On Amazonfail
This might not make me any friends. But I am stupidly going to post it anyway.
I've had an interest in the recent Amazonfail debacle, but mostly from the other point of view -- that of the developers. This recent blog post where an unnamed Amazon employee talked about the problem was especially interesting to me. The description of the escalation of the problem, being on call 24/7... that's the sort of stuff I did in my previous career as an engineer, once upon a time.
Even though what I was doing at the time was slightly different (optical networks -- you know, making the internet go), the biggest outage I had worked on at the time was having the network of a major bank collapse -- they were estimating losses of half a million dollars per every hour their network was down (the outage affected not only their computers, but ATMs as well). Total outage time: 17 hours. You do the math.
We worked around the clock to restore their network. It was a very high stress situation, and we took a lot of shit from the client (getting yelled at is never fun). I think people frequently forget that, yes, the people working to solve the problem are human beings too. And in the case of Amazonfail, you've got people being called in to work on an Easter weekend, no less, to try and fix the problem.
The thing that surprised me most about the Amazonfail problem was how quickly and angry people seemed to get in such a short amount of time, propagated no doubt, by the speed of Twitter. Yes, Amazon PR fumbled the ball in not issuing a public statement fast enough. Yes, there might actually have been something fishy about Amazon changing their data, but if it was an error in the roll out of their updated code, that error could easily have propagated itself to unwanted areas (like to all LGBT books, for example). Yes, the problem affected a group of people that have had a long history of things like this. But I really think this was a problem on Amazon's side of things that got blown out of proportion by Twitter users.
I don't see Amazonfail as a good example of what Twitter is capable of. All I saw was a lot of angry shouting by people before all the facts were in. It was internet mob mentality, and that scares me. If people stepped back when they found out about the LGBT sales rankings being removed on Amazon, and simply wrote to Amazon calling for an explanation and fix of the problem, wouldn't that have worked just as effectively?
At this point in time, given how fast the internet can be used to voice our opinions, I think it's important, now more than ever, that we step back and think about what we say before we say it. I know I'm still working on that one, but civil discourse will get us so much farther than blind, reactionary anger ever will.
ETA: Clay Shirky's most recent blog post sums up what I was trying to say much, much better.
I've had an interest in the recent Amazonfail debacle, but mostly from the other point of view -- that of the developers. This recent blog post where an unnamed Amazon employee talked about the problem was especially interesting to me. The description of the escalation of the problem, being on call 24/7... that's the sort of stuff I did in my previous career as an engineer, once upon a time.
Even though what I was doing at the time was slightly different (optical networks -- you know, making the internet go), the biggest outage I had worked on at the time was having the network of a major bank collapse -- they were estimating losses of half a million dollars per every hour their network was down (the outage affected not only their computers, but ATMs as well). Total outage time: 17 hours. You do the math.
We worked around the clock to restore their network. It was a very high stress situation, and we took a lot of shit from the client (getting yelled at is never fun). I think people frequently forget that, yes, the people working to solve the problem are human beings too. And in the case of Amazonfail, you've got people being called in to work on an Easter weekend, no less, to try and fix the problem.
The thing that surprised me most about the Amazonfail problem was how quickly and angry people seemed to get in such a short amount of time, propagated no doubt, by the speed of Twitter. Yes, Amazon PR fumbled the ball in not issuing a public statement fast enough. Yes, there might actually have been something fishy about Amazon changing their data, but if it was an error in the roll out of their updated code, that error could easily have propagated itself to unwanted areas (like to all LGBT books, for example). Yes, the problem affected a group of people that have had a long history of things like this. But I really think this was a problem on Amazon's side of things that got blown out of proportion by Twitter users.
I don't see Amazonfail as a good example of what Twitter is capable of. All I saw was a lot of angry shouting by people before all the facts were in. It was internet mob mentality, and that scares me. If people stepped back when they found out about the LGBT sales rankings being removed on Amazon, and simply wrote to Amazon calling for an explanation and fix of the problem, wouldn't that have worked just as effectively?
At this point in time, given how fast the internet can be used to voice our opinions, I think it's important, now more than ever, that we step back and think about what we say before we say it. I know I'm still working on that one, but civil discourse will get us so much farther than blind, reactionary anger ever will.
ETA: Clay Shirky's most recent blog post sums up what I was trying to say much, much better.