I recently read a book all about trees. This made me realise that I could identify a concerningly small number of trees, especially given how generally outdoor-savvy I would like to consider myself. I used a couple of online tree identifiers but was frustrated by those that don't let you enter the features that are obvious, but rather follow the flow-chart question-answer pattern.

So I wrote my own. I probably could have found a better one if I'd kept hunting (or paid for an app) but this seemed like something I would learn something from building. Also, it doesn't require any faffy deployment stuff, I can just deploy some HTML and JavaScript to GitHub pages and the job's done.

I should possibly have used this as a chance to make something with a fancy JavaScript framework, but I didn't. But it did mean that I did some more fiddly JS stuff. It's not like the tricky bit was manipulating the HTML. The tricky bit was fiddling about with the data. I ended up writing quite a few classes which I think I'm OK with. Python tends to push me gently away from writing classes unless I think they are definitely better than the alternative, purely because it is so easy to make code readable without them. But in languages that I find to be much less readable, classes give a nice way of making code more readable thanks to encapsulation. In addition, I just used jQuery and Bootstrap (because doing DOM stuff with JavaScript without jQuery is yucky, and making any web page look OK without Bootstrap is faffy).

The finished article is hosted here and hopefully it will evolve over time to get more extensive and easier to use.