I stumbled across yet another most unnecessary "DevOps" tool for ensuring config files adhere to a structure, schema, & satisfy naming conventions. It was written in a dynamically typed language using a very stringly (weakly) typed API that they exposed to users for extension. /1
-
Show this thread
-
Inevitably (from prior experience with that same language) this will grow out of control maintenance wise. I now know it is more succinct to check the first two reqs listed in a language that offers ML-style types & anecdotally (having exp on both sides) yields more correct code.
3 replies 0 retweets 6 likesShow this thread -
I am curious if it would make sense to write about this by building an equivalent tool to do the same evolving it to use a variety of type-based and functional techniques? As an industry we could save so much effort if only people knew of these methods. No hate.
2 replies 0 retweets 16 likesShow this thread -
I do not mean to sound like there is anything novel in my approaches; these are tried and tested typed functional programming approaches to solving this class of problems. My suggestion is making it more accessible and applicable to the realm of "DevOps".
5 replies 0 retweets 13 likesShow this thread -
Replying to @SusanPotter
Rust's derive(Encode) and Decode style (which comes from the techniques you're talking about) works really well for me.
1 reply 0 retweets 0 likes -
Replying to @wycats @SusanPotter
That said, are you referring to Ruby as a stringly typed language? Or a different language?
1 reply 0 retweets 0 likes -
Replying to @wycats
I didn't want to bitch about one specific tool just because I found it this morning (it is just the latest non-optimally productive I've found), but it does much more than derive Encode. I think the approach to deriving the level of "linting" is cumbersome, incurring maint costs.
2 replies 0 retweets 0 likes -
Replying to @SusanPotter
I ask because I object to calling Ruby stringly typed even though I agree with the rest of your assessment. I understand your desire not to attack a language though :)
2 replies 0 retweets 0 likes -
Replying to @wycats
👻A Registered Voter 👻 Retweeted 👻A Registered Voter 👻
People see what they want to see so they can launch into attack mode but I have a hard time seeing where you saw me do this here in a genuine attempt to understand the original tweet.https://twitter.com/SusanPotter/status/947842295977693185 …
👻A Registered Voter 👻 added,
👻A Registered Voter 👻 @SusanPotterI stumbled across yet another most unnecessary "DevOps" tool for ensuring config files adhere to a structure, schema, & satisfy naming conventions. It was written in a dynamically typed language using a very stringly (weakly) typed API that they exposed to users for extension. /1Show this thread1 reply 0 retweets 0 likes
I was definitely not trying to launch into attack mode and apologize if it came across that way. I talked about rust (and derive(Encode)) up front to show that we had some shared ideas about how to do this well.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.