Django shop owner here who would like to ask other (ex/current) Django developers, who made (or not made) the change to the Phoenix/Elixir world, to share real world stories about what went good and what went not so good.
We have been developing Django applications for more than a decade and we enjoy all the bells and whistles and also the accessibility of Python.
While I also have some fond memories with Erlang (game backend has been running for many-many years) giving a lot of confidence in the OTP platform and its concurrency model. Changing languages and frameworks seems quite scary.
I enjoy reading posts like
https://news.ycombinator.com/item?id=32530203 and it's been a while since it’s been asked.
It would be great, if you could share some recent examples and what pain point you had that you could accomplish better or worse in either of the worlds.
One thing that we hope to get from developing a project in Phoenix is that ECTO is much closer to SQL. While building a financial product where not just OLTP but also OLAP like dynamic queries need to be developed and maintained. It's a constant pain to get Django's ORM to behave the way we like and not put extra joins on the annotation/aggregation. Lot of the time we do think in SQL and then translate this into ORM, mostly to be Django friendly and to keep the safety checks from the framework.
If you read so far and have some React vs LV insights, that would also be great, although that’s for another Ask HN :)
We have been developing Django applications for more than a decade and we enjoy all the bells and whistles and also the accessibility of Python. While I also have some fond memories with Erlang (game backend has been running for many-many years) giving a lot of confidence in the OTP platform and its concurrency model. Changing languages and frameworks seems quite scary.
I enjoy reading posts like https://news.ycombinator.com/item?id=32530203 and it's been a while since it’s been asked.
It would be great, if you could share some recent examples and what pain point you had that you could accomplish better or worse in either of the worlds.
One thing that we hope to get from developing a project in Phoenix is that ECTO is much closer to SQL. While building a financial product where not just OLTP but also OLAP like dynamic queries need to be developed and maintained. It's a constant pain to get Django's ORM to behave the way we like and not put extra joins on the annotation/aggregation. Lot of the time we do think in SQL and then translate this into ORM, mostly to be Django friendly and to keep the safety checks from the framework.
If you read so far and have some React vs LV insights, that would also be great, although that’s for another Ask HN :)