Microsoft: MED-V won't help you escape WinXP end-of-life Slipping Windows XP inside Microsoft's Enterprise Desktop Virtualization (MED-V) to get around the PC operating system's end-of-life date won't work. That's the warning from Microsoft, which has advised customers to stop looking for ways to keep Windows XP going and finally migrate. Extended support for Windows XP finishes on …
If by "maintained" you mean newer versions are being released, then Windows fits that definition. If you mean that old, unsupported versions are no longer being maintained, then I'm glad you are not in charge so that source code for the old, unsupported versions of my company's software aren't released to my competitors, etc since current versions are based on that code as well. I suspect many, if not most, developers of commercial software would feel the same.
I've got win 3.11, 95 and 98 running for some mission critical (ie too expensive to replace while they are still working) systems.
Since they've been stable for so many years, and I've avoided putting any other software on them, and they are not connected to anything else outside their own little circle they should be happy for a few years. Or at least so long as my cannibalised parts bin from their departed siblings holds out.
To some extent I can apply the same processes to Win XP.
"I've got win 3.11, 95 and 98 running for some mission critical (ie too expensive to replace while they are still working) systems."
That's not what "mission critical" means. And if you've got truly mission critical stuff running Win 3.11 on bare metal then you'll be in for a treat at some point in the future. Consider at least virtualising it?
"ome mission critical (ie too expensive to replace while they are still working) systems"
Err , what? If they were mission critical my friend you'd have replaced them years ago since the hardware your win 3.11 must be on its last legs. If your business goes under because you couldn't be arsed to update some PC software in 20 years then serves you right.
It's not long since I managed to replace a DOS (6.20) system. It was not the same box as it was when the software was new but it was the same software.
(In fact it was probably much faster and easier to back up - bootable USB, copy folder on and restart PC. just copy the files onto the network where my predecessor would have used about 3 sets of 45 floppy disks and PKZIP.)
The new system does pretty much the same job through a web browser. More importantly, management can run off lots of nice reports whenever the urge takes them.
>Err , what? If they were mission critical my friend you'd have replaced them years ago since the hardware your >win 3.11 must be on its last legs. If your business goes under because you couldn't
You've obviously never worked in a small manufacturing company.. where I used to work we had a number of old Win 3.11 machines (386's no less). The reason? We had a custom-designed card and software to run the machine That Did Everything and said card refused to work in anything faster than a 386 and the software was DOS-only (and no - we didn't have the source because it was written in 1994 by a long-departed contractor).
Cost to replace? A lot - both in terms of time and lost production on the machine. Did I mention that we only had 2 cards (1 didn't work properly and was being used as a spart-donor for the other)..
Unfortunately WinXP machines are much likely to be networked, and thus exposed (however indirectly) to the wider internet and all the dangers that that presents. If your systems aren't connected to anything else at all, then you're a lot safer.
(Yes there's still the old Sneakernet danger, but it's a lot less serious than having some script kiddie prodding your ports until you bend over.)
"If it ain't broke don't fix it".
IE, the anon could be posting about a 200k CNC machine or something. You load in the design files (whatever format) and it spits out a part. Does it matter if it runs linux, CE, Chrome, Win, DOS, JAVA? To me, it does not. If it turned out to have Win 3.11 installed, then fine, why change it? Why risk a fatal exception causing the tool to mangle your expensive materials or the entire machine?
However, the point is that the machine does not need any operating system support anyhow. It's a closed system. :P
> IE, the anon could be posting about a 200k CNC machine or something.
The CnC machines I work with are usually Win 95 / 98 machines and connect to our software via RS232.
Often at 9,600 baud.
It's fairly surprising when some of these things cost a quarter of a million; but they are pretty much bomb proof. So long as they keep to the standard comms protocol it's really none of my business what goes on inside anyway.
And besides
Our CNCs running windows <spit> have big fat disclaimers on them stating "UNDER NO CIRCUMSTANCES WILL XXXXX MACHINE TOOL COMPANY BE RESPONSIBLE FOR DAMAGE/DOWNTIME IF YOU UPGRADE THE WINDOWS OS IN THIS MACHINE"
But then only 3 run windows anyway, the rest are on Linux or Custom Control OS
I feel your pain, brother. We have one customer with multihead embroidery machines that have dedicated control software running only in Windows 3.x. Believe me, we've tried to make it run on modern hardware and OSs (closest we got was using WINE, but it still fell over to often).
