% fortune -ae paul murphy

Things we need to learn more about

If you were asked to list the top five things you think are important for IT but which you don't understand and for which there's no obvious research base you can tap into, what would they be?

Here's my list:

  1. the impact of the differences among open source licenses on business decisions.

    Imagine that you're about to start a hardware or software oriented business and then ask: what characteristics of the business and its operating environment would favor each licensing approach -from purely proprietary to full GPL3?

  2. the extent to which, if at all, the use of suspect e-voting technologies has the effect of biasing elections.

    Imagine that you are a senior electoral management official -and your state or county has opted for brand X (and thus type Z) voting technology. Does the technology deployed in brand X affect your decisions? and, if so, is the effect sufficiently systemic to affect state wide or national election outcomes?

  3. the extent to which, if at all, technology can be used to limit or reverse the erosion of personal freedoms associated with current government drives to first identify everyone and then legitimise demands for proof of identity in many different situations.

    Imagine you're a senior analyst at a major American political think tank. Your executive board is concerned that demands for identity are eroding individual rights, but they're also aware that present technologies and methods require fool proof identification for the police and others to do their jobs. Now ask: can technology help find the right balance between these competing demands? and, if so, what would it take to make a real difference?

  4. the drivers for Microsoft's next operating system decision;

    Their network OS effort is, I'm told, in shambles; the objectified VMS foundations for the present series are tottering under the metaphorical weight of functional expectations and 70+ million lines of spaghetti code; games developers aren't meeting expectations for Xenon; and rumour has it that the IBM/Sony Cell consortium has finally gotten its memory compiler to work.

    Basically theirs is a rocks and hard places choice - continuing with the current code base while slopping more good money after bad on Erogoth (Xenon network OS), taking on IBM over Linux, or making Steve Jobs look prescient either by licensing MacOS X or by taking the hit on grabbing another BSD as the foundation for its next generation systems.

  5. the extent to which, if at all, today's communications technologies can be used to improve both formal and informal education.

    Computers are now used almost everywhere in education - from kindergarten and early years home schooling to research uses in Universities in the formal system and from teaching games to the web for the informal system. The question, therefore, is whether something can be done to improve the effectiveness of these tools at any or all levels: on what criteria is brand X more effective (i.e. a better choice) than brand Y and for whom?

    If the code for pre-kindergarten learning games like "Phonics made easy" is written on Windows by Windows programmers using Windows assumptions, is the learning experience better or worse if it's played on a Mac? If teachers expect Wintel PCs does eliminating the costs and hassles of Wintel support by switching them to Sun Rays pay off in teaching time and commitment? and, if so, under what circumstances?

Mysteries to me - but worth exploring, right?

Oh, and FYI: the "erogoth" reference above is a joke - just like the claim that you can get fired from Microsoft for calling it Xenix2.


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specializing in Unix and Unix-related management issues.