You did well to pull out early if you found this part of the process painful, unfortunately end to end it feels like death by a thousand paper cuts. Theres nothing glaring wrong through out but lots of annoyances which really add up, when compared to iOS or Android development
The bb version took far far longer than the equivalent iOS, simply because of other nonsense such as the pain in the signing process (isthesigningserverdown.com the fact it exists tells you all you need to know). Random exceptions being thrown that have no useful meaning and of course the joy of RAPC throwing errors such as "CreateProcess error=87, Parameter is incorrect" which we eventually found was due to the classpath being too long for RAPC to handle
However once complete the application it is something we are very proud of!
Articles can often be summarized in a few lines which make them seem trivial. What really counts for me is why those points mentioned above are worth complaining about.
The summarized points are good, but the article is great. I might read those points and think "hmm... not too bad", but read the article and think "hmm.. wtf!"
I feel that RIM is still missing the big picture. If RIM makes a phone/tablet that is 3x superior to the iPad and half the price, they'll still sell next to nothing.
Why? The answer is quite simple. Their software is shit. Their developer relations is shit. They are competing not vs hardware, but vs hardware + software.
They made great decisions too... lets ship the "Curve" model with such shit hardware and new software it was a downgrade from the previous top BB model (the one with side wheel). Old model = 3 seconds cold boot, new model = 3 minutes. The curve could not run any software that was remotely useful. It turned into a glorified email/phone device. The internet browsing sucked so bad I had no use for it. Not to mention that the hardware could barely handle the rendering.
So with good decisions like that, can you really expect any more out of RIM?
Honestly as an ex-indie PalmOS developer, I'm having trouble being sympathetic. I don't do blackberry myself, and don't have any particular affection for RIM, but this is just a little much.
As I understand, RIM's gonna do all the fulfillment & billing for you in the app store? This is the SDK for an unreleased product? On a new, unreleased OS? On a different architecture than your dev machine? Yeah, it may require some paperwork, and the SDK & dev docs might not be where all their developers' attention currently is.
Actually, the 10 app limit impressed me. I don't want my apps as #13 in middle of 25 variations of the same app from another developer. $20 a product just isn't expensive.
Also, why do you have to use VMWare fusion if it's an ISO? Couldn't you use any of its competitors if it's not in a proprietary file format?
It's not a question of whether you CAN make it work, but whether you should have to. RIM is in competition for developers with Apple and Google. Why would a developer choose to go with RIM when the experience is much better elsewhere. I think we can all agree that apps are one of the things that will make or break a smartphone platform in the current market conditions. If RIM can't attract talented developers to write apps for their platform, the outlook can't be good for them.
(1) Employers often pay for the BB, and the data plan. You're getting access to a lot of users who wouldn't otherwise be customers in the smartphone arena.
(2) As the phones are already part of the business's primary operations, you have some nice opportunities to sell apps that integrate with the business.
Considering that RIM's product isn't even out the door yet, and the specific strengths RIM holds, it's not 100% comparable to the iOS & Android side. You've got the chance to provide something that isn't a (a) fart app; (b) casual game; (c) train schedule; or (d) local wrapper for the website.
Frankly, the superficial weaknesses of the form fillout and download procedure just aren't substantial. That's what, half a day? The next few months are gonna be in the APIs, and the next few years are gonna be in the customer base, app market, maintenance, and support of the app.
I'm not sure the first advantage really makes a difference though. Corporate users are not buying a lot of apps. This is shown in the latest market share data, where the App Store took in $769 million in revenue, with RIM App World in second place with only $36 million. This is especially significant because RIM has a similar smartphone market share as Apple. So you've got two companies with similar market shares, and one is doing 21 times the app revenue than the other.
The 2nd point is definitely valid, but it's a niche market. I'm not saying ALL developers would abandon RIM, but just that most would. I don't think a major player in the smartphone market can afford to only go after a niche market.
I agree that a lot of the stuff I mentioned is small. But a lot of small problems add up to a big problem. It's well known in the development community that Apple and Google have superiors SDKs in terms of ease of use, as well as tools and APIs. With the Android growth explosion, they will have a much bigger app market, and Apple will probably have the most profitable market for some time to come because Apple customers have demonstrated their willingness to spend money. RIM is going to have to really up their game if they want to lure some of the talented developers away from the other platforms.
