Is there a support group for serverless devs worki...
# random
s
Is there a support group for serverless devs working in an organization that doesn't embrace serverless tech? You know, somewhere we could come up with thoughtful responses to questions like "can't we just do this in EC2?"
t
have you tried responding "shut up"
m
I don't know if this helps you, but I literally believe my org's strategy was they expected to watch me fail and then have to roll everything back. Things worked out a bit differently than that and you can't put the genie back in the bottle....
g
"We could do it in EC2, yea. We could also probably do it in C or perhaps assembly" ๐Ÿ˜„
on a more serious note, I usually list the pros and cons. Talk to the EC2 people, write down their concerns in a doc (in detail), try to address them
usually there is some hidden concern which may or may not be valid... if its addressed you might get a sigh of relief ๐Ÿ˜„
s
I think it commonly comes down to a lack of familiarity with a given tech, which is fair
g
I can tell you my typical concerns with serverless ๐Ÿ˜„
(which... actually might depend on whether we consider containers to be serverless or not)
t
The pushback you're getting isn't "real" pushback. It's mostly hesitation around any change, becoming a beginner again, acknowledging someone else is more aware of something than you, etc. How you deal with this probably doesn't have too much serverless specific approaches. Explaining why serverless is better than doing it in EC2 for a bunch of logical reasons probably won't get you a response besides "that sounds really complicated and we can already do all of that"
I'd suggest maybe thinking about it a different way where you find opportunities to build with serverless and build excitement around it. Avoid asking people to use it or suggesting it, just treat it like this other thing that you're playing with + sharing exciting things about it here and there
Eventually people will want to be in the know and research it themselves and then bring it up to you as though it's their idea
This is the only approach I've ever seen work at larger / established orgs where things feel entrenched
The root issue is if you're bringing something new and the response isn't excitement to at least learn as much as possible, you're dealing with an underlying resistance to change
s
Yeah, much of this resonates with my experience. The response I've seen often falls into two groups 1) I don't understand serverless so I won't use it and 2) I undersetand it and prefer not to use it. I think most of what I've faced is in the first group. They reach for the tools that are familiar to them, which is not serverless.
t
yeah I might be more cynical because of my experiences but it's been really hard for me to create organizational change like this when the culture isn't excited about new things
s
I deal with this every day of my life. Sometimes I get in the mood to start a DevOps podcast. This would be one of the top/recurring talking points. I think about it often, but Iโ€™m not gonna do it
s
Until then, you have a safe space here to vent ๐Ÿ˜†
s
Thanks man
Another slack Iโ€™m on has #cryparty for stuff like this. Iโ€™m not recommending it, itโ€™s kind of a shitshow, just saying
s
I converted a SLS app to SST and gave a demo today. One of the questions from the audience "if we push our development environment to the cloud, what happens when the services go down?" ๐Ÿคฆโ€โ™‚๏ธ
A common refrain from the mythical "I only develop cloud software without network connectivity" crowd.
m
I think that problem needs to be framed in terms of trade-offs.
You gain acceleration and understanding. You lose that "I can write cloud apps from my bunker with no wifi" glow. Or maybe say a rare outage will disrupt work. Likewise, an IAM messup (OTOH the IAM problem when you're on your way to prod is MUCH WORSE).
g
yeah that's why I'm big on pros/cons lists
if I acknowledge and review the concerns the other side is having and manage to present them fairly in the doc I typically have reasonable chance of success.