http://www.solie.ca/index.htmlBlog Updateensteven@solie.caCopyright 2009 Steven Solie2010-12-19T13:07:12-07:00 hourly 1 2000-01-01T12:00+00:00 Sun, 19 Dec 2010 15:29:21 -0700steven@solie.caNone2010-12-19T13:07:12-07:00http://www.solie.ca/files/2b72690664d26d3135dc7cf5db428dcd-13.html#unique-entry-id-13http://www.solie.ca/files/2b72690664d26d3135dc7cf5db428dcd-13.html#unique-entry-id-13Cat herding (managing software developers & engineers) is difficult enough. It certainly doesn’t help that your cats... er... people become demotivated when offered things like large cash bonuses. Why on earth would a large bonus not motivate people?

This is what I wanted to learn more about and this is why I picked up Drive: The Surprising Truth About What Motivated Us by Daniel H. Pink.


I discovered this book after attending a Calgary Agile Methods User Group seminar by Jonathan Rasmusson. His seminar was entitled The Surprising Science Behind Agile Leadership – Why We Do What We Do and you can find the slides at the link.

Plenty of scientific research has been done on what drives us to do what we do. The basic biological human drivers include hunger, thirst and sex. Responding to rewards and punishments in our environment is a second long understood drive we all have. The surprise comes from the third drive which some call “intrinsic motivation” which causes us to behave rather irrationally at times.

What can happen when the second drive is leveraged (a large cash bonus) and the third drive ignored (no autonomy) is that people can become severely demotivated. This intrinsic motivation is so strong that it can even give you more of what you don’t want: unethical behaviour, addictions and short-term thinking.

The third drive is not universally applicable which is likely why we don’t understand it as well as the other drives. You need a situation in which people are comfortable (paid enough) and use their brains a lot (knowledge worker). This pretty much describes life in many software development organizations although some will argue they are never paid “enough” of course. The point is that the intrinsic motivation dominates when the other needs are already satisfied. And yes, you want this intrinsic motivation to be satisfied.

Working crazy hours for nothing, coming up with fantastic solutions, higher quality, contagious enthusiasm, incredible dedication and smiles all around. This is what you get for free with satisfied intrinsic motivation. All the things the large cash bonus was hoping for.

Drive explores how to enable this elusive third drive and avoid the negative behaviours you get with disincentives. The trick is knowing when intrinsic motivation is the dominant factor in any situation. You don’t hope for intrinsic motivation to kick in at the local burger joint for example--wrong venue.

I found the book is very approachable so I refer anybody interested in this topic to give it a quick read and make up their own minds. It is rather short and includes some fluff at the end to pad it out. The one thing I really didn’t like about the book is the operating system analogy like Motivation 2.0 and how it needed an upgrade to Motivation 3.0. Being a software guy, the operating system analogy was annoying but tolerable.]]>steven@solie.caNone2010-11-23T22:12:19-07:00http://www.solie.ca/files/8aad8f29c6cdbd3fbd41194ed2f53723-12.html#unique-entry-id-12http://www.solie.ca/files/8aad8f29c6cdbd3fbd41194ed2f53723-12.html#unique-entry-id-12
Thanks to a patch from Richard Smolak (aka Cyphre), the REBOL 3 style browser now runs as well. The style browser is used to help test the REBOL GUI.

style-browser.r3
]]>steven@solie.caNone2010-11-18T21:45:26-07:00http://www.solie.ca/files/2793e262caf36dcaf2867dae91e80e73-11.html#unique-entry-id-11http://www.solie.ca/files/2793e262caf36dcaf2867dae91e80e73-11.html#unique-entry-id-11FreeType instead of the proprietary Windows font system. Another big thanks goes out to Michael Trebilcock who provided me with the latest FreeType port.

Additional changes were needed because of the assumed wide character support and also some lingering little endian issues. There is also some work left to do to enable bold and italics. We still need to resolve the issue of font naming as well since default Windows fonts are not available on other platforms.

