John Olson

Recently I was at a Sybase event and an IT supervisor/PB developer approached me and asked why the new features in PowerBuilder were almost entirely for Web and n-tier and only a few were for client/server development. First, tools that are purely for client/server development are dead and dying, though client/server development is alive and well. They are dead and dying because they have matured and met the needs of their customers. PB and VB were great client/server tools five years ago, and it was becoming a challenge to identify new client/server features that would compel users to upgrade. Many of the other tools quit there, but PB and VB began evolving into client/server and Web/n-tier tools. Both of these tools are still alive and well. Just as PB client/server developers are scratching their heads at all the whiz-bang new Web development features in PB, so are the VB client/server developers.

Back to the question though. Maybe I'm out of touch with reality, but I don't think there are many strictly client/server projects anymore. Most new development is either Web development, n-tier, or hybrid client/server with a high level of integration with other systems and external components. The world of software is opening up and integration and interoperability are major goals in most new development. After years of improving efficiencies within systems and through business-process reengineering, many companies have already fully exploited those areas and are looking for new ways to get an advantage over their competition. Now they are focusing on integration and interoperability, both internal and external. Web services is on the leading edge of this effort, but you don't need to go that far in order to make gains in those areas. PBNI, XMLDW, IIOP calls to external systems... all can be used to broaden the capabilities of client/server applications and provide two-way communication of data between systems.

Sure, each new version of PowerBuilder is a large leap in the direction of new technologies, but many of those same features should be seen as a boon to client/server developers. Rather than the problem being that PB is leaving client/server behind, the problem may be how to educate developers in the many ways PB can help them in the type of development they are doing. That form of education is more up to us, PB developers, than to the Sybase engineers who work on concepts rather than real-world applications. They spend significant effort exploring new technologies and determining what features to give us. But it is the PB developers who push those features to their limits and invent new techniques. It is up to us to spread the news by writing about these discoveries in our magazines and books, speaking about them at conferences, and sharing them with the masses on the Sybase newsgroups and on CodeXchange.

If you have figured out ways to do things you haven't seen before, then share it with us by writing about it and uploading examples. We often don't consider our accomplishments to be noteworthy, but sometimes a seemingly insignificant solution to one person may be very significant to another.

If you know you need solutions or don't keep up with the latest information and therefore may be unaware of solutions available to you, I highly recommend you spend a little time each day going through the CodeXchange contents, Sybase newsgroup postings, and reading the latest magazine issues and books. You may find the solution to a problem you don't even know you have.

Everyone works better when everyone works together.

Resources