At the closing session of DevConnections there was a jam packed auditorium of attendees and a panel of a host of speakers up front. Carl Franklin ran around the room (literally ) with a mike so people could ask questions. A larger percent of the questions were SQL Server related as that is the focus of one of the three conferences that make up DevConnections. There were a few questions about .NET, such as the future of the application blocks (which, by the way, is two-fold: much of what was learned in the app blocks is being folded into VS 2005 and they are also going to be part of Enterprise Services).

Yet one consultant stood up and asked an interesting business question. While there is so much push from Microsoft and from all of us who are excited about the new technologies, he is in a very familiar position. His clients' VB6 programs run their businesses. They use SQL 2000. Everything works great. He said that his clients are afraid that he will push them to the new technologies which would be a great expense for them, even though their current systems more than suit their needs. What's a guy (or girl) to do? He was looking for advice in handling this issue.

I have a client that was very happy on their VB6 apps and SQL 7 for a long time. I have been very fortunate that they trust my choices and although I continue to make minor changes to their enterprise VB6 application, every new piece of development in the past 2 1/2 years has been in .NET. This did not cost them anything extra since I spent my own time learning and experimenting. There are even pieces of their application that I have not charged them for, such as implementing WSE2 in one of the production apps. My ability to make these choices has also paid off in a very visible way with the use of a smart client application with a web services back end which solved a problem we had talked three years ago about solving with a very expensive T1 line to one of their satellite offices.

I have not leveraged anything new in SQL 2000 so that they were not forced to spend the large chunk of cash to upgrade from SQL7. This summer, it was discovered that their server had been hacked and in a quick build of a new server, the sysadmin chose Windows 2003 server which meant we had to move to SQL 2000. The 6 month trial was a saving grace in the emergency, but I know the client did more than grimace when those 6 months were up. Now they are asking me about SQL 2005 and Longhorn so that they can start planning ahead for the expense of upgrading all of their systems. I promised them that I have no plans to implement any of the sql 2005 features any time soon in their applications and that if I do find that I have a very strong business case for doing so, there would be plenty of time to prepare for the financial hit.

So, back to the guy at the conference with his question  - how will Microsoft support HIM in working with his clients to make these decisions and to assure his client that he is not needlessly costing them money? He  is in a rough place. One of the biggest reasons I left FoxPro years ago was because I was sick of defending it to potential clients who thought that Microsofot was going to bury it. This developer wants to learn and be proficient with the new tools. He also wants to keep his clients happy and to be a good consultant to them by helping them make the right choices and protecting them from unnecessary expenditures.

It's an interesting question. I agree that there should be something that Microsoft can do to help him, but it has to be to help give *his* clients the best advice but not scaring them into moving everything to the latest and greatest technologies so that they can sell another server license.

Kathleen and I talked about this and I think it also fits somewhere in the conversation that Joel Spolsky began (or perhaps just raised to new heights) in his article about the API Wars. I expect to see more on this discussion in the near future.