[Issue 27] API Language ADR #38
Conversation
| - Nava and HHS experience with Python and APIs built with Python | ||
| - Number of available libraries and frameworks | ||
| - Free and open source to use | ||
| - Ideal for data computation and complex database queries |
There was a problem hiding this comment.
maybe add that it has superior data science / algorithmic libraries compared to the other two?
| Python as a scripting language is fast and productive. Web-based applications can be created quickly and the code is highly readable. Python syntax is easy to understand, well-defined and extensible. APIs built over python are highly scalable and have an unmatched speed. | ||
|
|
||
| - **Pros** | ||
| - Nava and HHS experience with Python and APIs built with Python |
There was a problem hiding this comment.
Let's add something about broad developer familiarity, outside of just HHS/Nava, because that determines whether it's easy to hire people, whether it's easy for open source contributors, etc.
I would think Python and Node/JS have relatively equal popularity, with Java kind of falling behind, it doesn't seem as commonly used anymore by modern applications.
| - Free and open source to use | ||
| - Ideal for data computation and complex database queries | ||
| - **Cons** | ||
| - Single threaded and can be slower than other options |
There was a problem hiding this comment.
I think Python has multi-threading (if we wanted for some reason) and async support (on newest version)
| - Could use the same framework for frontend, simplifying infrastructure | ||
| - **Cons** | ||
| - Not as mature as other options | ||
| - Responsiveness may be slowed by any CPU intensive computation |
There was a problem hiding this comment.
I wonder if we went with a full-stack JavaScript implementation, if we could extrapolate data-intensive and computation heavy workloads into Python scripts anyway. Or maybe in that case just better to go with full-stack Python if we have a lot of use-case for that.
There was a problem hiding this comment.
Just saying, maybe there's a world where we pick both? :D
Co-authored-by: Aaron Couch <[email protected]>
Co-authored-by: Aaron Couch <[email protected]>
Co-authored-by: Aaron Couch <[email protected]>
There was a problem hiding this comment.
This looks great @SammySteiner I think you captured the pros and cons of going with Python well!
I left a few small suggestions to fix minor typos and had one comment about Python's unmatched speed -- which is at odds with the con about python's single-threaded locking just a few lines down.
|
|
||
| ### Python | ||
|
|
||
| Python as a scripting language is fast and productive. Web-based applications can be created quickly and the code is highly readable. Python syntax is easy to understand, well-defined and extensible. APIs built over Python are highly scalable and have an unmatched speed. |
There was a problem hiding this comment.
APIs built over Python are highly scalable and have an unmatched speed.
I'm a big fan of python, but I'm not sure this is a true statement. Python is outperformed pretty consistently by Go and Node.js even when using a ASGI-based web framework like FastAPI.
Co-authored-by: Billy Daly <[email protected]>
Co-authored-by: Billy Daly <[email protected]>
Co-authored-by: Billy Daly <[email protected]>
Co-authored-by: Billy Daly <[email protected]>
There was a problem hiding this comment.
LGTM! This was a lot of work for a basic decision, but I'm also really glad we're laying a strong foundation for our choices so we can move ahead with speed in the future. Nice work pulling this all together @SammySteiner !
Resolves #27
Discusses pros and cons of three popular api service languages.