To add on to your first point, I know people who have company-provided-and-paid-for blackberries, and they buy their own iphone out of pocket and use that more. the BB is just for the bare minimum compliance at work - all their downtime is with their device of choice - and that's where their money gets spent (on apps, add-ons, etc).
Would people using company-provided blackberries - the kind companies love because they can be 'locked down' to the 'enterprise's' needs - even be allowed to purchase apps on their own?
Not to mention this market segment is usually behind a product iteration, or two, due to corporate security policies. This bunch probably won't have the Playbook or whateverphoneisnext--that can actually do anything remotely worth purchasing--vetted for them until RIM's all set and done.
#1 is a big disadvantage. Those devices are all locked down by IT staff and software won't install. RIM is a tricky platform. I found that the market is split mostly between consumers who text and corporate/government employees who have their devices locked down.
Depends on the employers. Lot more employers now are willing to switch away from BB.
There were some reports few months ago that was showing how the BB market on corporate/enterprise was declining.
For me personally, I own both Droid and iPhone, and bought lots of apps as well, with my own money.
There's no way I will do that on BB.
As an old Palm OS developer and someone who wrote code for BlackBerry OS, I can tell you that BlackBerry is much much much much worse than Palm OS ever was.
> Actually, the 10 app limit impressed me. I don't want my apps as #13 in middle of 25 variations of the same app from another developer. $20 a product just isn't expensive.
Completely agree. It's like if email spammers had to pay 1 cent per spam email - spam would go away. With Android as an example, when there's no cost to publish per App, you get a ton of garbage. I think $20/app is a good value to discourage spam.
Why not just say Mac is unsupported? A friend of mine's company spent months developing a BlackBerry app, when he got me to do their iPhone app it took half the time and worked 100s of times better. Why? Because their developers were busy writing code trying to figure out what kind of connection the device had so they could open a TCP/IP connection, seriously the network interfaces aren't abstracted away from the developer.
This is a fair point, but its a fair point that completely misses the point. If RIM wants to attract the huge numbers of developers that are flocking to iOS (and, presumably, doing all of their dev work in OS X), they should be doing everything possible to make the switch from iOS development to Playbook development as smooth as possible. And there is no way to do that without addressing the question of Playbook development on OS X.
Maybe they want to attract the huger-still number of developers who aren't flocking to iOS. It's obvious they won't get the rock star developers, but there's a lot more non-rock stars. How is the experience on Windows?
The question is: do they want to compete with Apple for developers? Given Apple's huge success, I can't see how the answer to that question could be anything but "yes".
If RIM's strategy doesn't have some component that involves trying to woo iOS developers to their platform, I predict they will fail.
At the risk of sounding xkcd-esque, what is your definition of "fail"? I agree that RIM attempting to duplicate Apple's consumer-focused strategy is unlikely to yield Apple-like results in terms of market share and immediate profit, but let's remember that that's not where RIM started out. There's always a lot of money to be made in B2B, which just happens not to be the competition's strong point.
Finally, the whole smartphone area is ten years old, consumer-focused part of it less than five years old: less than mature, and way less than settled. In another ten years, this all might have turned out to be a silly bubble as Moore's law erases any effective differences between smartphones. Will Apple "fail" when its percentage profit per device and market share falls to current Macbook values?
That same thought occurred to me when I was writing it. The conclusion I came to was that if RIM is going to offer an SDK for Mac (or any platform for that matter), they need to do it in an integrated way, just like their competitors. I have no problem with NOT offering an SDK for a specific OS, but if you make the decision to do it, don't do it half assed.
I think it's less important how "good" an SDK is unto itself, and more important how good it is relative to it's competition. In this case you have to compare the Playbook SDK to Apple's and Google's because when developers are making decision about which platform to work with, they're going to be choosing between them.
Judging by the fact that the dev kit shipped with an ISO for running the simulator, I suspect you would need some kind of VM package in Windows as well. However Windows has the free VMWare Player.
