Moving from Windows XP Pro to Ubuntu

I decided to take the plunge with Ubuntu on my old laptop that had Windows XP Pro installed. First I pulled down the Desktop CD of Ubuntu 6.06 ISO and burned it to a CD. Ubuntu fits on one CD-Rom disk, nice. Then after setting the BIOS settings to boot off the CD-Rom/DVD drive I inserted the CD with the Ubuntu 6.06 ISO.

After booting from the Ubuntu CD, you can test drive it to see if you want to install it on your hard disk. If so, there is a handy install icon. Select the icon, answer a few standard questions for language, timezone, userID, and password and it will install.

Then, I inserted my Linksys WPC11 wireless NIC and Ubuntu recognized it lickety split. I added my WEP settings info via the Administrator|Network tool and I was browsing.

I have to say that this was the most trouble-free install of any linux distribution I have seen. Then, with the help of the Urban Puddle blog, I was able to get Ruby, Rails, and MySql installed.

See my messing with Interactive Ruby (IRB) below on my new Ubuntu install.



Also of note was that Ubuntu was able to browse to my Windows server shares as Server Message Block (SMB) shares. I did not even have to mess with SAMBA. Sweet!!!

If you are hosting ASP.Net 2.0 on GoDaddy and using the MySql .Net Connector for MySql please read

First download the Connector/NET 1.0 installer from the MySql site (do not get the source code only download as it is buggy. The installer download contains the source also.)

After installing the MySql .Net Connector, in VS 2005 Add an Existing Project to your solution by browsing to C:\Program Files\MySQL\MySQL Connector Net 1.0\src\ MySql.Data.csproj file.

After adding the project, open the MySql project and add the following to the AssemblyInfo.cs file in the perspective locations:

using System.Security; //added for GoDaddy Host
[assembly: AllowPartiallyTrustedCallers()] //added for GoDaddy Host

Build the project and then either reference the MySql project or browse to the new MySql.Data.dll assembly directly.

Special thanks to a post from Alek at GoDaddy’s Support team on Microsoft’s Asp.Net Forum site. Apparently, as of June 22, 2006 GoDaddy.com, "has updated the custom medium trust configuration to allow the MySql.Data.Dll to work in a medium trust environment for the .Net 2.0 development environment." He then goes on to instruct you to set the AllowPartiallyTrustedCallers attribute in the AssemblyInfo file. Also thanks to others for pointing out the referencing the System.Security namespace in the AssemblyInfo file.

UPDATE: By request I have posted the compiled MySql.Data.DLL Assembly. Get it here.

The value of multiple input

I was waiting for the app I was working on to build as I was listening to an old Led Zepplin tune and started thinking, what has the great bassist/keyboard player John Paul Jones been up to? A quick Google on John Paul Jones and I found his personal site at johnpauljones.com. A quick look at the photos and I was surprised to see John in a picture with Nickel Creek band member Chris Thile, at a mandolin symposium (See the image below).



For the unacquainted, Nickel Creek is a progressive, blue grass group that many would not even label as blue grass. Any way I always read of the various blues and other world music that was the influence of Led Zepplin. I also know John Paul Jones played mandolin at least on Led’s fourth album song, Going to California. The point of this post is to be aware of the value of various influences. One should resist the temptation to limit oneself to only what is comfortable. I then thought of the literary influences that were listed by futurist Alvin Toffler in a recent C-Span interview. He spoke of a array of topics on which he regularly reads. Moreover, he explicitly stated that one needs to expand the genre and subject matter of what one reads to get a better, holistic view.

How does this apply to technology and business in general and software development in particular? One general application is that inputs of various topics helps one to better anticipate the ebb and flow of the information technology market and adjust one’s direction accordingly. Secondly, and more specifically, a broader understanding of things makes one more able to understand the various business domains for which one will have to write software solutions. Finally, the understanding of various programming languages enables one to develop a broader knowledge base by applying the concepts, idioms, and problem-solving patterns of one language to another, potentially combining the strengths of two or more languages to a solution.

Learning the Ruby Array via Specifications with RSpec

Now that I am having fun messing with executable specs let me review how RSpec helped me to better understand and use the Ruby Array.

See the code and the outcome of running the executable spec.

RSpec and the Empty Executable Spec

One great side effect of utilizing the agile approach to software design is the first thinking about, at lease as well as one can given the available information, the desired end result. This is typically started a failing test in the Test Drive Development (TDD) camp. However, the more I am exposed on RSpec the more I like the idea of design that is driven by specifications that are fulfilled in short, iterative cycles. Therefore, instead of having a "failing test" one would have an "empty executable specification." (If you could not already tell, I am not attempting to coin a new phrase. The expression "empty executable specification" is neither clever nor catchy. To be honest, I really do not know what else to call it.) In fact, it is the use of existing terms in TDD that do not really communicate the great results of TDD.

Click here for more.

Hegel's (or rather Jim Weirich's) Ruby Dialectic: the synthesis of Design by Contract and Behavior Driven Development

At a great Extreme Programmers User’s Group meeting here in Cincinnati, Ohio, Jim Weirich gave a thought provoking presentation on the experimentation of utilizing the semantics of a Behavior Driven Design (BDD) and the more traditional Design by Contact (DbC) with the Ruby rSpec tool. Having come out of the VB/ASP/COM/.Net world and now working on the J2EE team at US Bank, I appreciate the semantic qualities that various languages possess and more importantly the "idioms" they teach. I agree with the classic Sapir-Whorf hypothesis that the language(s) you use can either aid or hinder your ability to form thoughts and even solve problems.

With that said, the potential of Jim’s experiment with BDD and DbC is rich in its bringing together two divergent methodologies into a potentially better one. The question that the experiment is attempting to answer, from my understanding of the presentation, is can the thesis of DbC and its emphasis on getting the specifications completed upfront, pre and post conditions, and assertions combined with the antithesis of BDD’s executable specifications (a.k.a unit tests in Test Driven Devlopment) and short iterations produce synthesis of the two--a clearly understood contract (or set of contracts) between customer and developer that is the requirement for a short, iterative BDD’s executable specification(s)? Moreover, are there any unanticipated results that will emerge from the experiment that can be immediately useful or spawn more questions? One might state this is similar to an "alchemist experiment to see what comes to the top of the mixture of the two elements."

In summary, after showing us his Ruby framework to support Design by Contract, Jim explicitly stated that the experiment’s result(s) are not intended to replace Test Driven Development, BDD or DbC. However, in my view, it could produce a client initiated contract that can be solved in a few iterations with BDD. Or, sticking with the previous alchemist analogy, Jim's experiment could provide some proverbial "gold from silver and copper."

Testing .Net Properties with FIT

Just got a comment on .Net and FIT from a blog reader. This caused me to remember that I submitted a proposed patch to source forge, artifact 1255429, which deals with using properties instead of member variables. The patch was submitted to source forge in August of 2005. Nothing as of this post has been done, that I am aware of, so I thought I would put it out for public consumption and comment.

In more detail, fixtures deriving from RowFixture have instantiated classes with public variables, instead of exposing the instantiated object’s members via public properties. With these modifications you can write fixtures directly against objects under test, which are typically the actual application classes, exposing properties instead of variables to the RowFixtures. This enables the .Net developer to avoid writing additional code beyond the fixture and the actual application objects under test.

Here it is.