I write Blackberry software as a major component of my day job. I was basically thrown into the position as part of my companies mobile division because nobody else wanted to work with the platform. After a few weeks of learning the platform I understand why.
The Blackberry OS was never designed with general consumers in mind. I'm assuming that RIM was under the impression that the Blackberry API would be used by programmers to develop in-house software. Why do I say this? Until OS 5.x, there's no built in way to determine how to get a Blackberry on the internet. In order to reliably get the device online, you have to sequentially attempt to connect using the six documented connection methods. Ironically, there's an undocumented seventh method that RIM will tell you about for $1,200. The details on how to access this without paying the RIM tax are readily available if you simply google "blackberry undocumented mds-public". The only rationale behind why they'd make this so difficult is that they didn't foresee the advent of 3rd party apps being written for the device, or alternatively, they just don't give a crap and think taxing people who develop for the platform is good business.
Forunately, as of 5.x, they finally provide a closed source class to get your device online; however, 90% or more of the phones on the market don't run 5.x, so you're still stuck figuring out how to do this yourself. This is discussed endlessly on the forums with each programmer posting his or her personal example of how to get their customers online and reading these posts has become pretty comical after a year of working on the platform.
Unfortunately, this is just the tip of the iceberg. Additional pains of working with Blackberries include the following.
* Five minute reboots between tests for any sizeable app
* HTTP requests must be constrained to a max of 100k
* Code signing requires a 3rd party server that's down frequently
* The simulator is constrained to 10k/sec downloads
* The Java API and UI toolkit are unbearably ancient
All this considered, I think RIM will continue for a good while in the same way Microsoft will. They're heavily entrenched and a lot of suits could give a crap about aesthetics. That being said, if you're thinking about exploring the Blackberry platform, there are better ways to spend your time.
Comments
I write Blackberry software as a major component of my day job. I was basically thrown into the position as part of my companies mobile division because nobody else wanted to work with the platform. After a few weeks of learning the platform I understand why.
The Blackberry OS was never designed with general consumers in mind. I'm assuming that RIM was under the impression that the Blackberry API would be used by programmers to develop in-house software. Why do I say this? Until OS 5.x, there's no built in way to determine how to get a Blackberry on the internet. In order to reliably get the device online, you have to sequentially attempt to connect using the six documented connection methods. Ironically, there's an undocumented seventh method that RIM will tell you about for $1,200. The details on how to access this without paying the RIM tax are readily available if you simply google "blackberry undocumented mds-public". The only rationale behind why they'd make this so difficult is that they didn't foresee the advent of 3rd party apps being written for the device, or alternatively, they just don't give a crap and think taxing people who develop for the platform is good business.
Forunately, as of 5.x, they finally provide a closed source class to get your device online; however, 90% or more of the phones on the market don't run 5.x, so you're still stuck figuring out how to do this yourself. This is discussed endlessly on the forums with each programmer posting his or her personal example of how to get their customers online and reading these posts has become pretty comical after a year of working on the platform.
Unfortunately, this is just the tip of the iceberg. Additional pains of working with Blackberries include the following.
* Five minute reboots between tests for any sizeable app
* HTTP requests must be constrained to a max of 100k
* Code signing requires a 3rd party server that's down frequently
* The simulator is constrained to 10k/sec downloads
* The Java API and UI toolkit are unbearably ancient
All this considered, I think RIM will continue for a good while in the same way Microsoft will. They're heavily entrenched and a lot of suits could give a crap about aesthetics. That being said, if you're thinking about exploring the Blackberry platform, there are better ways to spend your time.