A very critical error on the installation documentation (I think)

[alvations]

On the installation page, it says:

"We don’t recommend installing scipy or numpy using pip on linux, as this will involve a lengthy build-process with many dependencies. Without careful configuration, building numpy yourself can lead to an installation that is much slower than it should be. If you are using Linux, consider using your package manager to install scikit-learn. It is usually the easiest way, but might not provide the newest version. If you haven’t already installed numpy and scipy and can’t install them via your operation system, it is recommended to use a third party distribution."

Isn't it suppose to be We recommend installing scipy or numpy using pip on linux?

[GaelVaroquaux]

Isn't it suppose to be We recommend installing scipy or numpy using pip on linux?

No, it's not. We _do not_ recommend it.

[alvations]

Okay, so it's you recommend to use conda or other package manager. Thanks for the note. I thought I was seeing things =) I think it's good to refer to consider using your package manage, e.g. conda, to install scikit-learn

[GaelVaroquaux]

Okay? So it's you recommend to use conda or other package manager?

Yes. There reason being that scipy and numpy need tricky compiled libraries that are outside the scope of Python packaging. The formulation might be unclear. If you feel you can improve it, we'd love a pull request on the other repo (the main scikit-learn one). Thanks!

[alvations]

What if we change the wording to:

If you are using Linux, consider using your package manager, e.g. 
*yum, apt-get, dpkg, zypper, etc.*, to install scikit-learn. It is usually 
the easiest way, but might not provide the newest version.

Is that correct?