Time Marches on...

May Day, May Day... It is the first of May and I along with the rest of the nation are looking for better days to come.
The last few weeks I have been working on Java code for a good friend of mine Bob Arasmith. Bob is a great guy and one of the best engineers I have ever worked with / for. Our friendship goes back more than 25 years (yea I know, we don't look that old) when we first met at Verbatim in 1983. But back to the subject of working with Java.
The first thing I that hit me is that although I have been around Java from a QA perspective since the early 1990s I really had not developed a serious application with it. Java is pretty straight forward I guess but after working with Ruby and Rails for the past 3 years Java feels a bit heavy. My first project was to parse some very large text files and create a database from the contents. This was definitely not rocket science but it did give me a good task to familiarize myself with Java. I am now working on a second project that will incorporate JSW or perhaps YAJSW for monitoring a collection of applications and services. Wish me luck.

By the way, I am still looking for consulting, contracting or a permanent position, any leads would be most welcome.

I made the leap...

I finally made the leap and bought a new 17" MacBook Pro, let me just say that I believe this is the best personal computer I have ever had. My old PC workstation will undergo the transition to virtualized desktop using VMWare ESXi in the coming weeks. Once I have confidence that I can configure and maintain a virtualized environment I will move on to the server, should be a fun but challenging conversion as I am not an expert systems administrator.

I have several projects going right now, however I am on the lookout for a new contract or permanent job should the right opportunity present itself; any leads would be appreciated.

Checkout
Danny Faught's new blog Software Alchemy
Harsha Elchuri's new site TestForge.net


Until next time...

Cucuber and Selenium

In my latest contract I have been developing automated tests for a Rails application using Selenium and Cucumber. I was reasonably comfortable with Selenium, or so I thought but Cucumber was another story, no pun intended.

Cucumber allows you to define tests in natural language such as:
Given I am a valid user
When I log in
Then I should see my todo list


Pretty cool but the real beauty is in creating a DSL that lets you spit out tests nearly as fast as you can think of them. This is where the true benefit of Cucumber + Selenium really shines. Yes it takes significant time to build up the DSL but over time, you have a vocabulary that allows you to not only test the application but in true BDD form, specify it as well.

More to come over the next few weeks.

Is Development finally getting SaaSy

I spent a few minutes the other day speaking with Scott Price and getting to know a little about his company LoadStorm and their impending Beta. Cloud computing and SaaS seems to be "the next thing" lately, everyone (with the notable exception of Larry ) is proclaiming it to be the next big thing and for the most part I agree.

I think back two years ago when I started my journey as an independent contractor, I thought then there was a definite market for SaaS based tools. New companies were emerging that were more "hosted" or "ASP" than SaaS but Salesforce.com opened the door and with Amazon EC2 and S3 SMEs can afford to go green. IMHO, virtualization and multitenancy are going to have a significant impact not only on the way software is distributed, but how it is developed as well.

LoadStorm is not trying to compete with the big boys, at least not yet, but as more similar products mature and become availble, the big boys may find their market share eroding. LoadStorm, I think falls somewhere between the DYI with Open Source and the Enterprise that want and can afford the top of the line offerings.

I am going to kick the tires on LoadStorm's product and see if I can find a lightning strike.

Cloud Computing

Yesterday I heard about a new company called Zuora.com from a recruiter so I looked them up. Interesting company that is all about SaaS; I guess SaaS is a bit passe these days, everyone seems to prefer discussing "cloud computing" but no matter what you call it, it is on-demand, it is efficient, it is green and I believe it is the future of a large percentge of computer applications.
I think we will see in the near future, a large shift to cloud computing. An excerpt rom K. V. Rao, President and one of Zuora's Founders that I think sums it up very well.

What you build, you should sell.

Our customers are realizing they are diverting precious resources into critical, but non-core activities like subscriber management, billing, and payments instead of dedicating them to building and enhancing differentiation in core products. They fall behind in product differentiation and competitive advantage leading to loss of market share and revenue when operational laundry lists and details are usurping time that could be spent on revenue-generating activities.

What you can't sell, you should buy (or preferably subscribe).

This becomes even more important in tough economic times, when businesses get lean and mean, regain focus on building differentiation in the core business, and divest non-core activities or outsource them.

Addicted to Visine (use rcov to get the red out!)

So for years, being the QA guy that I am, I constantly was questioning if developers were bothering to unit test or complaining that their unit tests were not robust enough. What would I do if the roles were reversed? Well, they are not reversed as there is no QA team checking my application but still I was / am wearing the developer hat at times these days.
Big confession time.... it is harder than it looks to get into a test as you go rhythm. Test Driven Development is still not something I can claim. Not what you expect from a QA-Test guy huh? Well there you have it, I have taken the first steps on the road to recovery. I am a recovering non-testing developer. Actually I am well past that in the 12 steps, you see I develop in Rails and by habitually runing rcove I am developing a new addiction to green bars. I find myself getting anxious when I see any red these days; even when I can rationalize why I shouldn't worry. So now I develop a little application code, check the coverage then grap the Visine by coding the tests required to “get the red out”. Next step, rspec and TDD...

But what does that tell me about my years of QA experience? Was I just being a hipocrite? Perhaps a little but back when I was developing full time professionally, in C and assembler, the tools were not nearly as sophisticated. What I am really learning is that good practices, need to be habits and how do we get habits? By practice. So develop your addiction to visine and “get the red out”. It is a great “habit” for developers.