The following are the results of some of the tests:

text-test2.r


Note in the test above there is still an issue with the calculation of line length when using FreeType fonts.

caret-test.r
]]>steven@solie.caNone2010-11-09T22:34:20-07:00http://www.solie.ca/files/1fdcd718de2d3d2a237afe02d43906b9-10.html#unique-entry-id-10http://www.solie.ca/files/1fdcd718de2d3d2a237afe02d43906b9-10.html#unique-entry-id-10REBOL host kit test scripts. The scripts are being displayed in simple AmigaOS windows.

draw-test.r


shape-test.r


Although these images are quite complex they are rendered very quickly by REBOL.]]>steven@solie.caNone2010-11-07T11:02:04-07:00http://www.solie.ca/files/1b1ec011f9ff0063ec892c3af7662a72-9.html#unique-entry-id-9http://www.solie.ca/files/1b1ec011f9ff0063ec892c3af7662a72-9.html#unique-entry-id-9
REBOL (Relative Expression Based Object Language) is returning to the Amiga platform and I plan to be a part of it.

There is much already written about REBOL and what it is so I’m not going to repeat any of that here.

What is new is that REBOL is turning 3.0 and with that version change we are now able to collaborate a lot more with REBOL Technologies. Previous incarnations of REBOL kept much of the language close to home and REBOL Technologies was responsible for porting it to every platform along with all the services/components/etc. that go along with it. REBOL 3.0 opens things up a lot more and enables people like myself to participate in development. Only the core of REBOL remains closed source now.

Carl Sassenrath recently ported REBOL 3.0 to AmigaOS 4.1 and provided a working binary for the AmiWest 2010 show in Sacramento, California. The first feature I tried was chat. With chat you can interact with other REBOL users using a text based interface. The very next thing everybody wants to see is some kind of GUI. That is where I am getting involved.

REBOL 3.0 has something called a Host Kit where host-specific code is contained. Since AmigaOS has a unique GUI it requires a unique host kit to enable REBOL to interface with that GUI.

I now have windows opening and closing and reacting to events. Next will be adding more features like text rendering from REBOL and so on. I hope to blog my progress here so stay tuned.]]>steven@solie.caNone2010-07-28T20:57:56-06:00http://www.solie.ca/files/3f53db87b0514c21ede7eac901d2514d-8.html#unique-entry-id-8http://www.solie.ca/files/3f53db87b0514c21ede7eac901d2514d-8.html#unique-entry-id-8
I’ve attended a couple of seminars by the authors of Leading Lean Software Development. Mary and Tom Poppendieck did not shy away from the tough questions at the seminars which is a good sign. This book attempts to tackle some of those really tough questions and more and it does a good job at it.

There are many references to real corporations and situations which helps back up the claims the book is making regarding lean software and human behaviour in general. This is always a welcome addition for me because it is sometimes difficult to believe some of the claims because they are rather unintuitive and often against traditional western thinking.

The subtitle of this book is “Results Are Not the Point” which is prime example of going against traditional western thinking. The idea in this case is to focus on improving your software delivery system and stop focusing so much on the products. This will in turn help you produce better software more reliably because you won’t be just burning people out with overtime each time to get results. At least that is my interpretation.

Overall, I did find the book was a bit dry and perhaps a little too detailed at times. There is just so much information in there I suppose it was difficult for the authors to distill their knowledge down. The book is well organized but I think it also requires quite an investment to soak up all the information in there.]]>steven@solie.caNone2010-03-13T22:19:11-07:00http://www.solie.ca/files/ce2deca24ff983f631ae4d9b576c5baf-7.html#unique-entry-id-7http://www.solie.ca/files/ce2deca24ff983f631ae4d9b576c5baf-7.html#unique-entry-id-7
I like this book. It is obviously targeted directly to software professionals like myself. It also talks about that mushy brain thing. How could I not like it?

I never really meant to buy this book. I was actually trying to buy a copy of the Agricola board game. Since the store was out of stock when their computer claimed otherwise I received a nice discount on anything in the store. I walked by the computer book shelf and spied the wonderful title Pragmatic Thinking & Learning: Refactor Your Wetware. I knew the author Andy Hunt from Pragmatic Programmer fame. I also just have to buy anything with wetware in the title.

This is one of those books that is actually two books in one. On the one hand, it talks a lot about how the brain works and the latest theories. There are plenty of cool facts to chew on. On the other hand, the book tries to help you become a better software professional by fully utilizing your current brain. No expensive and inconvenient brain transplant surgery required.

There are plenty of other texts out there about how the brain works, why people behave as they do, how to change your learning habits, etc. What this book does is bring it all together and present it to the software professional in language we understand. There is even a handy reference card listing all 48 tips and references to where they are in the book.

You need the reference card because this book is only a starting point. You don’t magically become a better learner after reading it. You have to work at it and the reference card helps you remember the most important points.

