Book Review: Steve Jobs by Walter Isaacson

I am a sucker for modern history about the world in which my chosen profession lives.  In this period in our time, we give a lot of credit to Apple, and more specifically Steve Jobs, for driving the last three decades of technological evolution.  There is a theory in Anthropology that modern history is very inaccurate due to the fact that we all have a predisposition to bias our own feelings too much.  Fifty years from now, we will be able to look back on this biography and decide how important it really is. 

What is likely to stand the test of time is the example Jobs set by placing Products before Sales when attempting to ensure corporate longevity.  By simplifying a product line, Apple was able to focus on its core competencies and shed itself of lateral growth and compartmentalization that has plagues other large companies such as Microsoft, Sony, and Yahoo.

What is likely to NOT stand the test of time is the assertion that Apple really had such a large impact on the industry.  Being nimble enough to be a first-mover on capacitive touch screens doesn't mean you can claim credit for inventing a product that was already being developed by Microsoft and Palm.  Being a UI innovator doesn't mean you can claim credit for inventing a product to play music that already existed.  The same is true from the original Macintosh (the Mac wasn't the first PC) to the iPad (Microsoft had tablets for over a decade before Apple).  Sure, Apple has a habit of coming out with products that are among the best in their class, but I was very put off by the conclusions of the book that claim that Apple invented the classes it so frequently dominated over the second half of Job's reign.

Walter Isaacson tells a really good story and has an uncanny ability to research and communicate.  His frankness and honesty, even though he spent to much personal time with his subject, is amazing.  But, even while he discusses the "Reality Distortion Field" over and over again, he seems to get trapped in it from time to time, especially when describing details fewer than five years old where his own predisposition to bias that reality distortion field short of hard facts.

Book Review: God's Battalions: The Case for the Crusades

History books, especially those about the period of time before The Enlightenment, are always biased based on popular views and prejudices.  Among those controversial events was those of the Crusades - which traces its roots all the way forward to the current Israel/Palatine/Middle East fighting the persists today.  This book is an attempt to remove the religious bias that accompanies the study of post-Islamic middle east to provide a geopolitical study of why Europe kept sending people east.

The author attempted to make "a case for the crusades", which I think was a failure.  He didn't convince me that the crusades were a good idea, but the author did convince me that other people were convinced.  Moreover, he convinced me that his arguments were a much more likely set of arguments than those that are traditionally taught - religious extremism coupled with total anarchy.  These traditional arguments I always found hard to believe given the multi-national cooperation of Europe, Byzantine, and both the Latin and Greek churches.

Before reading this book, I had an American high school knowledge of the Crusades, which is to say, very little understanding of what happened and why.  So, I can recommend this book as an introductory text on the subject, but I can't attest to its completeness or validity.  Still, it was an easy read and quite enjoyable.

Introducing QUnitTestDriver

For the past month or so, I've been working on a library to make JavaScript testing integrate better with my company's Java build (junit/maven/jenkins).  All of the options we've tried have been failures because they either don't work well with jQuery or they require an external browser instance which makes them slow and buggy (Selenium, HTMLUnit end-to-end, js-test-driver).

What we really wanted to do was run very fast (sub-100ms) unit tests written in the language of our production code.  This lead us to qUnit.  The problem with qUnit is that there wasn't a convenient way to exercise those tests inside of our normal build easily.  Sure, there is the qunit-test-runner project, but that doesn't come along with any IDE integration.  It requires the tests be run from within an Ant script.

Enter QUnitTestDriver.  It's a simple library allows you to create a plain old jUnit test in any JVM language you choose that points to a plain old qUnit test suite HTML file.  Under the hood, the library lights up a jetty server to avoid "file://" URLs, and uses HTMLUnit to parse the output of the qUnit test results.  Here's an example Java class that would run an entire qUnit test suite that could contain many tests inside of many different files:

public class QUnitTest { @Test public void testPage(){ QUnitTestDriver.run("path/to/qUnitTest.html"); } }

That's it. I hope you find this useful!

You can find out more about the project on it's new Google Code Site Here.

Book Review: jQuery Mobile

Done.  A quick read.  I think that the jQuery Mobile API, unlike the regular jQuery API, is more "invasive".  jQuery Mobile really alters the way we are supposed to work with HTML - adding enhancements to HTML5 to better support mobile browsers.  I think this is a good thing.  The jQuery Mobile team really did a good job adapting HTML/CSS and Javascript to smaller devices.  One warning: this book is very Apple-heavy.  This really isn't a big deal, but I think that as this new book gains mass acceptance, Android will be the predominant platform developers will want to target first.  These platforms are, for the most part, interchangeable, but the screenshots are a little deceiving to those who aren't really familiar with the IOS browser which appears to bleed navigation into the page a little more, something I don't really like.

I thin that anybody who is writing mobile applications really should START here and move slowly to the world of apps.  This is a great books for developers familiar with jQuery and Javascript, but those uninitiated to web development might need a little introduction to Javascript and HTML first.

