Note: I originally wrote this commentary in another venue, and have edited it to be more appropriate to the blog audience.

When it comes to the vision for Notes and Domino Designer, I strongly believe there is a huge hurdle to consider that stems from the very nature of IBM as an organization.

Directions get taken based on talking with large IT shop management, but also very much on trends and goals within IBM which is itself one of the biggest customers of the products in the world. It is not, however, a good representation of the rest of the customer base. Hardware costs, licensing costs, and diversity of skills costs are very much skewed at IBM when compared with the rest of the world. There are very few companies out there who can just throw J2EE server management skills, or Domino management skills, or AD management skills, or programmers of virtually any language at problems nearly at will. There are very few companies out there who can allocate new 20,000 dollar servers at will. If compared as a country, IBM has one of the largest economies in the world. What drives the things IBM wants to do are very different from what drives a lot of the rest of the world.

IBM is so big, that to some extent Notes & Domino end up being the result of an internal corporate IT development project that gets shared to the rest of the world. If you don't believe that, consider that there is code in shipping beta (and possibly production versions) that is specifically related to internal IBM end user tools and would never be executed outside IBM.

To validate their ideas, IBM presents them to their big customers. These customers that IBM presents their vision to see a very clean and crisp representation of it. A very senior IT executive sits in a room and is presented an idealized picture of what the next generation vision from IBM will look like. He agrees that he wants more control over the client side desktop, that he wants the kind of integrated wiring represented by 'dashboards' and 'portals' (speaking generically, not about Portal itself). Its like taking a poll of the population and asking "Do you want clean air and water?" Of course. Everyone wants clean air and water. How many people are capable, when asked that question, of fully evaluating the different ways to get there and what those implementation details will mean on the ground?

Lawyers will tell you that in the process of a "Grand Jury", any decent prosecutor can get an indictment for a ham sandwich. That doesn't mean the sandwich is guilty of murder or that the case will be won in the open battle ground of a court. The IBM executive briefing process is very analogous to this. Given well crafted presentations, shiny slideware, and promises of integrated portals, secure desktop management, and cool looking clocks and things; its fairly easy to get executive buy in for the new product suite. When it comes time to capture the market share in a broader sense and the real work and costs of doing things come in, its much harder to win in court.

So you start with a vision and direction strategy that is based very much on a viewpoint that is radically divorced from many of the customers, then very skilled sales and presentation people get the customer buy in through these customer briefings, and you end up with a self-validating set of design plans.

IBM Lotus has had a "Design Partner" program for several years now. That it exists is no secret. I believe this has been a serious (and very welcome) move toward balancing the internally reinforcing nature of the design and architecture process. As painful as it is, I honestly think that the barrage of harsh criticism IBM gets from some of us has in fact helped the product. But not enough. As DP's we feel often that we're brought in too late to make the really important changes. I know I have a pretty strong vision for what a Notes front end needs to be in a next generation world. I could drive a couple of hours to Westford and present my thoughts -- and I'm sure I'd be given the courtesy of an audience, but I have no doubt that it would be met with at best blank stares, and at worst outright derision once I'd left.

There are some of us in the DP program that have a pretty realistic understanding of the real forces that push success and failure at the client and designer level. I don't mean just on a bug-fix or UI layout sense, but on a real architectural level. While we've had some input lately, it seems to often come too late for the kind of changes we're asking for to be implemented.

How can IBM do a better job at getting more of the kind of real core level feedback in time to make the right decisions?