What a nice surprise this book turned out to be. It is compact, very readable, Andy Hunt is hilarious, the book is inexpensive and it is loaded with great tips 100% relevant to what I’m doing right now. Can’t get much better than that.]]>steven@solie.caNone2010-03-04T17:38:39-07:00http://www.solie.ca/files/665b40716d465ccf5a053b4faaa575e1-6.html#unique-entry-id-6http://www.solie.ca/files/665b40716d465ccf5a053b4faaa575e1-6.html#unique-entry-id-6
The latest AmigaOS Software Development Kit (SDK) was recently released. The SDK corresponds to the AmigaOS 4.1 Update 1 system release. The SDK is available for download now and doesn’t require any sort of mandatory registration.

I had prepared the SDK before but this would be the first time new includes and autodocs were included. The includes are a collection of C source code header files. The autodocs are generated directly from the source code similar to JavaDocs, etc. However, the autodoc format has not really changed much in a couple of decades now and consists entirely of plain ASCII.

Many developers helped contribute to the SDK and I’d like to thank all of them. Once again, this truly was a team effort.]]>steven@solie.caNone2010-02-25T13:09:42-07:00http://www.solie.ca/files/f576c5e5a47f34c194d32b9ca107870a-5.html#unique-entry-id-5http://www.solie.ca/files/f576c5e5a47f34c194d32b9ca107870a-5.html#unique-entry-id-5

Lua Programming Gems is a collection of articles on Lua in the spirit of all the other “gems” books that came before it. This is simply a must have for all Lua programmers and touches on pretty much every aspect of Lua in some way.

I’m still relatively new to Lua myself so some of the samples are beyond my Lua skill level. Even so, I think the book will be a valuable reference later on when my skills have advanced.

The one important lesson I took away from this book is that Lua really is a language for building languages. Luas only data structure is the table from which you can build all your traditional structures like queues and lists but that doesn’t mean that you should. Instead, try to use Lua as it is and adapt to the way Lua does things instead of forcing Lua to be something else.]]>steven@solie.caNone2010-02-05T20:49:11-07:00http://www.solie.ca/files/2c82030bb241205596583e1677df8ef5-3.html#unique-entry-id-3http://www.solie.ca/files/2c82030bb241205596583e1677df8ef5-3.html#unique-entry-id-3
Clean Code: A Handbook of Agile Software Craftsmanship focuses on the importance of the source code itself. I’m already familiar with Robert C. Martin’s Agile Software Development: Principles, Patterns, and Practices so I expected something similar but focused on the code instead of on design. I was not at all disappointed.

Please don’t let “Uncle Bob”’s book titles scare you away. There is some focus on the agile methodologies but nothing is really being pushed on you. Agile enables more of a focus on the code and testing which is the main point.

Martin begins with an argument for writing clean code and caring about the readability of code. I did not require convincing myself but I’m pretty sure there are still a lot of people out there that still believe getting code working and walking away is still acceptable practice. What follows is the real meat of the book.

The book covers horizontal and vertical spacing, code comments, choosing meaningful names, handling third-party code, unit testing and everything else you would expect a book about coding would cover. Martin goes so far as to pick sides as well and does not hold back on expressing his opinions. I think we all respect an author that is willing to take sides.

The book then takes a turn into more unfamiliar territory. The systems level is discussed with ideas on design aspects, frameworks and even domain-specific languages. Emergent design, refactoring and the wonderful world of concurrency (ex. multi-threading) is also explored in some detail. These chapters were a surprise for me as I did not expect to find such a wealth of information in a single book.

Real examples on open source projects are presented along with all the advice. Martin takes us through how to take a badly written pile of bits and turn it into what he calls “clean code” step by step. There is a lot of great detail on how and why he chooses to do what he does but it does drag on for a while so be prepared to put some work into reading it.

The book concludes with a collection of “Smells and Heuristics” which help bring everything together.

The first section on concurrency was a little scant so it was decided to put a full appendix dedicated to concurrency at the back of the book. This is a valuable appendix with useful information for anybody dealing with concurrency in their code base.

