About five months ago, I had a difficult decision to make. My main development PC had finally died earlier in the year - something about an "Unmountable Boot Volume." My data files and source code were well backed up, but I had lost my unique collection of Visual Basic 6 controls and development environment, and I could no longer make changes to SurvivalWare the software. I could not create a new install file if only a few files needed updating. This was a six year old Windows XP, whose companion laptop had died a few years earlier. Many of the companies I had licensed 3rd party software from - for things like charts, zipping, and ftp functionality - had gone out of business.
As a software developer and entrepreneur, I find it embarrassing to write these words. How could I let myself be so exposed?
I had two choices:
Option #1 - I could spend time and money to recreate the old development PC, and continue developing under VB6, or
Option #2 - I could migrate SurvivalWare to the .NET development environment and start fresh.
Actually, there was a third choice. I could throw in the towel.
The fact that I'm blogging about it tells you my decision was probably NOT option #3.
My brother, Hank Luhring, thought I should take a serious look at option #1 - after all, how hard could it be to track down old disks and create a new XP computer? That would buy time to pursue the migration at a more leisurely pace. It would be the conservative choice.
But, maybe this was a blessing in disguise. The migration was already years overdue. I had been telling people it would cost $500,000 or more - not really knowing how much work was involved. I had made several attempts at at "Requirements Document" and getting design thoughts down on paper. I even talked to a couple of development firms. Things kind of stagnated when I could never find the time to get them a document detailing exactly what the software had to do.
I am a firm believer in what I'll call the "Nike Principle" for software project estimation: " Just do it!" (The software project). Note how long it took to do. That is the only reliable way to do a schedule estimate.
Hank and I engaged in a little contest. I turned over to him all my old disks and everything I had to help him load up a Windows XP computer with a Visual Basic 6 Development environment capable of continuing SurvivalWare development.
I would upgrade to the latest Visual Studio and write a "Hello World" program in VB.Net and maybe C#. And try to figure out what needed to be done and how long it would take. What were the key technology decisions? Could parts of it be outsourced? After 30 days, I'd decide which way to go.
SurvivalWare, the company, has the feeling of a startup once again. I am lucky enough to have my son, Nick Luhring, on board to help. Nothing like youth to push things forward, and keep one humble. He's been pushing me to start this blog for weeks. Thank you Nick - for your role in the birth of this blog, and the rebirth of SurvivalWare.
Next week: Brother vs. Brother
As a software developer and entrepreneur, I find it embarrassing to write these words. How could I let myself be so exposed?
I had two choices:
Option #1 - I could spend time and money to recreate the old development PC, and continue developing under VB6, or
Option #2 - I could migrate SurvivalWare to the .NET development environment and start fresh.
Actually, there was a third choice. I could throw in the towel.
The fact that I'm blogging about it tells you my decision was probably NOT option #3.
My brother, Hank Luhring, thought I should take a serious look at option #1 - after all, how hard could it be to track down old disks and create a new XP computer? That would buy time to pursue the migration at a more leisurely pace. It would be the conservative choice.
But, maybe this was a blessing in disguise. The migration was already years overdue. I had been telling people it would cost $500,000 or more - not really knowing how much work was involved. I had made several attempts at at "Requirements Document" and getting design thoughts down on paper. I even talked to a couple of development firms. Things kind of stagnated when I could never find the time to get them a document detailing exactly what the software had to do.
I am a firm believer in what I'll call the "Nike Principle" for software project estimation: " Just do it!" (The software project). Note how long it took to do. That is the only reliable way to do a schedule estimate.
Hank and I engaged in a little contest. I turned over to him all my old disks and everything I had to help him load up a Windows XP computer with a Visual Basic 6 Development environment capable of continuing SurvivalWare development.
I would upgrade to the latest Visual Studio and write a "Hello World" program in VB.Net and maybe C#. And try to figure out what needed to be done and how long it would take. What were the key technology decisions? Could parts of it be outsourced? After 30 days, I'd decide which way to go.
SurvivalWare, the company, has the feeling of a startup once again. I am lucky enough to have my son, Nick Luhring, on board to help. Nothing like youth to push things forward, and keep one humble. He's been pushing me to start this blog for weeks. Thank you Nick - for your role in the birth of this blog, and the rebirth of SurvivalWare.
Next week: Brother vs. Brother
RSS Feed
