So close. There are actually about eight of those tiny little pieces holding everything up, and yes, all those teams got slashed. https://t.co/vWhjNn5bpY
— Jim Redmond (@jredmond) November 4, 2022
“Reliability engineer on the Twitter Command Center team (TCC)”:
Oh, and while the press is focused on how advertisers will handle changes to content moderation, I really think they should be focused on this:
Nobody is going to advertise on *or subscribe to* a site that isn’t reliable.
And a whoooole bunch of SREs got laid off today.
For the record, “SRE” stands for “site reliability engineer”. These are the people who keep this place functional – who make sure your tweet is published and your DMs delivered and your Space usable etc. – and they are, quite literally, the backbone of this company.
It’s a tough job, but we had – past tense! – a world-class team of SREs keeping this place operational, even when whole datacenters go down (and then Queen Elizabeth dies, spiking traffic dramatically).
We had laudable uptime, to the extent that Twitter was where people came to see if something else was down. (Case in point.)
Our stuff wasn’t perfect, but we kept it going.
With skeleton crews of SREs, though,
* incidents will be more frequent
* incidents will be more severe
* incidents will last longer
* incidents will be more likely to repeat
This is not sustainable.
I’m still here, and I’ll still do my job while I’m here, but I don’t plan to stick around.
Perhaps that’s the point – if you get rid of those expensive, experienced reliability types then you can save a bit of cash up front – but it’s a deeply, deeply shortsighted point at best.
We will do our best.
— Jim Redmond (@jredmond) November 5, 2022