I'm having trouble sleeping in Mexico. It could be jet lag but I'm at the point where I should have gotten over it. I didn't have a siesta today and I only got 4 hours of sleep. Rain woke me for the second night in a row. I like rain, rain makes food grow. My roof however was designed to let light in during the day and instead of making it out of glass or thick plastic, it's made out of fiberglass. When rain hits it, it sounds like I'm playing the drums. But the rain isn't the only issue.
I'm working on a few projects that mean a lot to me and I don't get to spend enough time on them. Some of this is explained by life but some of it needs to get fixed. And it will.
If only lack of sleep was good for productivity.
Read more »
Nov 19, 2013
I wanted to post this to my Mexico blog here since I just told people that I'd be blogging there instead of here for the next two weeks, but I decided that a travel blog shouldn't have such an important announcement.
AI3 now has an alpha natural language interface. It doesn't yet query the database like it should, but it will. For now you'll have to be satisfied with thousands of possible conversations. What is a brilliant technology without a tutorial? Here we go.
Read more »As always, I have to post photos of my travels.
Day 1: Beautiful sunset from a house on a hill
Day 2: Me in front of a church
Figure 1: Español Attack v0.1
by Javantea, AltSci Concepts
Text version
Histogram version
I produced this learning tool before coming to Mexico. From basic knowledge of the language to a set of words and sentences I can throw together, this page of common words can help you train yourself what to learn first. I have done this type of language learning guide before. I also created a very nice Twitter - Google translate mashup, but both services were modified, the Google translate API became a pay API which broke my mashup.
This page represents the output of a histogram of words used in a small corpus of Spanish. This mathematical concept is quite easy to understand for linguists: tally each word used, sort by highest tally. I wrote a piece of software that does this for any corpus, including HTML and foreign texts.
Javantea out.
Read more »
