed to be caught offguard at ironic RTs of his statement, and that they've never even had their tweets leave their circle and seen the results... It just indicates a massive gap between what the developers see, and what users actually do with the platform.

It's like, no wonder they don't (apparently) understand concepts like people harassing or spamming. Most of the developers probably don't even have big enough circles to see it, let alone understand how it, or Twitter users, tend to operate at "scale" (larger number of followers, or simply very frequent use of Twitter despite follower count). You can't understand it by looking at analytics data -- you have to participate in order to understand how people interact. I think Twitter developers probably use the product, maybe even a lot -- but likely don't participate in the ways many of us do.