To all those who've never played in this space, the realities are that it costs tens to hundreds of thousands to replace a machine and they are built to run for decades. People running these systems for business have the attitude that it's worked fine up until now, we'll move on when the machine dies. You deal with it or they find someone else who will.
If MS was my company, I'd be finding out why the customers want to stay with it, perhaps it has features that the later Vista and Win7 don't? I'd be asking "Why the hell are we forcing people to take something they don't want!?" Make the upgrade path less painful, especially in terms of cost.
If MS see no money in Win XP, then charge for updates, charge for sevice packs.
I still use XP in Parallels for my CAD system on my Mac. Upgrade to Win7? spend £100+ for what? Don't think so. XP works just fine. Mind you I downloaded the preview of Win8, what a crock that is! Got so pee'd off it with it so fast.
Microsoft: Giving people what they either don't want or need and charging the earth for it.
Yes, but the "killer feature" that XP has which Fister and 7 lack is, er, IE6. MS want that POS dead, buried and its grave thoroughly danced on as much as the rest of us do.
At the end of the day, the real issue is "it works, so why would we go to the effort and expense of changing it?". There's not much you can do in a subsequent release to counter that. Even giving it away free and having it make the tea as well won't help.
In the real world, the cost of the product itself is peanuts when compared to the costs of application migration / certification / testing and the subsequent rollout.
Incidently, MS's view of corporate usage may well be correct, from their perspective. They'll be looking at the licenses. It's common knowledge that vast 7 desktop migration projects are ongoing around the world and it's quite likely that the majority of those are close to fruition and have already aquired the licenses for rollout.....
"It amuses me so much that MS's proprietary lock-in on IE6 is now biting them hard on the ass."
I wonder if at times it amuses them too? I mean, never could they have dreamed that they would have done such a thoroughly comprehensive and all encompassing job of it. In a way it has to be the most successful piece of code Microsoft has ever written.
"If MS was my company, I'd be finding out why the customers want to stay with it, perhaps it has features that the later Vista and Win7 don't? I'd be asking "Why the hell are we forcing people to take something they don't want!?" Make the upgrade path less painful, especially in terms of cost."
its well known why businesses have stuck to winXP. as stated in the actual article...
"The sticking points are Internet Explorer 6 and 7, as many big businesses have written critical apps for IE6 and 7 – which are off-limits on Windows 7."
The cost to re-write those apps so they work with alternative browsers that work on windows 7 is going to be MASSIVE... add that to the cost of upgrading all the OS, and on top of that, then the cost of upgrading hardware... your looking at a costs of several hundred thousand pounds for a large corp.
Personally, once it was realised that my software I use for my business wouldn’t be able to run on windows 7 the code was re-written and made sure that it was browser independent and therefore OS independent. It wasn’t really a big deal for me as my business is only a very small enterprise and for the majority of the time its only me or the missus using it so it does not have to be very polished !!
other businesses should have taken similar steps starting way back 7 years ago...
"other businesses should have taken similar steps starting way back 7 years ago..."
Yep, done that.
As a former MS sysadmin I know what a crock of shit Windows is, so when I started my business I made sure that I stayed well away from the MS treadmill. I only use Gnu/Linux and it serves me well. A few tussles with suppliers and customers who would send me .docx documents but a few polite requests sorted that out. Other than that I had a few problems with getting to grips with Gimp which is a story in itself. As a bonus I sidestepped all the viruses and malware problems.
Not that this is a solution for everyone trapped on the MS upgrade cycle but with a bit of planning and sadly at some cost it can be done. You just need to think through whether it makes sense for your particular business.
"If MS was my company, I'd be finding out why the customers want to stay with it, perhaps it has features that the later Vista and Win7 don't?"
