When we process a block request from user A -> user B, we remove the follow edges between user A -> user B and user B -> user A, and then add a block edge from user A -> user B.
When we process an unblock request from user A -> user B, we remove the block edge from user A -> user B.
I imagine that the "Aleksey checked that he was following me again" was either client caching, or eventual consistency latency. There's no nightly batch job or anything doing that.
Source: I work on the social graph service at Twitter.
After the test, I unblocked Aleksey, and this tweet includes the video where we saw that Aleksey could see my tweets in his timeline again. Something went wrong after that I guess.
https://twitter.com/muratdemirbas/status/1062763863332937728
Yet, another update.
It might be that since my tweet included a mention to him, Aleksey clicked on the notification and could see my tweets after that, and not in his timeline. We repeated the experiment, and found that after I blocked and unblocked him, he was indeed still unfollowing me.
All third party apps (except a single Enterprise tool) do not receive Promoted Tweets. They're only requested and displayed within the Twitter "Owned and Operated" clients.
That made me cringe a little. Especially when the rest of the page seemed pretty professional (save for the now industry-standard cartoon icons everywhere).
The "nerds" make 120-200k a year. The 25 year olds that went to Arizona State and studied drinking, getting laid, and "business" make up to 300k a year doing enterprise sales for startups.
The current culture is, "hey thanks NERD for geeking out on that distributed systems BS, really helped me out on my last sales call/commission."
Then the "NERD" gets to wonder what the F to do with their life come their mid 30s, while said sales person is set.
But yeah, let's keep calling ourselves nerds and allowing this to happen.
We also call ourselves "geeks," which is also quite bad, and we've done little to stop the media and industry rebranding us as "coders," a job title with little prestige or connotations of authority and professionalism, over our previous titles like "software engineer/developer/architect" or even just "programmer." Collectively, we're really pretty socially inept, aren't we?
Nerds seems to be a well used term there [0,1]. I agree that I'd prefer something else. Maybe they got the same guys that did the new logo to come up with it?
This was a bug. We rectified it as soon as the issue was brought to our attention.
The original version of this product launched in March 2013. Its intention was to just "bump" brands in your follow lists to the top, not to start placing Promoted Accounts randomly in the list—what the headline suggests. That was the bug.
Your account appears to be 100% anonymous and does not have any clear link with twitter that implies that you are really speaking for the company. Unlike say Matt Cutts saying something about Google handling spam.
I'm a bit skeptical about 'bugs' like this lasting for over a year and a half without there being some kind of intent involved. It's hard to imagine how this bug could have come into existence in the first place, harder still to imagine that code review and testing didn't catch it and that nobody complained about it in the meantime.
Random accounts appearing in followed/follower lists would be weird enough given that that is core twitter functionality but to have that happen specifically to promoted accounts would appear to be by design rather than by accident, especially if those promoted accounts did not appear in the lists to begin with. That would require some serious overriding and additional logic unless twitter is implemented in an unlogical way.
To put it plain: a change in sort order does not normally change the contents of the lists, that requires a lot more work and is usually not unintentional.
The 'bug' didn't last for over a year and a half. The product has existed for that long.
The bug was in the advertiser selection process. There was an issue with the job/dataset we use to select which advertisers to choose. It should only have been people in the follow lists.
As for "a change in sort order does not normally change the contents of the lists". Due to the size of some of the lists, we don't load the entire list when we show you the first 20 or so. Therefore, we would require an insertion process.
Thank you for clarifying this. I'm obviously not familiar with Twitter's code to any extent, but it seems reasonable you'd need the insertion process regardless.
Out of curiosity, can you speak to how long this bug actually appear in the wild before it was corrected?
This post perturbs me a bit - aside from the hardcore skepticism - I don't see how anyone who doesn't work for Twitter would be able to comment on the implementation of Twitter's follower lists.
It's strange because to me you are suggesting that I shouldn't believe an engineer at Twitter that it was a bug, because the bug should be impossible based on an imaginary implementation from someone who doesn't work at Twitter in order to prove that Twitter is 100% malicious.
I'm beginning to wonder is most posts like this play at our biases (we want to believe that Twitter is making some underhanded deals with "evil" advertisers), rather than taking a more neutral view.
It doesn't reflect very well on the company though when a bug like this is seen as something that people believe you would actually attempt to implement - even if this was a mistake. What does that say about what people think about that company?
Sorry, I hadn't seen it before, and my quick scan of it for joke-signifiers (date near April 1, funny names, strange references) didn't set off my alarms.
Only the engineering (and regulatory) challenge of a glass-bottomed plane seemed strange – and I was hoping for educated commentary here, because with Virgin/Branson, who knows?
"If a follow or friend relationship exists it is destroyed."