Not easy to discuss this political analysis. Not a lot about how, technically, things went wrong here. Those details about how the tech failed, why SWA couldnt adapt arent entirely mysterious & unspoken, but still arent super clear either. Saying that the bosses didnt get it is fine, but more elaboration on what they didnt get, how things really went so bad would be nice.
> but more elaboration on what they didnt get, how things really went so bad would be nice.
The article addressed this, imho.
1. Software from the 90s was neglected (cost center that didn’t improve roi, with roi being a focus of the new CEO from 2004).
2. The software didn’t grow with the airline or contemporary circumstances that the front line people were aware of a notified the management of multiple times to no avail.
The tl;dr is that the airline had grown but the processes were still largely manual rather than automated with a small enough margin of error that a bad storm might cause the system to collapse.
I wish we had more details of the specifics: what system failed? What OS was running them? How could phones be the bottle neck? Did scale and volume bring the system down? What are the details? Where is the actual breakdown?
Imagine a fully offline system. The only way to interact with it is by sitting directly next to it. Instead of a NIC it has an employee with a telephone on her desk.
I'm just making things up, but this seems like a way to position the phone system as a bottleneck.
7 comments
[ 1.9 ms ] story [ 40.0 ms ] threadThe article addressed this, imho.
1. Software from the 90s was neglected (cost center that didn’t improve roi, with roi being a focus of the new CEO from 2004).
2. The software didn’t grow with the airline or contemporary circumstances that the front line people were aware of a notified the management of multiple times to no avail.
Edit:
This answers your question somewhat:
https://www.reddit.com/r/SouthwestAirlines/comments/zxg6op/t...
The tl;dr is that the airline had grown but the processes were still largely manual rather than automated with a small enough margin of error that a bad storm might cause the system to collapse.
I'm just making things up, but this seems like a way to position the phone system as a bottleneck.