H/T @pwang
-
-
-
We're always hearing from non-data folks that conda is redundant; we should just use pip. Thing is, we tried that. It didn't work.
-
things have changed hugely, and for Python packages (not pure-Python) pip+wheels+manylinux works beautifully.
-
How about different matplotlib versions/backends in virtualenvs? That was another of my pain-points in pip that conda solved.
-
to be fair, matplotlib might not be well-structured in that respect.
-
it may be that matplotlib depends on non-Python components. pip/virtualenv are not well-structured for that :)
-
wheels, etc. are a really good enhancement, but from what I've seen they're too-little too-late.
-
It's not too late to use and improve common tools. Too many splits: Py2/Py3, pip/conda, CPython/pypy/jython/pyston
- 4 more replies
New conversation -
-
-
I love conda, but I wish it were more compatible with the virtualenv ecosystem (tox etc). Lots of friction at my last job.
-
Me too! Begging for community helphttps://github.com/conda-forge/staged-recipes/issues/1139 …
-
I know
@brandon_rhodes has a hack for this too. Would be nice if we could just fix it.
End of conversation
New conversation -
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.