But in these cases We still have judgment here; that we but teach Bloody instructions, which, being taught, return To plague the inventor: this even-handed justice Commends the ingredients of our poison'd chalice To our own lips. -Macbeth: Act I: Scene VII: Lines 7-11.
OK, it's not often that you see a quote from the ol'e bard in a piece aboot computers. However in this particular instance, it appears that Shakespeare hit the nail on the head. Yesterday Microsoft declared that it had released an 'embedded' version of NT 4.0 that it hoped would be used to run mission critical devices in hospitals, eventually being used in pacemakers and the like.
No, really! If you had visions of red lights flashing in emergency rooms across the world as thousands of embedded NT systems coughed up BSODs, then you're not alone. We need to send Microsoft a message: They can sell us half-baked products to run our homes, our offices and our factories but when it comes to the systems our very lives depend upon, they had better stay away - far, far away.
I'm not trying to bash Microsoft here. They do make some fine products indeed (with that many talented programmers working for them, how could they not?) like Excel, which is easily the best spreadsheet application ever written. But NT is not one of those fine products. In fact Microsoft's proficiency at writing operating systems is so far below adequate that it has given them a reputation for writing buggy software. The only reason they even bother to keep playing in the OS market is that they have a de facto monopoly, which makes the opportunity cost of not selling new OSes too high for them to pass up the opportunity.
Traditionally, Microsoft has managed to sell a phenomenal number of licences for it's OSes because they sell products that, at first glance, give the most bang for the buck. Typically, corporate types see the low sticker price of something like NT (compared to the big unices) and plunge in headfirst. Once they are firmly entrenched in a Microsoft environment, they figure that using Microsoft's other products will ensure smoothly running systems. When they run into problems with the OS, it is easier to create a temporary workaround (known as a kludge) and hope that the problem will go away in the next version.
This is true for just aboot every corporate environment running on a Microsoft OS. Over time so many of these kludges emerge that new employees are often stunned at the amount of quirks that the people at the company have been putting up with. When the next version of the OS is released, Microsoft makes a big deal aboot having fixed the problem (if they managed to fix it) and trumpet all the new features they have added. Companies upgrade their systems, hoping that everything will work properly now. They soon realize that, although the old problem has vanished, some of the new features (most of which they didn't want in the first place) have introduced new problems that need to be worked around so business can proceed as usual.
This flagrant disregard has worked for Microsoft because businesses are generally too stagnant to even think aboot leaving the Microsoft fold. Microsoft uses FUD (fear, uncertainty and doubt) to keep defectors in check. They go to great lengths to prorogate the notion that alternatives like Linux, MacOS or BeOS are largely 'untested' or 'unsupported' and therefore not viable options. The logic generally goes something along the lines of: MacOS is nice but nobody uses it anymore because there aren't any applications for it (this despite the fact that Microsoft sells Office, their flagship product, for the MacOS); Linux is too hard to set up and there are so many different versions that you will get confused (and yet they sell Win98, NT server, NT workstation, WinCE and now embedded NT); BeOS, what's that (although they are quite capable of acknowledging it's existence to the DOJ)? Business people tend to believe what Microsoft's army of independent support technicians (the MCSEs) tell them, unaware that if Microsoft looses their stranglehold on the OS market, these technicians would find themselves out of work (giving them a pretty good incentive to promote Microsoft products).
While this may cut it in the business world, the hospital is a slightly different customer. If your insurance company has a computer crash, it may take a few more hours for your claim to be proccessed but you can live with that. Now if the intravenal drip system - the one keeping your child alive in the hours after a car accident - decides to crash, the results may be far more painful. Last year a huge US military ship running Microsoft's NT was left dead in the water after NT crashed. Now how do you feel aboot letting hospitals run mission critical systems on an operating system whose principal claim to fame is for being unstable?
It's high time that ordinary people became concerned aboot the dangerous consequences of allowing Microsoft to push it's crash prone operating systems upon them. There are several embedded OSes that have over a decade's history of being used in mission critical devices: QNX, being perhaps the most technically suitable for medical equipment because of it's legendary stability. There is no need to use embedded NT, unless you enjoy living on the edge: NT tends to require a reboot every week or so. Yeah I can just see them putting it into pacemakers now...