4 ms·
I like the idea of using a crypto lib, but it will probably take some thought and experimentation to figure out what will fit into roughly a ~1hr exercise. That
by joncalhoun 9y ago
I like the idea of using a crypto lib, but it will probably take some thought and experimentation to figure out what will fit into roughly a ~1hr exercise. That is always the hard part - not picking something too large.
- Vendan 9y agoGo's crypto libs are pretty well made, and the ssh libs are less crypto and more "here's an interface similar to os/exec that lets you run commands on a remote system"
- distrill 9y agoWhy limit the complexity/interestingness of everything to always fit into a single hour? These seem to be aimed at people who have finished introductory material and are potentially looking for (semi-)/nontrivial material. If you're married to the idea of ~1hr segments, you can always split them into parts of a series.
- joncalhoun 9y agoI'm not stuck on 1hr segments, and I am willing to split exercises up, but only when it makes sense. For an example of this, see exercises 4 & 5 - the fifth one uses the code from the fourth. What I am weary of is a single exercise that isn't easy to break up. Or one that even if you break it up, each exercise isn't entertaining or educational enough to stand on its own. I also prefer having exercises (at least for this course) that can be completed in someones spare time - say in an afternoon or evening - and if an exercise takes me 1hr to code in a screencast, chances are it will take someone newer to the language longer and that doesn't account for time learning about new std libraries, googling questions, etc. In short, a 1hr exercise for me might be a 3+ hour exercise for some students.