“First, explore how to doubt everything you think you know, and remember that all your ideas, even the most successful, are hypotheses within you – and not set in stone. Challenge the boxes that determine how you perceive the world, and think creatively about how you’re defining the specific problems you’re hoping to solve.”
Luc De Brabandere and Alan Iny in their book, “Thinking in new boxes.”
When is it time to move to the next box? Henry Ford was so enamoured with the Model T that he couldn’t move on to develop new vehicles. This allowed other manufacturers to take market share and Ford was nearly bankrupt by World War II.
A fear for anyone creating OSS products is that they are the owners of an old model in a marketplace that has moved on. Or a technology platform that has long since been superseded.
Sure, OSS the world over are being constantly enhanced by thousands of developers. But at some point it is inevitable that you can no longer enhance but need to overhaul. How do you throw away millions of lines of code, man years of effort or all the customisations you’ve developed for specific purposes? And then consider all the time required to replace it.
Is the marketplace really ready to move on? Does the sales data only tell you the market has shifted when it’s already too late due to the lead-time required to create a new product? Sales occur so rarely in our industry that it might even be hard to tell whether the market has moved on or whether the currently active buyers just aren’t your target customers (although I’m sure your sales team will already be atuned to it).
The key once again is complexity. The larger and more complex your products, the harder the decision becomes to replace rather than refresh.
Alternatively, the smaller, more modular, less complex solution, the easier it becomes to overhaul one product or module at a time (assuming you’ve also designed flexible interfacing capabilities between modules), This also means hard-coding less cross referencing within the product and providing it through flexible data query / report engines.
This modularity is another benefit of the Christmas tree approach by the way.
There is a major shift already underway in the OSS industry. Are your products well suited to replacement or massive regeneration to avoid disruption?