Free Books

Two free ebooks today.

The first is Breaking the Law of Averages, an introductory statistics book that I recomment. Matt Briggs approaches the subject in a wholly modern way, relying on your computer to do the math, allowing the student to focus on the more important task of understanding what you are doing. Briggs applies objective Bayesian probability theory in this book, allowing for an easy introduction to the subject. This book also has a simple R primer in it.

The second is Modern Perl, a primer on Perl 5.

Paint.net

I have been using the GIMP for a couple of years, but I found that it really had too many features for me. Based on a recommendation from John D Cook, I switched over to Paint.net. Paint.net serves my needs a lot better than the GIMP does. It is simpler, so I feel like I am making better use of the power of the program, and things are easier to find. I just find the toolset easier to use, although it honestly strikes me as more like the old MacPaint than Microsoft Paint.

John D Cook also recommended Inkscape as a good vector graphics program, but I haven't yet had the time to investigate it.

Secret Projects

I have been pretty busy of late. To get an idea of how busy, check out the Reading Log for April as opposed to March. Partly, this is just life, but I have also been working on a secret project that is now ready to be unveiled. I have been redoing the Dead Philosophers Society website, rebuilding it on a more modern platform that ought to make it easier to maintain.

The platform I chose was WordPress. It hasn't actually been too bad, but I have been rather spoiled by Squarespace, which hosts my site. Squarespace is pretty much the easiest web editor I have ever used, and it can do pretty much anything I can imagine, and then some. WordPress in theory can do as much or more, but the work involved is a couple of factors of magnitude larger. You have to go track down the right code or plugin, and then make it work with everything else you already have. It took me about a month to get the Dead Philosopher's site running well, whereas I was up and running here in a few hours.

However, hopefully most of the work was on the front end, and now it ought to be relatively easy to work with. I also got my introduction to CSS, PHP, and MySQL out of this, so now I probably know just enough to be dangerous.

Box, Hunter, and Hunter

I uploaded some datasets from Box, Hunter, and Hunter. It is hard to work through the examples in a stats book if you don't have ready access to the data. I looked around, and I found very little on the Interweb from BHH, even though it is a common book. All of the datasets are from the 1st edition, so if someone wanted to learn some stats, the book would be inexpensive.

I also started with R. R is a language and environment for statistical computing and graphics distributed under the GNU General Public License. R is admittedly a little intimidating even for me, because it all run through the command prompt. I decided to give it a go however, because R is powerful, and more imporantly, free. I use JMP® at work, and I learned JMP in school because NAU had a site license. JMP is really pretty good, it is powerful, easy to use, and well-written. It also costs money, so I don't have a copy at home.

JMP makes output like this:

Pretty nice eh? This is the data from Table 2.1 of Box, Hunter, and Hunter.

I fooled around with R today, and I made this:

Not precisely the same thing, but JMP has it's own way of doing things. I could replicate all of the output that JMP made in R with time. The boxplot is the standard way, so that is easy in R. Over time, I hope to build up a library of datasets from BHH. I hope to learn R along the way, but it is time-consuming at first to learn the commands. Once I get it down, I think I could almost anything with it and R includes many, many more routines than JMP, but they are harder to get to.

LaTeX ReDuX

Well, that was...challenging. I downloaded a common Windows distribution of LaTeX called proTeXt. I installed it, and proceeded to start transforming that piece on Bayes' Theorem into a .tex file. Learning the markup is a little tedious, but everything you need to know can be found pretty quickly. My problem came when I tried to compile the file last night, and I kept getting an error regarding the location of the pictures. I puzzled over that for quite some time before giving up. Turns out the problem was that I was using the LaTeX=>PDF option, which means that you do not use the standard .eps files for pictures, but rather JPEG, GIF, or PNG. Ah. One of those things that is so simple, but not obvious to the new user.

The file turned out pretty well, all things considered. It took a lot longer than I would have liked, but I'll keep plugging away at this. I tried to cut and paste as much as possible when doing this conversion. The idea was to see how the same output could be made with new software. This was hard, because I would have done it differently now, so I had to refrain from copy editing.

If you would like to see the gibberish contained in a LaTeX file, look at the source for this new version. It is just plain text.

LaTeX

Last night I downloaded a LaTeX package. That's 'lay-tech', not the stuff you find in exam gloves or trashy clubwear. One of my purposes here is to play with free software that can be used to produce really nice work. LaTeX is a kind of markup language used to typeset documents. Actually, it would be more accurate to say that it prevents you from having to typeset documents. The idea is to simply write the content, and LaTeX knows how to format it properly, so you don't fiddle with it. It is also very good for scientific works because it handles the display of equations well. I am planning on using LaTeX to redo the Bayes' Theorem paper I posted previously. We shall see how it goes.