To view parent comment, click here.
To read all comments associated with this story, please click here.
1 and 2) In the case of MS Windows, it might not make sense to expect killer apps until the official release of an OS update. However, for most other operating systems, the killer apps are often quite visible during the beta stages of an OS release. Linux OSs in particular, but also for MacOS. The killer apps just kind of develop along with the OS proper.
Where applications are sold with the intention of providing support and committing resources to them, it doesn't make sense to release something that's not sufficiently tested. Where there's no money or hope of money being made, and no contractual obligation in any way of worrying about testing or other things, it doesn't matter nearly as much. The Killer Apps for a proprietary OS often do develop in parallel to the new OS revision, that's true, but there are associated pains with developing any application in parallel with any OS, proprietary or not.
As for the BeOS thing, isn't that exactly what Microsoft is counting on? Selling Vista based on its use in several nifty Internet-based applications? The problem is that, even if a bunch of web developers create innovative sites using Vista technologies, that doesn't help sell Vista to Joe User. If Microsoft is resting its laurels on web services, then consumers are going to (eventually) figure out that they can run them in Firefox on Linux just as well.
There's what JLG of Be, Inc. and Apple Inc. fame referred to as "tractor apps" which aren't nearly as significant as "Killer Apps" which is what Be, Inc. was hoping to do with BeIA with a single app, being the "Killer App" that would hopefully save their bacon. Microsoft is likely sufficiently happy if there's no single "Killer App" that entices people to buy into Vista, but the whole "ecosystem" with a bunch of smaller "Tractor Apps" causes them to choose Vista.
As for your "killer app dependencies," I think those only apply to proprietary OSs. Open source OSs are generally stable enough even during active development for application developers to continue development based on nightly builds and such. How do you think Linux distributions come to release with their entire complement of supported applications ready to install?
It doesn't appear you've been clued in about MSDN, you know, Microsoft Developer's Network? Granted, it isn't free as in money type of free, and perhaps it doesn't allow everyone access to nightly builds (if Windows can even be built in a single night and packaged properly without too much overhead) but developers do get access to more betas of things during development than the general public sees. There's a funny thing about MSDN: you get more insight into how much the API's of whatever is being worked on change between early versions and the final version. Wait, you didn't seriously believe that it was any different for a proprietary OS compared to an OSS one, did you, or that "stable" doesn't only apply to whether or not the OS crashes, but in the API itself and the semantics? Why do you think those API's change during the development process? Do you think it's all entirely because Microsoft said "Hey, let's change it!" without there being some form of feedback from customers/developers? They aren't living in a complete vacuum, which is a good thing, because that would really suck. Your statement should change to say that As for your "killer app dependencies." that applies for proprietary applications that are sold and expected to be supported, but isn't tied to the OS being proprietary.
Open source distributors accomplish their OS and application debugging concurrently, not in the strictly serial fashion you describe. Whether or not the pundits like what they see, open source applications tend to generate buzz with little or no marketing budget.
It appears you misunderstood what I was saying: it isn't strictly serial in that not all things that are mentioned later must wait for the previous things to complete. However, for commercial applications or where people don't do it purely for the love of doing it because resources aren't infinite and free, it largely happens this way because of practical concerns.
It appears from your answers that you're very much slanted to interpret everything towards supporting your view that a particular ideology and way of doing things is good, and something else other than that is bad, without actually stating objective facts about how things work if it doesn't suit your purpose. Please don't try to state that your opinion is fact in such cases.





Member since:
2005-07-08
1 and 2) In the case of MS Windows, it might not make sense to expect killer apps until the official release of an OS update. However, for most other operating systems, the killer apps are often quite visible during the beta stages of an OS release. Linux OSs in particular, but also for MacOS. The killer apps just kind of develop along with the OS proper.
As for the BeOS thing, isn't that exactly what Microsoft is counting on? Selling Vista based on its use in several nifty Internet-based applications? The problem is that, even if a bunch of web developers create innovative sites using Vista technologies, that doesn't help sell Vista to Joe User. If Microsoft is resting its laurels on web services, then consumers are going to (eventually) figure out that they can run them in Firefox on Linux just as well.
As for your "killer app dependencies," I think those only apply to proprietary OSs. Open source OSs are generally stable enough even during active development for application developers to continue development based on nightly builds and such. How do you think Linux distributions come to release with their entire complement of supported applications ready to install?
Open source distributors accomplish their OS and application debugging concurrently, not in the strictly serial fashion you describe. Whether or not the pundits like what they see, open source applications tend to generate buzz with little or no marketing budget.
So, in response to your post as a whole, yes, the article is a little silly. Microsoft's development and release strategy is outmoded, inflexible, and inefficient, and therefore it is outrageous to expect 3rd party killer apps at this stage of the game or anytime soon... or ever, for that matter. Microsoft is finding it lonely at the top, and their wildly successful network of independent partners is crumbling. If they want killer apps, they're going to have to make a withdrawl from their massive war chest and develop it themselves.