I’ve been thinking a lot lately about how to spread the word about programs, such as MathPractice and ReadingPractice–a couple Tablet PC education-oriented apps I’ve written. The problem is…uh…they are Tablet PC programs. The reality is that most people don’t have Tablet PCs and therefore can’t try out the programs–or don’t even know what a Tablet PC is.
I can provide free or trial demo versions of the apps, but the problem is they’ll only run on a Tablet PC OS. Well, not quite. If they were drawing-only apps, it would be different. The ink license agreement allows an ISV to distribute their app to anyone. A Tablet PC is not required. ArtRage can leverage the Tablet PC ink SDK, for instance, and distribute a version for any XP machine so people can try it out. Of course, the experience won’t be as good as on a Tablet, but a whole bunch of people can learn about the program, use it, and dream of upgrading to a Tablet. This was a good move by the Tablet team.
However, there’s a catch to this. For applications that involve recognition, you’re out of luck. I can’t create a version of MathPractice for a standard XP machine for people to try because it uses the Microsoft handwriting recognition engine and that can’t be redistributed.
Of course I want people to try out MathPractice, PenCircuit and my other programs–Tablet or no–so I’m at a decision point in terms of purchasing a third-party handwriting recognizer or developing a custom one. It would be interesting, for instance, to be able to leverage the xThink recognizer, for instance. I know there are some others too. Of course, Microsoft could loosen the licensing restrictions a bit and allow partner ISVs–that have demonstrated that they are developing for the Tablet platform–to redistribute and install the reco engine on non-Tablets and yes maybe even make some money along the way.
Hmm. More to think about.
The same problem applies to Web-based applications. I can’t do text search in my ink-based Wiki for the same reason. My problem’s a little more tractable than yours, though, since I can have Tablet PCs do the recognition and upload the results, where they’re stored in the database.
Would MathPractice be crippled by the time required for a round-trip to an Internet-connected Tablet PC serving as a “recognition server”?
I’ve wondered about putting together an Internet-based version of MathPractice for a variety reasons, but I’d stopped because of the reco issue. You bring up one possibility–a Tablet PC recognition server. I don’t know if this violates the license or not. Good thing to check out.
The reco needed for MathPractice is a much smaller subset than what other apps would require, which makes the approach of using a custom recognizer not that bad. It’s just that it drives the application away from the Microsoft SDK–which may not be a bad idea since it leaves the door open to porting to other platforms or providing browser-based solutions.