So. . . . Here is one of the reasons I really like messing with DSL. . .
Most Linux distro's - even the "tiny" ones, assume you have (relatively) massive amounts of memory and hard-disk space. And on those systems, if you do something sub-optimal, it is likely that you won't see the real impact of your mistake - it gets drowned in the sheer size of the distribution.
In DSL, you can try to run this beastie on hardware that is really ridiculous by any other standard. And if you 'naff-up, the system tells you in short order by crashing, freezing up on you wile you thrash swap, or something similarly obvious. It is because DSL is such a small distribution that it becomes useful - the small size makes it easier to see how the pieces fit together, how they articulate, and how one aspect of the system affects the others.
Also, it's a real challenge to try to get a system that is both stable, and responsive in something insane like a system with only 16 megs of memory.
And it's because you can run DSL on systems that are insane by any other reasonable standard, that you have to really pay attention. I've learned a thousand times more about setting up an efficient system by crashing DSL, than by reading a hundred books on the subject. And it has served me well - I have a 15 Tb server, running Mint/Cinnamon, on a system that reached it's prime in the '90's - and it runs just as well as someone else's 8-core i7 with mondo amounts of memory. It's all about knowing where to trim the fat, and knowing what fat is important to system stability and performance.
And then, there's the sheer challenge of it: Trying to make a system useful that someone else would not even bother to heave out, because it would be a waste of his time and effort.
And that is the main reason I admire DSL. I don't always use it, but I sure respect the hell out of those who maintain it.
What say ye?