teh hayley

Mark it. Yesterday was the first day that I didn't make my 1,667 daily word quota. Now, if I play by the NaNoWriMo rules, I'm still good, because I had enough days where I went over that, on average, I'm still at the 1667.

Today's todo list is going to look fairly similar to previous days with the exception that I got some good work done on the BMI calculator yesterday.

As mentioned yesterday, I was getting really tripped up by how to handle the BMI ranges in a don't-repeat-yourself (DRY) fashion.

My first task was to declare whiteboard bankruptcy and erase everything so I had some actual room to work.

Whiteboards are fantastic. I tend to think it's the ease of erasing, the non-permanence, the ability to write in pseudo code... all of these things seem to make it perfect for tackling mind boggling programming problems.

Or maybe it's the fact that you're standing up and moving around as you write. Or maybe it's the fact that you're a little high from all of the marker stench. Whatever it is, it made it so much more simple to break down what it was that I was actually trying to accomplish.

And since I did figure out a DRY way to do things (the if/else if block that preceded it was really ugly), now I can easily implement support for the country specific BMI ranges (US, Japan, Singapore).

The whiteboard was also a huge help when it came down to trying to solve the math of... I know my height, and I know my target BMI, how do I find my desired weight?

It was kind of insane. The steps in flipping an equation around were completely escaping me. I had to google it.

And even then, I managed to completely botch the numbers somehow. Maybe it was a simple typo, I don't know since I used Alfred's built-in calculator to do it (it doesn't store a history).

But at the whiteboard, everything was clear and I nailed it on the first try (I still have no idea what I goofed up at the computer).

I've got some visualization ideas for the calculator, but I think I'm going to leave it for now with one last todo item.

implement the ability to enter height/weight in multiple formats

Instead of how other calculators make you input your height and weight into a completely different box when you watch to switch between metric and imperial, I think I can do better. I'll just have certain formats for entering weight for instance:

Something along those lines. And then I'll support different height measurement units by doing something similar.

finally roll out the integrated IMDB scraper site

I got a start on it last night. And by start, I mean I copied a bunch of the files over... that was really it. Election coverage got in the way after that. And then sleep got in the way after that.

Now, that I've resigned myself to just spinning off the current code into a different site instead of trying to figure out how to manage it all from one code base, it really should be decently trivial to implement. Like a 30 minutes to hour job.

Of course, I'm also terrible at estimating project time. However, I generally tend to overthink things and plan for them being a massive effort with lots of head scratching. Ya know, things that, once you just freaking start, are often way easier than you were expecting.

MVP of the custom quiz on Hack Spanish

I've mentioned previously that one of the HUGE problems I'm dealing with right now on Hack Spanish is that I'm getting zero search engine traffic (this wouldn't be such a big deal if I actually knew how to promote my own sites instead of relying so much on google).

Presumably, the problem is that to google (and the other search engines, but they still matter far less unfortunately), there's hardly any content on my site at all because it's all being injected at runtime via JS.

But thinking about it, I realized that Angular might actually present the perfect solution. As I understand it, a search engine will see your raw Angular template (with things like {{variable}}), though I haven't seen this confirmed yet.

What I'm wondering is how difficult it would be to basically just show your entire set of quiz questions as part of the "raw" HTML page and then reveal them one by one using JS. It's surely doable. And then the search engines would presumably have something to look at, and something to deem your site worthy of appearing in search results.

It's definitely something I need to try.

At the very least, I want to get my custom fill-in-the-blank quiz going today.

[try low carb][] recipe restructuring

This was on the list a couple of days back, but to recap, the site is currently built in an unsustainable way (basically too much duplication of data across multiple files). I'm wanting to dig into document_mapper to see if I can basically turn my individual recipe files into a database equivalent. Then I know I'll be more motivated to update the site, because I won't have to update the data in 3 million separate places.

Currently, this site appears to be doing the "google dance" that you'll so often hear about (if you follow the SEO crowd, of course). It's danced its way completely off the map as far as I can tell (based on a unique phrase that it previously ranked #1 for).

Such is the life of someone who's never learned how to self promote. I'm completely dependent on the mighty google to bring me traffic. Well, to a point of course. Like, how once people see Cyclocane, it gets shared elsewhere. But it was still that initial google search that made them aware of the site in the first place.

And I've made this blog a lot about self promotion in a way, by actively talking about what I'm working on.

But no one comes to this blog either, so that's not much help.


teh hayley