Re: [R] Key colour problem with lattice plot

From: Alex Brown <>
Date: Wed 25 Oct 2006 - 14:42:13 GMT

On 23 Oct 2006, at 23:52, Deepayan Sarkar wrote:

> On 10/23/06, Alex Brown <> wrote:

>> >>
>> >> Can anyone explain why the key is inverted versus the bar
>> >> order?
>> >
>> > Because the key goes from top to bottom, while cartesian
>> coordinates
>> > go from bottom to top. Neither can really be any other way by
>> default,
>> > if you think about it. I did promise someone that I would add an
>> > option to reverse the order in key, but looks like I haven't
>> done that
>> > yet.
>> >
>> In the case of barchart, what would be the justification for not
>> ordering the bars (in a group) as for the key, or vice versa? Please
>> forgive my ignorance - I am curious to understand the design
>> principles. I am certainly very pleased to have the power of
>> auto.key in lattice - it's the main reason why I stopped using the
>> default plot function.
> No particular reason, it's just (I think) one of the weaknesses in the
> Trellis design. There is no actual communication between legends and
> panel functions....

> -Deepayan

Thank you for your excellent explanation. I look forward to your fix.

-Alex mailing list PLEASE do read the posting guide and provide commented, minimal, self-contained, reproducible code. Received on Thu Oct 26 01:19:37 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Wed 25 Oct 2006 - 16:30:14 GMT.

Mailing list information is available at Please read the posting guide before posting to the list.