[R] Latest revision of lme4 requires 'Matrix' >= 0.995.2 but that version is not available on CRAN

[John Logsdon]

Doug and list

No, the change as reported on my XP PC was Matrix package 0.995-1 was
found but lme4 requires >= 0.995.2.

The format change was from a dash to a dot.  

Maybe a few more would spell SOS!

But it's the sort of thing where sometimes even the geniuses very close to
the work may read the string they think they wrote by gestalt.  As long as
it is what was intended and works, it doesn't matter.

But what about working in logits.  Each new mini-version would have a
constant amount of logit added, ultimately to be changed to the next
release...:-))

Best wishes to all

John

John Logsdon                               "Try to make things as simple
Quantex Research Ltd, Manchester UK         as possible but not simpler"
j.logsdon at quantex-research.com              a.einstein at relativity.org
+44(0)161 445 4951/G:+44(0)7717758675       www.quantex-research.com


On Mon, 23 Jan 2006, Douglas Bates wrote:

> On 1/22/06, John Logsdon <j.logsdon at quantex-research.com> wrote:
> > I've just hit this problem as well and as I am slumming it on XP at the
> > moment, I don't have the compilation tools to use the .tar.gz version.
> >
> > I also notice that the numbering format has changed.  Is this intentional?
> 
> Which numbering format?  If you mean the use of 0.995 rather than 0.99
> it is because I keep finding changes that need to be made in the
> underlying format for mixed-effects representations (i.e. mer objects)
> before a 1.0 release.  The current sequence of numbers is 0.99 (99%),
> 0.995 (99 1/2%), 0.9975 (99 3/4%), 0.99875 (99 7/8%), ...
> 
> It is a pattern like, but not as inventive as, Donald Knuth's
> numbering sequence for TeX releases.  It has the desirable property of
> allowing for an infinite number of releases before you need to declare
> the package to be at release 1.0
> 
> There will be at least a 0.9975 series of minor releases (I need to
> add another two slots to the mer object).
>