The real trick is ensuring that application functionality and familiarity remain intact. I believe there is such a tool that will allow this to happen indeed, it has worked well for many computing environments already.
One approach: run Windows software off of the Windows server while allowing the end user the freedom to use older hardware running a thin client called Thinstation. Taking this approach means both access to full fledged Windows software while enjoying the ability to utilize old hardware dragged up from the basement as a client machine.
This may be a situation where older workstations running Windows 2000 Pro are in use, but are not of value to the company due to budget restrictions that mean no new software purchases for these older PCs. This is not a problem. Just look to Free Software/Open Source for some assistance locating applications that might enable you to get some more life out of the now dated PC.
Often using various open source software applications that are available, though not able to be run natively in a Windows environment, will more than fit the bill for the needs of these old workstation PCs. Assuming the abilities of the software can match the abilities of the workstation hardware, most open source applications run on all semi-recent versions of Windows.
So what types of software can be used on an existing XP box from the open source archives? Generally speaking, everything from Web browsers to office suites. Some favorites include:
You get the idea...
As you can see, there are a number of applications that open source software can lend a helping hand with. Unfortunately, it is not a panacea for everything that a company needs from a modern workstation. Sometimes, those pesky legacy apps can get the best of us, no matter how hard we might try to embrace something with a more liberal license going for it.
When closed source software remains a legacy need
Now many businesses face the reality that closed source software will remain part of the company despite any desired open source migration ambitions. Whether it be some legacy application that is simply not available in an open source alternative format or even if the existing open source alternative just is not measuring up.
In either case, sometimes there is just no escaping it. Even with the desire to keep things in an open source atmosphere, often it proves to be more cost effective to simply bite the bullet and stick with the legacy application.
That said, I would point out the inherent danger when using closed source software with any level of long term reliance. For an example, grab a document made with Microsoft Publisher 2002 and attempt to open it with Publisher 2000.
That's right, it wont open.
That type of vendor lock-in is my biggest issue with proprietary software and the document formats they often support. The moral of that story is to avoid using proprietary applications whenever possible, but to remain flexible to using them if nothing else will do. It's an unfortunate reality, to say the least.
At the end of the day, it may be a necessary evil, but if there are open source alternatives, be sure to exhaust attempts to switch to them completely before tossing in the towel.
Cutting costs, idealism aside
Thus far we have to examine a couple basic ways of cutting your IT costs: using Linux thin clients connected to both Windows and Linux servers. And stand-alone workstations using the existing Windows installation, yet relying heavily on open source software so that the company is able to keep licensing costs down to a dull roar. After all, the idea is to cut your department's overhead, not change the world of software in a day.
And so everything described above, in a perfect world, can work without too much trouble. The biggest challenge to its success is facing your ability to implement these changes without getting into the idealistic disputes of closed source vs. open source. In short, get over the idea that one way of looking at software clearly must be better than the other this is nonsense.
Once you overcome the one way is best mindset, your company is then free to utilize what both the open and closed source worlds have to provide. This will be the winning formula of commonsense that enables the enterprise to harness the power of merging these software technologies together.
A perfect union, perhaps not. But it will definitely work and keep costs down at the same time and prove your ability to think outside of the box during tight economic times.
Thinking about servers
The final area I want to touch on is the need for servers. Obviously, theyre a fact of life in today's business world.
Yet, once again, choosing based on absolutes may not be the most logical answer. Outside of the Web server market, it seems like most businesses immediately latch onto anything Windows-based for their server needs. Clearly, that is where the support is right?
Not really. While Microsoft does well in the enterprise markets, so does Red Hat and Novell, with Canonical chugging away with its own brand of support. In the end it could very well be the most cost effective approach to provide both Linux and Windows servers, as it could save on expenses in the licensing dept.
And let's face it, not everything in the enterprise requires a Windows licensed server. Sorry, hate to be the one to point this out, but Linux often makes for a smoother solution for some tasks not relating directly to the desktop needs of Windows users.
So what is the hold up then? If this entire vision of mine is such a great idea, why are we not seeing more businesses and government agencies taking this mixed bag approach more seriously?
Well, I have seen indications that the US is badly behind with Linux adoption as a collective whole when compared to the rest of the world. Not just with businesses either. This is also a problem with schools and government agencies alike. Here in the US, if it requires a shift in what we perceive to be normal, forget it. Apparently, this is showing no signs of changing anytime soon, either.
So here is the challenge to the IT departments across the US and it all starts with you. Rather than continuing to put all of your eggs in one Windows basket, consider blending in Linux thin clients wherever possible. Even if they are connected to a Windows server, with the assistance of a Linux support team, making the "blend" should not be all that difficult once you set your mind to it. Best of all, you will be the IT guy that was able to breathe new life into the PCs from the basement that were otherwise on their way to the recycle bin.
Remember, I am not advocating going exclusively Linux here. As nice it might be from an ideological standpoint, its a pipe dream in the world's current infrastructure. Just understand that it is definitely doable for you to begin the blending process now while there are still (hopefully) some available funds to get thing into place.