This turned out to be the most problematic of all my statements in the interview. A buyer, whether of large-scale vendor-supplied applications, or shrink-wrapped software, will lose all control over their data if the format in which it is stored is proprietary. Many licence restrictions exist to prevent end-users extracting their data from the application they thought would help them.
The key point is that the data belongs to the users - not the vendor. Vendors, however, want you to continue to pay, again and again, for the 'service' provided by their software. The recent chat about 'software-as-a-service' (SaaS) is led entirely by vendors, looking to gull otherwise sensible people with fear, uncertainty and doubt.
The growth and undoubted success of open-sourced software is seen by these people as an aberration, suitable only for niche applications, such as web servers. But even in these niches, fear is spread that the applications are unsupported, amply demonstrated by the lack of corporate contracts to provide support. This is totally wrong, of course. The main method hitherto of getting paid in the open source space is to hire out your expertise to others who need it; think Red Hat and Linux.
Even if your data is locked inside a proprietary database, which your licence forbids you to reverse engineer, perhaps there are tools to extract the data? Maybe programs exist to export your data from inside the box to another format, perhaps .csv? Don't be fooled. The reason these export tools are so weak is to make it hard for you to get your data out while appearing to be 'open'. Your data should wash seamlessly into and out of applications as and when you call upon an application to help you with your problem, right now. One-stop, full service solutions are not the answer either - what if you want to re-engineer your processes? Or decide to do Chinese rather than Indian after the regular early afternoon card game?
The main benefit of open sourced production, and there are many, is that you regain control of your data. Nobody who sells you software wants to make this easy for you.
Simple, I thought. No elaboration into information asymmetries or game theory. Just common sense. New tools arise and are discarded all the time. A few years ago I used Yahoo for all my searches, and then Google came along and swept the board. The Microsoft vendor model is already dying. And this last proved to be my undoing.
"You mean we'll all be using Linux on our desktops, and not Windows?" she said. We might, I replied; the future is not set, but you can see the trends today, and there is no reason why Linux shouldn't become mainstream.
She couldn't see it, and hence couldn't see a future for me in her company. Neither could I, thankfully.
