On 10 June 2011 21:59, Joe Cheng [RStudio] <j...@rstudio.org> wrote: > Sebastian, how simple is simple? If it is REALLY simple (i.e. a bunch of > scalar parameters, checkboxes, and drop-down lists that feed a plot > function) then consider RStudio's manipulate package: > http://www.rstudio.org/docs/advanced/manipulate > I've seen manipulate, and it is very nice. But in my case 'simple' is more the workflow and therefore GUI design, multiple interactive plots being visible and some update handlers are necessary i.e., some options modify data, others only affect plot features, with the large data-sets the resultant/temporary data-sets must be cleverly cached. For some features manipulate would probably suffice, but I don't want to become tied into RStudio just yet -- maybe I'll prototype some of the interactions with manipulate. I will let you know how it turns out, it will probably be a while before I know what is working well though.
Cheers, Seb > > <http://www.rstudio.org/docs/advanced/manipulate>Note that this will only > work in RStudio (both desktop and server); the manipulate package simply > won't work in any other environment. > > > On Fri, Jun 10, 2011 at 1:32 PM, Sebastian Mellor <seb...@sebble.com>wrote: > >> I agree, if we were hoping to load a lot of data into an interactive table >> or provide a more tailored experience with a smooth UI, tooltips, dynamic >> context menus and customised layouts I would opt for a more powerful GUI >> language (or at least move down a layer or two gWidgets -> Gtk/Qt/Java), >> but >> in our case the functions to wrap are comparatively simple (not combined >> in >> any complex overly complex ways) and fairly well parameterised, the plots >> should provide most of the visual information feedback. I may yet find a >> need for more control. >> >> [[alternative HTML version deleted]] _______________________________________________ R-SIG-GUI mailing list R-SIG-GUI@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-gui
