You’re right Peter. At least from my perception, I’m quite novice, Homebrew is quite good covered and supported. Besides, right now R is part of Homebrew core, so there is a good amount of people taking care of the installation scripts. I’ve had myself some problems in the past, and rJava and Java, like any other installation, give you problems from time to time. But usually are they are fixed in a couple of days.
It’s very true that if you use the Homebrew to install R, you aren’t free of problems down that road, and you need to know more or less what are you doing. When you compile R you need to compile most of the packages you download and sometimes there is problems from time to time. Anyhow, I agree, that the more people knowing R at lower level the better. There are several tutorials out there about how to install R with Homebrew and get you what some people call a “100% Homebrew R Edition”. I want to write my own soon since there is none there for High Sierra. If people is interested I can share here too. Finally, have a good Finnish Independence and Spanish Constitution day to you all. Best Regards Luis Puerto http://luisspuerto.net > On 6 Dec 2017, at 15:27, peter dalgaard <pda...@gmail.com> wrote: > > Homebrew is not particularly unrecommended either, but you could get in > trouble with compatibility vs. CRAN package binaries. "Unsupported" is more > precise: None of us are trying these builds so if they break, you could be on > your own. (From the Open Source perspective, the more people who know how to > deal with R at source code level, the better; and homebrew does appear to > have a substantially better reputation than the older MacPorts system did.) > > From that point of view it could be easier for you just to grab R-patched > from r.research.att.com. Those binaries are unsigned, though, so you need to > make a leap of trust that noone has hacked their way in and replaced the > binaries since they were built. Pragmatically you need to download, open the > Downloads folder (in Finder!), right-click the package, choose to use > Installer.app, and accept to open it even though yadayadayada in the dialogue > window. > > -pd > >> On 6 Dec 2017, at 11:24 , Luis Puerto <luiss.pue...@gmail.com> wrote: >> >> Hi Federico, >> >> I know that it isn’t recommended by R Core Group, but you can always instal >> it from source using Homebrew. I’ve believe that there is even a bottle >> already and you don’t need to compile. I’m not sure. >> >> I’ve using a pure Homebrew install in my Mac and runs faster then the CRAN >> one, with openBlas and OpenMP. >> >>> On 6 Dec 2017, at 11:26, Federico Calboli <federico.calb...@helsinki.fi> >>> wrote: >>> >>> Hi All, >>> >>> I know that I can get the latest and greatest binaries from >>> http://r.research.att.com/ but I am puzzled why there are still no OSX R >>> 3.4.3 binaries on CRAN. The Windows installer has been up a few days >>> already. Is there anything I am missing concerning this delay? >>> >>> Cheers >>> >>> F >>> >>> _______________________________________________ >>> R-SIG-Mac mailing list >>> R-SIG-Mac@r-project.org >>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac >> >> >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> R-SIG-Mac mailing list >> R-SIG-Mac@r-project.org >> https://stat.ethz.ch/mailman/listinfo/r-sig-mac > > -- > Peter Dalgaard, Professor, > Center for Statistics, Copenhagen Business School > Solbjerg Plads 3, 2000 Frederiksberg, Denmark > Phone: (+45)38153501 > Office: A 4.23 > Email: pd....@cbs.dk Priv: pda...@gmail.com > > > > > > > > > [[alternative HTML version deleted]] _______________________________________________ R-SIG-Mac mailing list R-SIG-Mac@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-mac
