Education is
just barely starting to come out of what I call the "... but on a computer!" [1] era, which is when you try to do exactly the same thing you did before, but now it's on a computer (or a tablet in this case).
There's basically no advantage to taking existing classroom practices, but sticking them on a computer. You'll have to rewrite the entire education process to figure out how to harness computers.
Let me give a concrete example. I am now a fantastic speller, of a language that is one of the harder languages to spell. Am I fantastic because I'm just that awesome? No! Am I fantastic because I have spent years studying lists of words and taking spelling tests? Hell no! I'm fantastic now because every major text input I use now has the red-squiggly line, and the instant I misspell a word and hit space, bam, red line. Backspace & correct. Right-click and select if I really, really can't work it out. (But with backspace & correct actually being way faster, I prefer that.) With instant feedback comes a high rate of learning. I can tell this isn't "cheating", because when I do function in an environment where I don't have this functionality, I retain my spelling skill, the golden standard of whether a skill exists in the tools or the brain. In fact I write many of my blog posts in emacs with no spelling support on, and while I still scan over my posts with a spell checker before posting them, I routinely experience writing several thousand words without a single misspelling, because I have harnessed solid technology to train my brain. I also spell well with handwriting.
This is unbelievably fantastic technology for teaching spelling so efficiently and so effectively that it hardly even need be explicitly taught; sit students down in front of this technology and have them simply write and let the red squiggly handle the spelling training. (You'll still need vocabulary training, but, again, let the red squiggly handle the spelling of the vocabulary while the vocabulary training focuses on meanings.)
By contrast, doing spelling tests "but on a computer" is a terrible way of doing business! First of all, if the students have to "turn in" the test before they get feedback, you've already missed the sub-second feedback rate that would have been so much better anyhow, but second of all, the fact that you're having a spelling test at all is a huge waste of their time. That's the sort of second-order effect we need to be harnessing with computers in the classroom, and we are only at the very, very beginning of that road.
My first grader is just starting spelling tests. He's not quite computer savvy enough to start writing on a computer yet, but you can believe I'll be setting him up with a red squiggly as soon as is practicable.
I still believe computers are going to rewrite education from top-to-bottom, but I find it is proving to be one of the most entrenched, hidebound, nearsighted, and even perhaps terrified of real change elements of society we can possibly imagine. We're so scare of what bad things might happen if we don't stuff a conventional education into our children's heads (moreso than ever, thanks to ever-more test-based education) that we're terrified of experimenting with anything new, and the only acceptable change to the "conventional" education is to precommit even more time to it, making it that much harder to work any experimentation into the system. There are major religions that treat their holy texts as more flexible than the way our culture treats the official line of How Children Are Taught.
[1]: http://www.jerf.org/iri/post/2916