Interested to see how penalizing accounts for 'repeatedly tweeting at non-followers' is going to work when most of what I see on the platform is people messaging celebs who don't follow them back
This is what I was going to post, seems to be about 90% of the activity on twitter. They seem to only care about the experience from the perspective of the haves, anyone else just isn't part of the conversation.
One of the main problems with Twitter is people sign up and don't know what to do with it - except message famous people. If Twitter prevents the plebs from shouting at the intelligentsia then it will make the platform even more irrelevant. The last time I used it, it was like talking into a void and that was with hundreds of real followers.
So much for the strategy of tweeting at a company to resolve consumer disputes. Normally, I would say that's a good thing, but sometimes these companies just don't respond to non-public channels. Google, for instance, is notorious for making it difficult to reach an actual human in customer service.
That got me banned a couple of weeks ago. I wrote an application that recommended an event to the user once he\she would tweet "I am bored". I made the mistake by giving providing a fall back once the location could not be determined. This lead to ~40 tweets with identical texts. Twitter quickly pulled out the banhammer.
The listener is a pretty much off the shelf Tweepy listener writing into a Redis queue. For the recommendation is comes via an API where I can't share the code.
>Except when Twitter makes that a heuristic for detecting bots
There are several dozen. Maybe not turned on in prod, but that's a whole another story.
In fact, within three months of joining, way back in 2012, one of my very first tasks was to write a standard datamining job that would compute the difference between the GPS location during office hours (9-5pm) and the GPS location during home hours (7pm-7am) of everyone who tweets. A histogram of those differences would tell you about the commute distance of the average American who tweets. You could then bucket by region and say interesting things like the average NY tweeter commutes 25 miles more than the average CA tweeter.
Looking at the results we got, it was clear there was a substantial percent of bots, because their location varied so widely, minute to minute hour to hour. Haversine of GPS diffs will be reasonably stable, because your IP maps to the GPS ( we used the standard Maxmind geoip2 API) , and those IPs are relatively stable....Except if you are a bot and switching IPs willy-nilly.
This was just one instance, but there were several such projects...usually interns and new employees would work on these to get their feet wet, and then move on to more substantial projects.
Provenance of data was not in scope, 'twas more of a standard datamining "see if you can dig up something interesting" project.
Like I said, there were scores of these - one of my colleagues wrote the famous soda vs pop thingy which once again put location stats to good use- http://blog.echen.me/2012/07/06/soda-vs-pop-with-twitter/
With quantitative analytics being used and abused by ever more businesses, the advantage will go to those that can apply qualitative checks to their assumptions - and scale it.
Are you suggesting to get a Gmail account, turn on the setting that makes Twitter email you notifications, mark tweets as spam in Gmail, and then use Google's spam detection as an existence proof of a solution if it's successful?
It took them years to let users upload images (when there was a clear demand and parasite services grew out of this) and there's obvious spam accounts that could've shadow banned with very simple heuristics.
There are enough people talking to each other to make it interesting, at least for me. But the issue that most people are having is political nuts who just go around doing a search for "Donald Trump" or whoever and then abusing anyone who said anything remotely critical. Even worse are the actual white supremacists who do that or worse.
The bots are a problem as well, but they aren't what runs people off the platform.
That sounds like an informal description of a heuristic to flag accounts as potentially abusive rather than a specific rule with a specific punishment.