Recently PC World has named Vista as #1 biggest tech disappointment of 2007. The dubious honor adds to the list of anti-Vista criticism that keeps coming in ever since the latest Windows version was officially released – mainly revolving around performance and compatibility issues.
An interesting thing is that Vista's main competitor is Windows XP, which is currently packed with “good enough” set of features, faster (and will get even faster after Service Pack 3), and has practically no compatibility issues. To be fair, Vista does have a number of features that XP lacks, and the reluctance to upgrade also happened when XP was first released (at that time, Windows 2000 was XP's strong competitor). What is particularly bad about Vista's situation is that XP excels in enough number of aspects such that some computer manufacturers insisted on providing "downgrade" to XP as an option, and many end users actually still chose XP over Vista. Worse, it seems the imminent release of Vista's SP1 won't help much with the performance issue. A person even wrote a satirical review about upgrading from Vista to XP.
From my perspective as an outsider, Vista is not so much a “horribly bad” product and more like a “lackluster” product, especially considering its 5-plus year of development time and system requirements. My reasons are as follows:
There are indeed a number of improvements, both in visible aspects and in the OS internals, like the reworked graphics driver model, better networking support, and ReadyBoost. From reading Mark Russinovich's informative 3-part overview of Vista's kernel and its features (part 1, part 2, part 3), I think a number of the core OS improvements are very neat.
Performance issues will probably be fixable relatively easily through later updates. At least, it's easier for everyone if an OS maker focuses on architecture first and optimizing the speed in later updates instead of writing “fast” initial release and change the internals and APIs change in subsequent updates. Case in point: Apple's OS X, whose initial version was much criticized as being sluggish, but became snappier with each release.
Compatibility issues will diminish as hardware manufacturers release better drivers. For this particular issue, Microsoft is not really at fault. After all, there has been ample opportunity for hardware manufacturers to develop and test drivers on pre-release versions of Vista.
That being said, there are some things that hinder improvement efforts on Vista:
"Baggage" due to need for backwards compatibility will likely make it more difficult to clean up the codebase and make the internals better. Backward compatibility is very important, but perhaps Microsoft should've just provided an emulator for this purpose, so that they wouldn't have to make much compromise when designing and developing Vista. Providing an emulator for legacy applications would prevent old cruft from hindering future improvements, and, since Microsoft already has a virtualization product (Virtual PC), this also seemed to be well within their capability.
Bloat and extra complexity due to stuff like more pervasive DRM also increases difficulty to clean up the codebase and can cause problems. This bloat is worse than the backwards compatibility baggage because in this case, the extra stuff does not provide any benefit to end users. Reliance on WGA is also problematic -- when Microsoft's WGA server failed earlier this year, many owners of legitimate copies of Windows were unable to validate their software.
With another Windows version supposedly planned for 2009 release, there's a chance that many people will just wait longer and switch directly from XP SP3 to that version, skipping Vista altogether. In addition, existing compatibility problems reduces the advantage of using Windows compared to using another OS. Hardware support in Linux, for example, keeps getting better, and so is Wine's ability to run Windows applications. All these can make the next few years quite interesting in desktop OS world.
0 comments:
Post a Comment