[statnet_help] Nightly build Windows and MacOS binaries are now available for almost all Statnet packages

Pavel N. Krivitsky p.krivitsky at unsw.edu.au
Thu Dec 24 14:28:10 PST 2020

Dear All,

A Christmas present for everyone!

The current public development versions of the Statnet packages are
hosted on GitHub. These contain bug fixes and experimental features
(but also potentially new bugs) that have not yet made their way to

Until now, if you wanted to use those features and fixes, you needed to
either wait until the package made its way to CRAN or set up a
development environment (compiler, etc.) to install these development
versions. Until now.

For the following packages, you can now download the Windows and MacOS
binaries for the latest public development versions from the links in
the README file on their respective GitHub pages:

* ergm.ego
* latentnet
* network
* statnet.common
* tergm
* ergm
* ergm.count
* ergm.rank
* rle

Typically, the GitHub URL of the package is of the form
https://github.com/statnet/PACKAGE . Please go there and follow the
instructions. (In particular, for technical reasons, the installable
.zip and .tgz files are stored in zip files, from which they need to be
extracted before installation.)

Those of you using Linux will still need to compile yours, but this is
the case even when installing from CRAN.

A few caveats:
* These builds have not undergone the level of testing that packages on
CRAN have; and see above about the bugs.
* We generally only update the NEWS file before the CRAN release, so to
see a full list of what has changed since the last release, you will
need to look at the commit log.
* They may break some packages that depend on them (such as those for
pretty-printing model fit results), since their developers have not had
a chance to update their packages in turn.
* Since these versions are, in some sense, transient, results obtained
with them may not be reproducible without additional documentation
(e.g., the Git commit ID).
* As always, when running software downloaded from the Internet, standard
warnings apply. I CANNOT GUARANTEE that the third-party tools and apps
involved in getting those compiled binaries to you have not adulterated
them in any way.

I hope this helps,


More information about the statnet_help mailing list