[R] Dirty Rotten Hack. (reversing tickmarks on axes?)

[Peter Dalgaard]

"Allen S. Rout" <asr at ufl.edu> writes:

> I feel dirty.
> 
> 
> I have some graphs I'm building to communicate chargeback rates and service
> usage for our backup system here at the University of Florida.  These come
> down to daily data points on a graph of number-of-bytes transferred and
> stored. 
> 
> Since we chargeback on the same basis (price per MB this, price per KB that)
> the same chart with a different scale can be used to communicate bytes and
> dollars. I set about trying to accomplish this like so:
> 
> http://nersp.nerdc.ufl.edu/~asr/media/r-foo/try1.png
> 
> Those axes are a little messy. I tried nudging them around
> 
> http://nersp.nerdc.ufl.edu/~asr/media/r-foo/try2.png
> 
> which is better but not good.  What I really want to do is tell my axis()
> function to reverse the tick direction:  put your ticks and labels "inside"
> the graph.  Something like
> 
> http://nersp.nerdc.ufl.edu/~asr/media/r-foo/dirty-hack.png
> 
> which I accomplished by telling axis() 'line=-47.7'. 
> 
> Eugh.
> 
> Note that the distance between the left side and the right is different
> between the right side to the left. :)   I don't particularly object to this,
> when you abuse a tool in this manner you need to expect oddities.
> 
> I've wandered through the mailing list logs, and haven't seen reference to
> this particular desire.  Am I alone? :) 

Possibly in space, but not in time and space: Someone seems to have
been there before. Check out

plot(0,tcl=.5,mgp=c(-3,-2,0))


-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907