I’m getting ready to spin up a few new web development projects. I think I want to do them on Ruby On Rails. That means getting a good RoR development environment installed on Finrod.
About 5 levels of yak-shaving later… it looks like I should install RVM so I can get the version of Ruby I want. And most instructions on installing RVM say, “point a command-line web client at this URL and pipe the result into your execution shell.” To which my first thought is, “Oh, hell to the no! Shoot some unknown, untested, executable code straight into an interpreter? What kind of moron do I look like?” (Don’t answer that. It was rhetorical.)
Upon further investigation, this really does seem to be accepted practice nowadays. And I wondered what’s happened to the days when we’d download a tarball that has an autoconfiscated install package with a makefile and all that?
Hmmm, you know… speaking of untested code that I don’t examine before installing…
I guess this newfangled way of doing it really isn’t any less secure than what we were all doing back in the late ’90s. It’s just… honestly, it may well be more convenient, with fewer unnecessary steps (like unpacking that tarball), and it may result in fewer files lying around cluttering up my hard drive afterward.
I’m not sure. I’ll re-evaluate that after I’ve actually done it.