In what some less incisive souls are viewing as ammunition for the counter-attack, a recent Gartner 'first take' (the research outfit's free info-gobbets) warns that Munich city council's decision to migrate to Linux should not be seen as proving the case for Linux on the desktop. Which is no more than, as we say over here, the bleedin' obvious, but although the conclusion is sensible enough (in a bleedin' obvious sort of way, that is), the note also includes some claims that some people might interpret as suggesting Munich's decision was thoroughly perverse.
Gartner does not say outright that it thinks the Munich switch will turn out to be a costly failure, but it seems to question the move in terms both of cost and methodology. The migration, it says, will cost around €30 million, whereas an upgrade to Windows would have cost €27 million, excluding the extra discounts from Microsoft which Munich spurned. Alongside this, Gartner claims that "many applications will not migrate to Linux" but will be run either as thin client systems or "using virtual machine software, such as VMware."
Gartner then points out that too many VMware implementations "or other workarounds" would result in a higher residual reliance on Windows than planned. As it would - this too is obvious, but it seems to us that we have here a tiger largely of Gartner's manufacture.
For starters, there are as yet no lucky winners of the Munich Linux contract, therefore no definitive infrastructure architectures, and therefore no proven case from Munich for Linux on the desktop, or anywhere else for that matter. Munich has taken an in principle decision to switch to Linux, but although it seems likely that local player SuSE, and IBM, will get slices of the action, the contractors have not been formally decided on. After the appointments, the design and the rollout we (should it all work) might conclude we have an indication that Linux can play on the desktop, but proof? Not from one rollout, no matter how large, obviously.
Gartner misses the point rather more drastically when it comes to 'VMware and other workarounds.' Certainly you could count use of these as "residual reliance on Windows", but it's a curious way of looking at a setup that will have to be designed in order to handle existing applications currently in use by the customer. Would, one wonders, a 'rip and replace' make Gartner happier, and what does the outfit make of Microsoft legitimising (if this could ever be said of the company) virtual machine technology by buying some?
Although Munich's implementation hasn't yet been firmed up, it would not be in the slightest bit surprising if it used a combination of thin client, "other workarounds" (SuSE, for example, could be expected to offer Crossover Office to run some key Windows client apps) and virtual machine technology. The first of these three would require Windows app servers, the last higher spec hardware, but all three are perfectly reasonable mechanisms for deploying a new Linux infrastructure when you have numbers of existing Windows users to cater for. And it's a bit of a stretch to jump from this (although Gartner itself doesn't do this) to assuming that Munich will merely be deploying Linux boxes then running Windows wholesale on them under emulation.
And the tab? €30 million versus sub-&euro$27 million doesn't sound particularly clever, but we have known since the decision was made that Munich spurned Steve Ballmer himself, bearing an extra-special deal that undercut the Linux bid. The price tags offered to major customers that Microsoft wishes to stop defecting will frequently be lower than the alernative offers, but to presume that this must mean the outfits turning them down must be mad is again to miss the point. Government is well to the fore among these customers, and they are far more concerned about the long-term costs of lock-in than they are about short-term discounts.
Gartner claims that Munich currently uses "many older versions of Windows," observing that this class of migration is easier to cost-justify, but wasn't that a point we just saw galloping by, apparently unnoticed by the Gartner analysts? If you are currently using Windows 9x or even 3.1, then you must surely be thinking about upgrading. If you were to upgrade to WinXP, then it will be more costly and difficult for you should you later decide to migrate somewhere else. So a cheaper Windows deal now is only cheaper if you accept lock-in, and rule out migration for a period of years.
Aside from worries about the general hazards of lock-in, customers are concerned about the more specific ones evidenced by Microsoft's new licensing model. To many of them this provided a pretty clear indication of how hard Microsoft is prepared to squeeze when it knows you have no choice, so there's more than a little sense to taking steps to retain choice, the sooner the better, before doing so gets even more expensive.
All of that needs to be factored into the cost, and customers' increasing grasp of this will make it increasingly difficult for Microsoft to retain business via slush funds and the rapid deployment of senior executives.
How, though, do you escape? The significance of Munich is that it will provide useful data, whether it succeeds or not, on a number of possible escape routes, Linux on the desktop merely being one of them. But it shouldn't be viewed as an 'either or', with success meaning the triumph of Linux, and failure the continuing domination of Microsoft. Either way, the desire for an alternative will remain, and if the current best shot at an alternative doesn't work out, a better one, sooner or later, will be built. ®