you could completely respect the decision to deprecate some support, but the marketing speak comes across distasteful.
from a user's perspective, selecting a tool should be about what's "best" for you and not what is the "best" out there (from some arbitrary measure nonetheless).
pandas made sense and might still make sense for many, especially who are migrating from R and are already comfortable with dataframe paradigm. using the same paradigm and just saying "we run faster" does not solve the ergonomics or the reason why people adopted this way of processing data.
there are plenty players in the space going about this from the stack. would be nice to see more discussion on alternative generic data processing paradigms instead.
this is a blog for the Ibis project, directly from one of the main developers/maintainers of the project, about deprecating two (of 20+) backends the project supports and giving the justification
as a project, we're big on using the right tool for the job -- of course we often think that's Ibis. a lot of effort has been put into the ergonomics, taking learnings from pandas and dplyr and SQL and various other Python data projects the team has been instrumental in. if you weren't aware, the creator of pandas also created Ibis and the current lead developer (Phillip Cloud) was also very involved in pandas. we have plenty of other blogs and things on the documentation about the project's ergonomics and thoughts on data processing paradigms
this specific blog isn't intended as "marketing speak" or to somehow be distasteful, it's intended to inform the Ibis community that we're deprecating a backend and justifying why
Pandas support as input and output isn't going anywhere. People can continue to bring their pandas DataFrames to Ibis and run Ibis expressions against them.
> pandas made sense and might still make sense for many, especially who are migrating from R and are already comfortable with dataframe paradigm.
Yes, and nothing is different about the paradigm after the changes outlined in the blog post, it's still a dataframe paradigm.
> using the same paradigm and just saying "we run faster" does not solve the ergonomics or the reason why people adopted this way of processing data.
This seems to indicate you _do_ understand that the paradigm isn't changing but then does a complete 180 into complaining about Ibis not solving pandas ergonomics issues. Which is it?
from a user's perspective, selecting a tool should be about what's "best" for you and not what is the "best" out there (from some arbitrary measure nonetheless).
pandas made sense and might still make sense for many, especially who are migrating from R and are already comfortable with dataframe paradigm. using the same paradigm and just saying "we run faster" does not solve the ergonomics or the reason why people adopted this way of processing data.
there are plenty players in the space going about this from the stack. would be nice to see more discussion on alternative generic data processing paradigms instead.