It seems reasonable to say that the best way to develop a killer product might be to make one that you yourself would pay for. One of my recent business ideas (not the one I shared, obviously) was to make a sort of ‘worksheet engine’ that would automatically generate ESL worksheets from a corpus of text.
Naturally, I’ve since had other great ideas (it should manage lists of vocabulary to review, so that those words come up more often in activities, it should make it as easy as possible to re-use work done once already…) I’m focused on implementing the intial idea.
And, while I’ve had very little difficulty with any one aspect of the project (lets hear it for Python!), there is one thing that’s difficult: I’m finding it hard to concieve of a system for categorizing grammar activities.
The idea is that there would be a ‘recipe’ for how they’re made that the program will later be able to follow with a new set of text, generating a worksheet tailor-made to fit the needs of an individual or group.
It’s easy to write something that will remove all articles in a sentence and replace them with a blank space. Or to write something else that will find all the verbs in a sentence and change them to a space followed by the infinitive in parenthesis.
What I haven’t been able to do is to find a scheme to describe these things in such a way that I’m confident that I’ll be able to describe my next worksheet project using it.
Naturally, that means that things are growing as-needed now, and I suppose that’s a perfectly reasonable way to go about doing things. But, I’m a bit frustrated to think that I’ll never be able to be confident that the finished system will be able to quickly realize a new idea.
That is, I think, what the goal of the project is: to know that there’s a syntax by which I can say “this is the kind of worksheet I want to make, focusing on these particular constructions, and with this vocabulary showing up as often as possible” and have it quickly guide me through the steps of turning out a well-designed worksheet.
I guess I’ll just have to call this the beta-version.