4 ms·
Perhaps failing is so normalized in tech that I'm naive to how the rest of the world thinks, but I don't think failures of the kind Caliphate made should cost s
by madrox 6y ago
Perhaps failing is so normalized in tech that I'm naive to how the rest of the world thinks, but I don't think failures of the kind Caliphate made should cost someone their job no more than an outage should cost an engineer theirs. A blameless retrospective should occur with the appropriate updates to process so it can't happen again. This was an institutional failure, and I don't suddenly feel better about the accuracy of the New York Times because Andy Mills is not there. I doubt anyone else feels different.
To me, where this went off the rails was when Caliphate turned into a story about privilege and an individual's character, regardless of the truth behind it. I can see why it was allowed to happen, because it distracts from the institutional pattern of behavior by the NYT. Based on the characterization here and my understanding of the Caliphate story, this does not feel like justice. It feels like a witch hunt.
- AstralStorm 6y agoCome on, are you advocating for corporate cover-ups of bad behavior? Because "institutional failure" is an excellent way to diffuse blame and get nothing fixed.
- madrox 6y agoScapegoating an individual is also an excellent way to ensure nothing gets fixed. Was it a cover-up of bad behavior when Amazon refused to release the name and/or fire the engineer whose error caused the 2017 S3 outage? We don't fire or shame those people, and they arguably do just as much damage to the credibility of the organization they work for. Instead, a retrospective is held, they determine how it was possible such a thing could happen, and new checks and balances come out of it. Often you can even read about the entire thing, as is the case in the example I mention[1]. The Caliphate situation tells me that such a thing will likely happen again at the New York Times without better process to prevent these issues. That the institution of journalism wants to blame the person instead of generate learnings, I'm inclined to think nothing changed. [1] https://aws.amazon.com/message/41926/ https://aws.amazon.com/message/41926/
- zepto 6y agoIt’s not comparable. It’s both an institutional failure and an individual one. We know bugs and mistakes happen. If an organization has a QA or testing process and the code passed through the accepted processes and yet the still failed, we may not hold the engineer responsible. If an engineer on the other hand, ships code which has not been tested, when a testing process exists, then we might well hold them responsible.
- madrox 6y agoMy understanding is that this story was run through the New York Times fact checking department. Is that not the equivalent of testing?
- zepto 6y agoIf that is what happened, then I agree. My understanding is that the fact checking department exists, and that Mills was told he didn’t need to do his own fact checking. But it wasn’t clear to me that the fact checking department did the fact checking. In this situation, it wouldn’t be unreasonable to expect Mills to be responsible for making sure the fact checking has been done before considering his work ready for publication even if he wasn’t the one doing it. Just like a release engineer may be responsible for making sure the release has been through QA.
- catlifeonmars 6y agoThe two aren’t mutually exclusive. It’s always a process failure when something like this happens. Sometimes it’s also an individual failure.