We know why. The article even says why. Because some of these companies have mission critical software that only works in IE6 and they either can't or wont spend the money to re-write that software until the point that they absolutely are forced to. It's not because, as you imply, the management of these companies have strong feelings about how much more lovely the interface is in XP over Win7 or that they feel they really can't risk moving to an OS with a better security model. It's replacement of archaic software.
Very good point.
A lot of posts/media say it's old software (e.g. non-compatibility with IE7+) holding migrations back, but I'm really wondering whether a big factor could be that the new OS just does not have much more to offer in an enterprise setting.
Of course, driver support for new machines is one of the ways in which Microsoft gets people to upgrade regardless of whether they can afford it/see any net benefit.
Given the latter, my earlier points are probably academic anyway...
(icon directed at myself - interested in your thoughts)
I have been looking at reactOS (http://www.reactos.org/en/index.html), basically an open source version of windows, for some legacy software/hardware. So far some apps works on it, some don't, so it is a bit hit and miss, but the project is reaching a point where it is usable for some things.
Perhaps going with them is better than trying to fight MS with clinging on desperately to windows XP?
With Wine may applications work fine, and some don't, so there the picture is similar, but unlike ReactOS, with Wine the basic platform underneath the compatibility layer is a mainstream operating system, not something that is still in early stages. Yes, I know ReactOS has existed for many years, but its development has beem quite slow, just a hobby of a small group, whereas the same critique no longer applies to Linux. This matters for future hardware support. I suspect the Wine layer also has more momentum behind it than ReactOS.
I agree, the reason I went with ReactOS is because I had hardware with windows drivers. Linux only has NDIS for networking, and getting windows drivers working in Linux is near impossible. Far easier to make use of an OS that allows use of windows drivers by design :)
Wine and reactOS complement each other, rather than compete. I believe they co-operate on the API/userspace area already.
Drivers isn't really much of a problem in Linux anymore actually. One of my Linux box is running on a very exotic setup (two GeForce450s in an SLI configuration, and a SoundBlaster X-Fi E-MU PCI card). NVidia proprietary drivers for the graphics card aside (which required me to agree to enable a repository in Ubuntu and running the proprietary driver detection tool), the rest of the hardware were recognized and worked without a hitch on installing Ubuntu.
If they suceed in getting compatiability of reactOS up, then you don't need to rewrite IE6, you could just rip it out of an old windows install and run it as is. That is the benefit. I doubt wine (for example) will ever support activeX, as while it has API compatability, the OS-specific bits in IE6/ActiveX will never mesh with the linux system (at would be a massive job to do).
Out of interest made a search for activex at forum.winehq.org, and some forum posts there talk like IE + activeX works on Wine after IE is installed, at least in some cases (since the forums are for problems, you naturally see more comments about situations where it does not work). Someone with serious activex legacy code problems might want to take a closer look at it.
Well, here's hoping ReactOS 1.0 goes gold the day MS shuts off support for XP (or worse, shut off XP's activation servers). I'd really love to dump all the MS crud I have and switch to ReactOS, but sadly, there's just games I have that cannot be run under ReactOS or even WINE because of stupid copy protection mechanisms that for some reason fails in ReactOS, or because it used undocumented calls that ReactOS cannot handle. Hopefully, by then ReactOS would've ironed out all their bugs and problems by then.
We're just about to start a rollout of Win7 to ~2500 machines in September.
Plenty of planning and pain so far, but the bigger picture is that we've XenApp'd every single app in use so that in future we can shift to pretty much any OS without affecting users to any great degree.
Doesn't bode well for any aspirations Microsoft have for charging for Windows 7's successor.
In a corporate environment, LAN security is probably pretty tight to begin with, desktops will be firewalled and stripped of unnecessary services, and all end-users will be running with limited privileges (as Dave Cutler intended). To that, add the fact that XP has comparatively few security issues at the OS level. Microsoft Update delivers patches each month but it is apps that are being patched more often than not.
Corporate admins also have options that aren't open to ordinary users, such as using a Linux (or Mac) desktop and RDP-ing into a Win2k3 Terminal Server for those badly-behaved mission-critical legacy apps we keep hearing about.
Lastly, bear in mind that anyone still running XP has already decided not to jump to Win7. If they have a compelling reason to avoid Win7, Win8 isn't even going to be considered.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
