But my kids? Despite being "digital natives" don't even really know how to use computers. (Yes my failure as a parent, but also indicative of the environment.)
Personally I grew up reading Dr. Dobbs, Byte Magazine, a slew of Amiga journals and Scientific American. I probably started reading the latter at a point where I understood maybe 40% - but it was a great way to learn English.
On the plus side, you can now go and play with and read the code from the likes of Bernstein, Percival, Torvalds and things like the entire Solaris system. And there are great resources, like the Arch Linux wiki, the FreeBSD documentation and dev/user lists. But the good stuff is increasingly hard to discover in an ocean of mediocre stuff. And I have yet to find any good canonical resource for "the good stuff". One of my biggest disappointments with college was that academics appear to be doing a terrible job of keeping up to date, or helping students seek out good extracurricular resources, despite the fact that these resources keep growing and growing.
I hope that the trend toward open publishing of research will continue, and maybe see some sister-resources in the form of free popular science resources pop up -- I could easy imagine universities and research institution pooling resources in order to help researcher edit and publish popular science articles in addition to their standard papers -- as a combination of general science education, and to rise awareness/advertise research.
Have you considered books? For example, this book is shaping up to be the canonical tutorial for Haskell:
I do think it's important to point out that there are a lot of open books out there, facilitated by the web, like the haskell-book. But there's always been good books, it's not enough with out good librarians.
My first set of magazines were the Input series, a British monthly magazine about coding for 8-bit systems.
I invested the majority of my allowance in such magazines, "The C User's Journal" (later C/C++ User's Journal), Dr. Dobbs, PC Techniques, GDC Mag and lots of books.
Luckily my university being highly focused in systems programming, had a huge collection of books and SIGPLAN papers, which opened my eyes for the world of safe systems programming. Thanks to it I was able to delve into the work from Xerox PARC and ETHZ, as well as, the research that was happening with ML (Caml Light was rather new back then) and Prolog.
Input was a great series.