There is a lot of good information in Clean Code but it does have the same flaw as all other books on coding have. The author must choose a programming language to make the book useful. This time it was Java 5 that was chosen so it would be a good idea to brush up on your Java to get the maximum amount of use out of this book.]]>steven@solie.caNone2010-01-15T14:31:16-07:00http://www.solie.ca/files/2fa3294ad52a498912dc821a4a27ad7f-2.html#unique-entry-id-2http://www.solie.ca/files/2fa3294ad52a498912dc821a4a27ad7f-2.html#unique-entry-id-2AmigaOS 4.1 Update 1 is finally out for all supported AmigaOS 4.1 platforms: AmigaOne, Sam440ep and Pegasos II. The Sam440ep platforms are still manufactured by ACube Systems out of Italy and still readily available. If you want to get into AmigaOS, I always recommend AmigaKit because they are simply the best dealer out there.

I was personally quite involved with this release of AmigaOS. I contributed many hours of my time (mostly at no cost) to this product and I’m pretty proud of it. The team behind this product is pretty large and very international with members all over the globe. Hyperion Entertainment controls the AmigaOS source code and allows a smaller subset of the entire team to work on the code base of which I am a part of. I enjoyed my time working on this release and I look forward to contributing again in the future.

A big thanks to the entire AmigaOS 4.1 Update 1 team. It was a blast!]]>steven@solie.caNone2010-01-04T15:43:25-07:00http://www.solie.ca/files/b3964c6b22bbbcd73f6c42a36ee63001-1.html#unique-entry-id-1http://www.solie.ca/files/b3964c6b22bbbcd73f6c42a36ee63001-1.html#unique-entry-id-1
Outliers: The Story of Success is a book by Malcolm Gladwell which I received as a Christmas present. I already read Gladwell’s The Tipping Point so I appreciated his writing style and focus on simplicity. The book is very approachable and written for a large adult audience. This helps with defusing the reader. This is important considering he is making the point that it is likely those people we admire that are considered geniuses, elite, etc. are likely incredibly lucky.

Gladwell has a knack for pointing out the obvious and putting a new spin on it. With outliers, the obvious point is that we rely on others for success. One of his many examples included Bill Gates who cofounded Microsoft. Many consider him some kind of genius but I know I’ve always remained skeptical. Gladwell pointed out something new to me in that Gates had access to one of the only time sharing computer systems at a time when very few others had that opportunity. Armed with 10,000 hours (a magic number that pops up repeatably in many contexts) of computing experience when the rest of the programmers in the world had maybe hundreds of hours sure gives you an edge. Add in a couple of other lucky breaks and we have a billionaire with incredible timing versus some kind of genius.

Outliers isn’t poking fun at the successful of the world or those that are smarter than the average bear. What Outliers is doing is pointing out that armed with some smarts, some great timing and a great cultural background and you have the recipe for success. Without all these things, the underdog really doesn’t have a fighting chance and it is very unlikely we will find many that are considered a “success” which have not had a lot of critical help along the way.

I think Gladwell’s point regarding culture being a larger influence on behaviour than most would admit is an interesting one. He asserts that cultural differences linger a lot longer and a lot deeper than might be obvious to many people and we should not dismiss culture so readily. This came as a surprise to me as well. I would never have thought the attitudes or our ancestors who immigrated to Canada would still have a large influence on my behaviour today but on reflection, he just might be on to something.

This book really does make you think which is always a good thing.]]>steven@solie.caNone2009-12-06T22:39:15-07:00http://www.solie.ca/files/0d6353798bf9b02159142589e5e89449-0.html#unique-entry-id-0http://www.solie.ca/files/0d6353798bf9b02159142589e5e89449-0.html#unique-entry-id-0

Implementations Patterns is a book by Kent Beck of Extreme Programming fame. These are patterns which are applied at the implementation level versus the usual design patterns. It has been described as three books in one and I tend to agree.

The first part of the book includes Chapter 3: A Theory of Programming which delves into the values and principles of programming itself. Professional programmers will nod in agreement with much of what Beck is saying because much of it has been implied for many years. It is really nice to see it written down and explained.

The patterns themselves don’t necessarily follow any particular formal template and they focus solely on the Java programming language. Given these are implementation patterns it does make sense Beck had to focus on a programming language and he chose Java. Being more of a C++ guy, I found a vast majority of the patterns (if not all) can still be applied in C++. Any object-oriented programming language can easily make use these patterns.

The final part of the book delves into frameworks and includes a handful of patterns and some solid advice for those implementing frameworks. Anyone that has designed and implemented a framework will find this advice very familiar. Again, it is really nice to see it written down and explained.

Kent Beck has a knack for articulating the complex world of software development and this book is no exception. I now understand why it is has been so highly recommended.]]>