Re: [Gretl-devel] note on Rlib

I just want to add my 2 cents regarding R integration in gretl. The concern that the relevancy of gretl may be questionable given that R has been so successfull and getting a lot of attention in the recent years has been brought in the gretl conference back in Bilbao. Part of the concern was that gretl has only a handful of packages where as R has thousands of them. I think that R is great but when it comes to econometrics, gretl is probably bigger and this probably won't change in the future as well. This is partly because many of the R packages are known to be "not-too-good" (to be polite) and there exists a very small number of packages relating to econometrics. Also, R packages are perpared and maintained (or not maintained) by different different people so it is difficult to rate the quality as a whole package for econometrics. R is essentially a statistical environment, which represents only a part of what econometrics is. It is true that R provides a great basis for creating econometrical functions and procedures but so does gretl especially in the latest versions and it is getting better and better in this department as well. To my undrestanding, some of the motives for R integration, in the past, has been in order to be able to reach functionality not yet available in gretl. This includes using R's graphical power. Comparing results is probably another important motive as well. In conclusion, my humble opinion is that gretl is quickly becoming a great environment for econometrics and maybe we should aim at making gretl more beautiful and be careful in taking a step that can turn it into a freak of nature.