Until RIM makes its own desktop operating system I don't think it's comparable at all. And, once again, iOS is the market leader. They get to be different. RIM doesn't.
False dichotomy. Everyone knew that RIM and Nokia weren't even playing the same game. This is clear from the fact that neither had App Stores. You'll also remember that when the iPhone was introduced there was only web development, meaning iPhones had no special restrictions.
Comments
The complaints are:
$200 price to develop with 10 app limit.
Multiple registration forms.
Requiring multiple downloads.
Requiring to purchase vmware fusion.
Installer that just copies an iso.
Having to password protect the simulator.
Bad documentation on how to load an app on the simulator.
The comparative ease with which you can do these things with android and ios.
Now I feel like an idiot for spending an hour and half writing it, lol.
You did well to pull out early if you found this part of the process painful, unfortunately end to end it feels like death by a thousand paper cuts. Theres nothing glaring wrong through out but lots of annoyances which really add up, when compared to iOS or Android development
We recently developed this so speaking from some experience http://appworld.blackberry.com/webstore/content/28561
The bb version took far far longer than the equivalent iOS, simply because of other nonsense such as the pain in the signing process (isthesigningserverdown.com the fact it exists tells you all you need to know). Random exceptions being thrown that have no useful meaning and of course the joy of RAPC throwing errors such as "CreateProcess error=87, Parameter is incorrect" which we eventually found was due to the classpath being too long for RAPC to handle
However once complete the application it is something we are very proud of!
Ha. Clicked the link and came to a full page error:
Our apologies:
It looks like you are using an unsupported Operating System. BlackBerry App World was designed to support the following:
Windows
Although you can browse BlackBerry App World with your current browser, some features may not be functional.
To their credit they did let me continue.
This ones even more amusing on the actual Blackberry App World
http://failblog.org/2011/02/04/epic-fail-photos-blackberry-a...
Articles can often be summarized in a few lines which make them seem trivial. What really counts for me is why those points mentioned above are worth complaining about.
Yours was a lot more funny than the bullet points.
The summarized points are good, but the article is great. I might read those points and think "hmm... not too bad", but read the article and think "hmm.. wtf!"
I enjoyed reading it however. I will now mercilessly mock RIM along with you.
No time wasted. And you're a good writer to boot. I thoroughly enjoyed the article.
I feel that RIM is still missing the big picture. If RIM makes a phone/tablet that is 3x superior to the iPad and half the price, they'll still sell next to nothing.
Why? The answer is quite simple. Their software is shit. Their developer relations is shit. They are competing not vs hardware, but vs hardware + software.
They made great decisions too... lets ship the "Curve" model with such shit hardware and new software it was a downgrade from the previous top BB model (the one with side wheel). Old model = 3 seconds cold boot, new model = 3 minutes. The curve could not run any software that was remotely useful. It turned into a glorified email/phone device. The internet browsing sucked so bad I had no use for it. Not to mention that the hardware could barely handle the rendering.
So with good decisions like that, can you really expect any more out of RIM?
Honestly as an ex-indie PalmOS developer, I'm having trouble being sympathetic. I don't do blackberry myself, and don't have any particular affection for RIM, but this is just a little much.
As I understand, RIM's gonna do all the fulfillment & billing for you in the app store? This is the SDK for an unreleased product? On a new, unreleased OS? On a different architecture than your dev machine? Yeah, it may require some paperwork, and the SDK & dev docs might not be where all their developers' attention currently is.
Actually, the 10 app limit impressed me. I don't want my apps as #13 in middle of 25 variations of the same app from another developer. $20 a product just isn't expensive.
Also, why do you have to use VMWare fusion if it's an ISO? Couldn't you use any of its competitors if it's not in a proprietary file format?
Finally, as for all the 'cryptic commands'... write a shell script. ( http://tldp.org/LDP/abs/html/ )
If you want a button, write an applescript app that shows a button & runs the shell script. ( http://www.mactech.com/articles/mactech/Vol.21/21.04/IntroTo... )
It's not a question of whether you CAN make it work, but whether you should have to. RIM is in competition for developers with Apple and Google. Why would a developer choose to go with RIM when the experience is much better elsewhere. I think we can all agree that apps are one of the things that will make or break a smartphone platform in the current market conditions. If RIM can't attract talented developers to write apps for their platform, the outlook can't be good for them.
