As a developer, not only do you need to look at what platforms are out there right now, but what has been available in the past and what might be in the future. The stuff in the past is important because designing an application for say Windows XP without regard for previous versions of Windows dramatically reduces your target audience when with a (usually) minimal amount of effort you can support older versions as well.
However, the important thing to look at is what the future holds. Not only does it help to know so you don't code for soon-to-be obsolete features, but it gives you the scope to begin planning
Microsoft have been amazingly tightlipped with their future plans. Of course, they're raising every single flag regarding Windows .NET Server and pushing everyone's attention that way, but for Longhorn (the codename of the next desktop version of Windows) there has been little announced or confirmed. Looking for confirmed facts about Longhorn, and possibly more importantly the versions of Windows AFTER Longhorn is like the proverbial search for the needle.
I've spoken with a number of Microsoft staff who remain keen evangelists of their products despite not saying too much. They've mentioned a few things - some are obvious, other elements have started being announced in various keynotes as the corporation gets ready to lift the lid. I've also talked to a variety of Windows developers to see what they are hoping or dreaming Windows will be like in the future. It appears that some of their dreams are going to come to fruition in the next few iterations of the Windows desktop.
Even the moniker -desktop" may not mean as much in the future. One of the biggest areas of change is the way the user interacts with their PC. Arguably, the Windows XP desktop environment is much the same as the original Microsoft Windows interface back in the 80s. It's agreed that it looks much prettier, and has had several UI improvements (well, more than several) such as the taskbar and Start menu, but it's still primarily a static environment in which the user requests actions to be performed.
In the future Windows, starting with Longhorn, will have a refresh of the interface. It's still difficult to determine how -revolutionary" it will be, but one thing that has come through is that it is a much more feature-rich experience. Incorporating not just two-dimensional graphics, the intended -desktop" will be a three-dimensional environment, with layers of 2D images mixed in with video, animations and depth.
These changes are going to challenge developers who rely on their own applications seamlessly integrating in a visual sense with the operating system. This user interface model is going to allow users to more intuitively share and organise their information.
The current desktop model doesn't provide this in light of the huge amount of data computer users are presented with on an ongoing basis. Instead, windows upon windows hide information from the user and cause productivity losses as workers miss opportunities to harness the power of the operating system. The future iterations of the Windows environment will not only change that by default, but will come with a much easier to manage environment for developers to hook into. This new framework is currently planned to have its own developer -toolkit" as it were.
One area that has started to see some change already is the mobility space. However, there is still a long way to go at getting highly successful and automatic integration of mobile devices such as PocketPCs and Tablet PCs with traditional desktop and server machines.
The next few versions of Windows are going to bring all of these disparate devices together in a more cohesive fashion. As wireless networking becomes much more accessible and access to the internet becomes easier - more and more Windows components will take advantage of -hotspotting" where synchronisation of data happens when possible as opposed to real time, all the time.
Other speculations by outside developers that haven't been denied outright by Microsoft are an increased usage of alternative input devices. We've already seen the launch of the Tablet PC, and that used in conjunction with some of the new UI advances is going to make interacting with our PC's a different enough experience. But consider adding to that proper voice recognition running right throughout Windows by default, including security measures with your -voiceprint".
Expect the announcements to start coming thick and fast over the next few months about Longhorn. It is currently scheduled for sometime in 2004 but with past history, and Microsoft's re-commitment to getting it right first go, look to a 2005 release date at earliest.
A final caveat: In the home and small business markets, Microsoft Windows definitely has most of the market share with Linux and Mac taking up only very small percentages. Of course, this could all change in the future - who knows, in five years' time, users may not even care what Windows 2005 is like; instead they might all be on Linux. As I said in the introduction, as a developer you need to look at everything - are you headed in the right direction?
In the Future, Windows will be obsolete.
Open System, and especially Open Source, will be dominant within the next 3 years.
Services will be the ony thing to sell to a customer. Moreover, Microsoft dominance will have been vanished by 2005.
As a developer, you better get onto THAT bandwagon...