[R-sig-ME] Release Dates lme4 Version

Viechtbauer Wolfgang (STAT) wolfgang.viechtbauer at maastrichtuniversity.nl
Tue Mar 11 16:08:55 CET 2014


Great! Looking forward to the new version on CRAN.

Best,
Wolfgang

> -----Original Message-----
> From: Ben Bolker [mailto:bbolker at gmail.com]
> Sent: Tuesday, March 11, 2014 14:14
> To: Viechtbauer Wolfgang (STAT); r-sig-mixed-models at r-project.org
> Subject: Re: [R-sig-ME] Release Dates lme4 Version
> 
> On 14-03-11 06:35 AM, Viechtbauer Wolfgang (STAT) wrote:
> > Hello All,
> >
> > This is actually more of a question directly for the lme4
> > maintainers, but I think others may be interested as well, so I am
> > sending this to the mailing list. I noticed that there are already
> > versions 1.1-0, 1.1-3, and 1.1-4 of lme4 on GitHub (with some
> > fixes/changes that I am quite keen on), while the CRAN version is
> > currently at 1.0-6. I am just wondering whether there are any
> > definite (or semi-definite) plans on when to release a new CRAN
> > version (yes, I have installed the dev version from GitHub; this is
> > more about benefitting from those fixes/changes when using a package
> > that depends on lme4 and that is installed by users in the usual way
> > via CRAN).
> >
> > (and as usual: thanks to the lme4 maintainers for all of their hard
> > work!)
> >
> > Best, Wolfgang
> 
>   Very definite: we're in the process of submitting 1.1-4 to CRAN right
> now (we had to hurry up a bit because there have been changes in R-devel
> and upstream packages that we needed to accommodate).  There are a small
> number of downstream packages that our new release breaks, and we're
> working on contacting those maintainers/helping them fix the problems.
> 
>    The thing most users will notice in installing the new version is a
> wider range of convergence warnings.  Overall we think this is a good
> idea: users who didn't know that their fits were dodgy will find out.
> On the other hand, we also know there will be some false positives, and
> we look forward (!!) to reassuring users ... (this will be a big
> motivation to get improved diagnostic/troubleshooting documentation set
> up, within the package or on glmm.wikidot.com/faq or elsewhere ...)
> 
>   cheers
>     Ben Bolker



More information about the R-sig-mixed-models mailing list