Hello, I tried Deducer earlier in the week (on Ubuntu Linux) and encountered some hiccups, but I suspected that it had something to do with the idiosyncrasies of my machine rather than the Deducer package proper. So last night I did a *clean* install of Ubuntu 9.04 and all of my other software. Today, everything runs perfectly, without a hitch, first time.
Some things I am doing which may be relevant (or, not): 1) using the latest version of R (session info below), installed from CRAN according to the Ubuntu README http://cran.r-project.org/bin/linux/ubuntu/ 2) only using package binaries from cran2deb, namely, r-cran-deducer, r-cran-jgr, etc; did not build *anything* from scratch 3) am using sun-java-6, as opposed to some of the other Java implementations (iced-tea, etc) 4) *must* do sudo R CMD javareconf Furthermore, not only is JGR/Deducer working fine, but also rgl works too (something that has caused me trouble in the past). Note, however, that this does not address the 100% CPU issue with JGR reported elsewhere earlier. I have tried Deducer without JGR, and everything seems to be operating according to plan. The JGR console isn't there (obviously), but there are command line selections to navigate through the menus, and they seem to be functioning properly. It is a little bit strange to use the command line menus, though, because once you've made a selection in a submenu, you can't go back one step if you change your mind; you can only enter 0 to exit deducer() completely and start over from the beginning. The JGR console is much more convenient with its GUI and 'cancel' options. In fact, JGR is already loaded as a dependency of Deducer... it would appear to be easier overall for a person to just run JGR() at the beginning to operate Deducer through the JGR console and sidestep the command line entirely. There are apparently other reasons why using command line Deducer is desirable, from other discussions that I have not read. The other question was whether Deducer would be a good thing for introductory statistics, and in my opinion the answer is "yes". The layout of the menus seems reminiscent of something like SPSS, and in particular, many of the plots are integrated into the "Analysis" menu items, rather than having a top menu of their own as in Rcmdr. There are advantages and disadvantages to both approaches. I really like the "Data Viewer". This is great. It makes it much easier to handle factors, factor levels, ordered factors, and contrasts. The main goal in my view is to get the students started, and I think Deducer would accomplish that goal very well. Cheers, Jay sessionInfo() R version 2.9.1 (2009-06-26) x86_64-pc-linux-gnu locale: LC_CTYPE=en_US.UTF-8;LC_NUMERIC=C;LC_TIME=en_US.UTF-8;LC_COLLATE=en_US.UTF-8;LC_MONETARY=C;LC_MESSAGES=en_US.UTF-8;LC_PAPER=en_US.UTF-8;LC_NAME=C;LC_ADDRESS=C;LC_TELEPHONE=C;LC_MEASUREMENT=en_US.UTF-8;LC_IDENTIFICATION=C attached base packages: [1] stats4 splines grid stats graphics grDevices utils [8] datasets methods base other attached packages: [1] Deducer_0.1-0 lawstat_2.3 VGAM_0.7-9 foreign_0.8-36 [5] effects_2.0-3 nnet_7.2-47 MASS_7.2-47 lattice_0.17-25 [9] multcomp_1.1-1 survival_2.35-4 mvtnorm_0.9-7 car_1.2-14 [13] JGR_1.6-7 iplots_1.1-3 ggplot2_0.8.3 reshape_0.8.3 [17] plyr_0.1.9 proto_0.3-8 JavaGD_0.5-2 rJava_0.6-3 loaded via a namespace (and not attached): [1] tools_2.9.1 > -- *************************************************** G. Jay Kerns, Ph.D. Associate Professor Department of Mathematics & Statistics Youngstown State University Youngstown, OH 44555-0002 USA Office: 1035 Cushwa Hall Phone: (330) 941-3310 Office (voice mail) -3302 Department -3170 FAX VoIP: [email protected] E-mail: [email protected] http://www.cc.ysu.edu/~gjkerns/ _______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-teaching
