This week’s assignment is to discuss using a pre-configured VM instead of building our own, for the purposes of this class.
I think that the repetitive nature of installing new VMs from scratch was generally good practice, and I don’t begrudge the time I spent repeating the same process multiple times throughout this class, and 672 before it. Especially since our instructions for completing each assignment varied (e.g., editing the repositories, using wget, using FTP, etc.), I think that by the end of the class, we’ve gotten to practice a number of different ways of doing the same thing and generally have an idea of our preferred methods and their pros and cons.
From that pedagogical perspective, it might have been interesting to vary the new VMs — running different distributions for each install, perhaps, instead of doing the same basic Ubuntu Server install over and over. I really liked having one unit (I think it was DSpace?) that used Tomcat instead of Apache, and PostgreSQL instead of MySQL. A lighttpd one might have been neat, too. On the other hand, I imagine the difficulty troubleshooting such different setups would have been frustrating for many, and for those with slow internet connections, downloading multiple distros would have been very time-consuming.
If the advantage to using pre-configured VMs would have been getting to spend more time playing with the installed program, I would argue that taking the time to install and configure that program gives a lot of depth and insight into its workings that merely using it pre-installed wouldn’t add.