Do all the things like ++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatarSign Up
AlmondSauce989029dNah, there's loads of potential stumbling blocks that, if you're not familiar with it, can be head-scratchingly and/or cumbersome to overcome.
Stick with it and you'll get there, but I don't think anyone new to this stuff looks at it and becomes an expert in a day. The nice thing is that when you *do* become an expert in it, it makes a whole bunch of workflows a lot easier - it's another tool in your toolchest, so to speak.
Inxentas53629dThis is what I call the "inadequacy fallacy". It's the idea that pretty complex things can appear simple for others because they use it so regularly it has become part of their daily vocabulary. That doesn't mean it's simple, or can be learned in a day.
I feel the same way about unit testing. I have the impression everybody does those while I don't yet see why and how I'd use them for my own work.
PaperTrail751829d> I sometimes think I'm too stupid to use it.
I hear ya'. I'm struggle with automating deployments and making sure different containers can 'see' one another (among probably 100 other issues). Its very, very fragile. I see demos, read blogs, etc, but keep running into the 'Draw the owl!' situation.
100110111113529dWhat's this about remembering the creator of Docker tweeting that if WASM was around back then, there would have been no need for Docker? I wonder if that would imply one would be better of learning WASM than Docker... guess Docker retains use cases even though WASM now exists, but somehow I don't see learning Docker as as important now than I saw it earlier
>no need for Docker?
We'll be using Docker like crazy for our integration environment testing. Being able to create SQLServer containers, web service containers, etc all within their own 'network', is going to be awesome. We're already doing Docker 'by hand' and once I can settle the political disputes (long story) other devs should be able to click a button, and be able to have full integration path testing. Build+run the containers, copy some data, fire up the tests, destroy the containers, and the coffee never gets cold. The 'by hand' method has nearly eliminated deployment issues for our critical warehouse systems and once we automate, we'll be able to create scenario based tests in minutes vs days+weeks (long story, involves inter-dept politics).