Thread by @swardley

X : Why such a fan of Lambda?
Me : You have serverless i.e. the runtime as a utility (focus on code, price per function etc) plus its event driven plus a mass of services to consume and events to create in your serverless architecture. Why wouldn't I be a fan?
X : Kubernetes?

Me : I could argue about how Kubernetes misses the point, wins a battle that is irrelevant and loses the war but ..
X : Yes?
Me : ... can't we just wait? The war on serverless is over next year, spoiler AWS won and it's only a few more years until it's recognised as the new norm.

X : The war hasn't started in serverless,
Me : It started in 2014. It takes 5-8 years for the winners / losers to be decided (2019-2022) and 10-15 years (2024-2029) to become the "new norm".
X : How do you know this?
Me : You don't. But it's the same timings I used for cloud.

X : It's early days in serverless.
Me : No, it's not. It's nearly the end of war (winners / losers chosen) then we're into the washup (becoming the new norm) and finally the slow slog of the laggards. I've said this before -

https://twitter.com/swardley/status/1066000066966708224

If you really want a guide, here's a handy table ... bit out of date, a few years old but it'll do.

X : I think you're wrong on serverless.
Me : I've been told that I will be wrong on so many things and so many occassisons that I've stoped caring ... apparently the future should have been VCE or OpenStack or now Kubernetes ... well, it wasn't and it won't be.

X : You don't think Kubernetes is useful?
Me : I didn't say that. It's not the future but it has immediate use in dealing with inertia, legacy etc. Don't underestimate these things, you might have to make compromises as I tried to explain here - dxc.com/us/en/insights…