Universal WR7: Say Goodbye To Your Stack Of Remotes

If you're like me, you (at least used to) have a stack of remotes you juggle to watch TV, adjust the volume, switch between devices, etc.  Even if your cable company provided or your television came with a "universal" remote, those "universal" remotes only work with a approved pre-programmed list of manufacturers and models.  Roku?  Nope.  Asus O!Play?  Nope.  Cheap Chinese HDMI switch?  Yeah right!

Enter the Universal WR7.  There are a lot of remotes that bill themselves as "Universal" especially a few knockoffs from RCA and brand-X comapanies on one end of the spectrum that are no better than the remote that comes with most cable company's boxes or perhaps your TV.  On the other end of the spectrum are Harmony remotes that require a Windows or Mac and hundreds of megabytes of software updates to program and cost hundreds of dollars.  What everybody should buy is a remote from Universal.  Unfortunately, all remotes say "Universal" so you have to look for this logo:

How is this different?  Well, this company makes remote controls that cost anywhere from $10 to over $500.  All of them, as far as I know, are "learning remotes".  This means that the remote has an Infrared Receiver capable of listening to another remote control and mimicking back the signal.  This is huge because the remote can learn to behave like nearly anything that sprays Infrared which not only includes anything you point at your TV to do something, but also all kind of other home automation products.

Is it complicated to set up?  Not for me, but I have a degree in Computer Science.  Here's how you do it:

1. Let's say we want to program the "1" button on the "dvd" setting.  Press and hold [dvd] and [ent] at the same time.  
2. Then press [1].  
3. Then point your DVD remote at this remote and press [1] on the old remote.  The new remote flashes three times and your done.  
4. You can continue to program more buttons by repeating steps 2 and 3.
5. Press and hold [dvd] and [ent] at the same time to save.

Pretty easy!

This remote also supports:

• Punch through.  Any button can be programmed to use the button from a different mode.  For example, you can say that volume controls always go to the receiver or TV regardless of what mode you are on.  Channel changes can always go to the cable box even if you are in DVD mode.
• Macro Mode.  You can program a button to send more than one signal to more than one device.  Example: Turning everything on and off with one button.
• Old fashioned setup like a normal universal remote (the old three digit code and a giant table in the back of the manual or "press up over and over until the device turns off to search for the right code").

So, for those of you who have been wishing for a $200+ Logitec Harmony, run down to target and grab the Universal WR7 for $30, or save yourself the trouble of a complicated setup and get one of the more expensive URC remotes.

Book Review: History Of Western Philosophy

One of a few passions I discovered back in college was law and philosophy.  Lucky me, I actually enjoy reading about this area quite a bit (as you can see if you look back on my previous books reviews).  This is considered by many to be the seminal work on the history of the study of philosophy, so when I discovered it, I had to read it.  I have been listening to this in the form of an audio-book for a few months between podcasts.  The book is divided up into three main sections (unfortunately called "books").

The first 1/3 is a methodological summary of the most important ancient philosophers leading up to, including, and just after Socrates, Plato, and Aristotle.  Most of these ancients set the groundwork for science and philosophy, but have for the most part been thrown out by scientific method and the separation of church and state.  The only real quantitative remaining ancient philosophy appears to be the foundation of western religion.

The second 1/3 is a horrible slog through the beginning of Judaism right through the Roman empire into the 1400's.  This entire period seems to be a scientific stupification of man-kind through the oppression of religious leaders.  But, just as I prepared to give up on the book, it started to get interesting in the final chapter: The Eclipse Of the Papacy.

The final 1/3 starts off with a brilliant summarization of The Renaissance - Man kind finally started thinking! The final third was the only section worth reading if you aren't actually a student of Philosophy.  Much of what was part of cultural understanding was extincted once religion was set aside for logical and empirical development of thought.

As this final 1/3 of the book neared an end, Russel's own philosophy began to come into focus as he spent increasing time refuting and agreeing with the principals of his predecessors.

History of Western Philosophy was written by one of the great authors on the subject from the point of something rare: both an expert and a talented communicator.

Dual Book Review: Desgning Interfaces & Don't Make Me Think

I read these two books over the Christmas Vacation weekend.  Both were great, but serve different purposes.  Don't Make Me Think is teaches us how to discover what works best in the applications we are building while Designing Interfaces is just a catalog of best practices. 

Designing Interfaces is great for developers who have to work in teams and would like a higher abstraction language to use commonly when discussing interface design patterns that might be used to solve specific design problems.  This is as important as software engineering design patterns (flyweight, singleton) that can encapsulate a lot of information into a single word to make design decisions easier to make for a team.

Once those decisions have been made, then the practices outlined in Don't Make Me Think let us determine whether our assumptions about the users is correct.  Designing Interfaces is best skimmed through once and kept as a desk reference and Don't Make Me Think is best read cover-to-cover a few times over for content.