The User Experience (UX) Of Yesterday's Slack Outage
At InVisionApp, Inc, we're an entirely distributed team. We have about 150 people, all around the world, communicating exclusively through digital channels. A huge part of that workflow is our paid Slack subscription. Slack has become a critical part of our value chain. It's how we coordinate deployments, ask for code reviews, manage emergency "incidents," brainstorm, and generally chat about all the things. So naturally, when Slack went down yesterday, everyone in the company noticed.
And, here's the crazy thing: Life went on. Work still got done.
When you work in the SaaS (Software as a Service) industry, you spend a large amount of time and effort making sure that your service doesn't go down. You live and breath the service from a Provider standpoint (except, of course, when designing the product). So, for me personally, when I have an opportunity to truly and genuinely be the "user" of a service, I find it to be a wonderful opportunity to feel the user experience (UX) from a different standpoint - from the user's standpoint.
I am a "user" of Slack. When Slack went down, it sucked. But, in reality, it was just an inconvenience. There are things that I couldn't do. But, there were many more things that I could continue to do just fine. And so, the user experience of the Slack outage was getting work done. And, it made me so proud to see so many other people here at work having the same exact reaction.
This is not the first time I've seen this happen. At my previous company - Nylon Technology - when we would lose internet or lose electricity, nobody stopped. We found things to do. We cleaned desks. We cleaned the office. We organized papers. We kept working.
When a system or a service goes down, the user experience (UX) of that outage is up to you. It's what you make of it.
The timeliness of this is actually kind of perfect. Over the weekend, I watched CT Fletcher: My Magnificent Obsession. CT Fletcher is a world-record-holding weight lifter who has become famous for his infectious drive, passion, and outlook on life. He is well know for saying:
As he explains in this video, ISYMFS is exactly that - choosing to get things done or choosing to be defeated. And when you hear his story (he's died several times and has a mechanical valve in his heart), it sort of puts things in perspective.
Now, I'm not saying that there aren't critical systems out there that simply have to keep working (where people's lives are in the balance). I'm just suggesting that maybe those are few and far between. When a system does go down, you can choose to adapt and to move forward. Or, you can choose something else. The user experience is up to you.
Great thoughts Ben. I did experience that moment of panic as well, "Is it just my internet? Switch to phone, nope. It's Slack... What am I going to do?!?..." Breath, and keep working :)
Ha ha, yeah I was like, "if I can't Slack people, how do I know if Slack is down?" :D
The biggest point of friction for me was not being able to ask for a code-review. But, I just waited till Slack came back - it worked itself out.
You couldn't just send an email or use skype to get a code-review?
lol... this article gave me a chuckle... and funny to see how integrated some people get with a system with no SLA... come on Ben...
should have your messaging for CI and alike on a more stable messaging system than slack... (the opposite context to your article, but still worth asking the question, even if the answer is... slack is basic but it's enough)
We've really been enjoying Slack at work. It just seems to work really well (latest outage aside). Plus, it seems to be fairly extensive. I know the ops team just built some sort of Slack bot that integrates with our deployment tracking process. Pretty cool stuff.