Conversation
olivielpeau
left a comment
olivielpeau
left a comment
There was a problem hiding this comment.
Looks good!
One oversight though IMO: we should also seriously consider switching to a pure functional programming language for the Agent. I'm concerned that going with an imperative language will not create as much hype as we'd want.
| ## Constraints | ||
|
|
||
| 1. Solution must be packaged with Omnibus Software | ||
| 2. Agent must run integrations in Go and Python |
|
|
||
| - Strengths | ||
| - fast as C, slick as Ruby | ||
| - its Ruby-like syntax will make Omnibus specialists feel at home |
|
|
||
| ## Constraints | ||
|
|
||
| 1. Solution must be packaged with Omnibus Software |
There was a problem hiding this comment.
It's great to see this in the requirements. that's definitely the #1 requirement for a smooth developer experience 💯
|
|
||
| ## Open Questions | ||
|
|
||
| *Note any big questions that don’t yet have an answer that might be relevant.* |
There was a problem hiding this comment.
I'm surprised we haven't considered C, it seems like a strong candidate for a full rewrite of the agent. Plus its hype score is very stable: https://trends.google.com/trends/explore?date=all&geo=US&q=c
|
|
||
| - Strengths | ||
| - fast as C, slick as Ruby | ||
| - its Ruby-like syntax will make Omnibus specialists feel at home |
There was a problem hiding this comment.
What about crystal's release cycle? It changes fast and regularly ships breaking changes (https://github.com/crystal-lang/crystal/releases). Looks like a green light to me.
4 participants
What does this PR do?
Design proposal for a POC of a new Agent. The prototype would help us assessing the current development cycle is the best we can get and will let us plan in advance in case we want to steer the project in a different direction.
Rendered link