> What a hobbiest license would do is make it possible for z/OS > and z/VM to survive the coming retirement of 80% of all > the experienced programmers. By creating a very inexpensive > training/development environment, IBM would make it possible > for that market to continue.
I don't believe a hobbyist license would have the slightest effect. The days when mainframe users recruited self-taught people are long gone - they look for experience in the type of environment they have, and any training they fund they expect to be professionally given. The "hobbyist license" proponents number at most a few dozen around the world, and their demographics are no different from the pending retirees you mention - there is no great mass of young would-be mainframe programmers held back solely by lack of access to the software. Maybe a few - but nothing like enough to make any difference. One piece of evidence to support this is that the "hobbyst license" lobby has never yet managed to organise itself enough to make a formal approach to the right interfaces at IBM. All I have ever seen is disorganised bleating on the fringes of user groups and on mailing lists and bulletin boards. So far, this has all been highly counter-productive, especially in the case of Hercules: a) Redbook chapter deleted b) AD/CD locked to Flex-ES c) "IBM has taken a business decision not to license its commercial software to this platform" now returned to every commercial license request Terrific result, huh? You should also take a long hard look at the high-end strategies espoused by IBM. There has only recently been a major reorganisation of the executives to support "on demand" computing - prior to that (and still current) the push towards self-managing autonomous systems. Both of these initiatives presuppose lower skill requirements at the customer end - so how are you going to sell a strategy whose purpose is the reverse? Deskilling of customers is to a large extent deliberate. IBM does not want its customers to say: "[some piece of middleware] is great, but we can't deploy it because we don't have the skills". You cannot provision this requirement in advance - you can't train scads of people and have them sitting around in case something like this turns up. New business requirements are created on much shorter timescales than traditional staffing and training-up can hope to cater for. Hence eLiza, then autonomous self-managing computing, and now "on-demand computing". All parts of a strategic development that hobbyist licenses have no relevance to whatsoever. > It would take a wild leap of imagination for IBM to make that move... > and I doubt they have it. They see a valid business strategy and are > investing in it. I don't know if you've ever met Bill Zeitler, who currently runs all of IBM's servers and storage systems, but the very, very last thing I would accuse him of is a lack of imagination. He's one of very few people I know who actually thinks while you're talking to him - it's very disconcerting to get a fully thought out and considered answer to a complex question the instant you stop speaking. Back when he was running the AS/400, I once saw him give a presentation at Rocester. Someone asked a question, and he answered it, but he obviously thought on. About half a minute later he froze for an almost interminable period, though the audience (of around fifty) later agreed on 45 seconds. All life signs ceased - I'm certain he stopped breathing. Then he came back to life, pulled out a notepad, made a furious note and continued with his presentation without comment. I've gone into the prerequisites for an application for a hobbyst license structure for z/OS (or z/VM) _many_ times. First of all - IBM earns one and a half times as much from software as it does from hardware - it's actually a software company. Gross margin percentages are in the eightieth percentile of $13 billion. It's a critical business. Into this process you wish to introduce a set of Ts&Cs that will certainly add no short-term revenue (IBM is quarterly-driven, remember) and has a potential of downside if you get it wrong. So the Ts&Cs have to go through the same legal review process as something like Workload Level Charges - the lawyers' time alone costs millions. Right at the start - who pays for that? Because of the critical nature of software revenues to IBM, there is no doubt whatsoever that these new Ts&Cs would have to be approved by the Management Committee. This organisation reacts in a veto-like structure to changes in the status quo - change is generally resisted unless there is a clear positive business case. Change with no obvious short-term benefit would be rejected instantly - even one voice raised against it would be enough. Given that the top IBMers are political animals (less so than in other organisations, but even so) who would you find to sponsor it? If, e.g., someone from zSeries - how would you deal with potential vetoes from Software Group? Given the issues IBM currently faces, how would you even get time for a revenue-neutral project? It goes on. And on. None of this has been thought through by those demanding hobbyst licenses. -- Phil Payne http://www.isham-research.com +44 7785 302 803 +49 173 6242039