RIM's got two nice advantages to it:
(1) Employers often pay for the BB, and the data plan. You're getting access to a lot of users who wouldn't otherwise be customers in the smartphone arena.
(2) As the phones are already part of the business's primary operations, you have some nice opportunities to sell apps that integrate with the business.
Considering that RIM's product isn't even out the door yet, and the specific strengths RIM holds, it's not 100% comparable to the iOS & Android side. You've got the chance to provide something that isn't a (a) fart app; (b) casual game; (c) train schedule; or (d) local wrapper for the website.
Frankly, the superficial weaknesses of the form fillout and download procedure just aren't substantial. That's what, half a day? The next few months are gonna be in the APIs, and the next few years are gonna be in the customer base, app market, maintenance, and support of the app.
I'm not sure the first advantage really makes a difference though. Corporate users are not buying a lot of apps. This is shown in the latest market share data, where the App Store took in $769 million in revenue, with RIM App World in second place with only $36 million. This is especially significant because RIM has a similar smartphone market share as Apple. So you've got two companies with similar market shares, and one is doing 21 times the app revenue than the other.
The 2nd point is definitely valid, but it's a niche market. I'm not saying ALL developers would abandon RIM, but just that most would. I don't think a major player in the smartphone market can afford to only go after a niche market.
I agree that a lot of the stuff I mentioned is small. But a lot of small problems add up to a big problem. It's well known in the development community that Apple and Google have superiors SDKs in terms of ease of use, as well as tools and APIs. With the Android growth explosion, they will have a much bigger app market, and Apple will probably have the most profitable market for some time to come because Apple customers have demonstrated their willingness to spend money. RIM is going to have to really up their game if they want to lure some of the talented developers away from the other platforms.
To add on to your first point, I know people who have company-provided-and-paid-for blackberries, and they buy their own iphone out of pocket and use that more. the BB is just for the bare minimum compliance at work - all their downtime is with their device of choice - and that's where their money gets spent (on apps, add-ons, etc).
Would people using company-provided blackberries - the kind companies love because they can be 'locked down' to the 'enterprise's' needs - even be allowed to purchase apps on their own?
Not to mention this market segment is usually behind a product iteration, or two, due to corporate security policies. This bunch probably won't have the Playbook or whateverphoneisnext--that can actually do anything remotely worth purchasing--vetted for them until RIM's all set and done.
#1 is a big disadvantage. Those devices are all locked down by IT staff and software won't install. RIM is a tricky platform. I found that the market is split mostly between consumers who text and corporate/government employees who have their devices locked down.
Depends on the employers. Lot more employers now are willing to switch away from BB. There were some reports few months ago that was showing how the BB market on corporate/enterprise was declining.
For me personally, I own both Droid and iPhone, and bought lots of apps as well, with my own money. There's no way I will do that on BB.
As an old Palm OS developer and someone who wrote code for BlackBerry OS, I can tell you that BlackBerry is much much much much worse than Palm OS ever was.
As someone who, at one point, tried to write for Palm OS: wow, that must be awful.
ON the bright side, I think the Playbook is based on QNX, and hopefully will have a different experience from BlackBerry OS...
> Actually, the 10 app limit impressed me. I don't want my apps as #13 in middle of 25 variations of the same app from another developer. $20 a product just isn't expensive.
Completely agree. It's like if email spammers had to pay 1 cent per spam email - spam would go away. With Android as an example, when there's no cost to publish per App, you get a ton of garbage. I think $20/app is a good value to discourage spam.
Isn't the requirement to get VMware Fusion Mac-specific? Might as well complain about not being able to write iOS applications on Windows.
Why not just say Mac is unsupported? A friend of mine's company spent months developing a BlackBerry app, when he got me to do their iPhone app it took half the time and worked 100s of times better. Why? Because their developers were busy writing code trying to figure out what kind of connection the device had so they could open a TCP/IP connection, seriously the network interfaces aren't abstracted away from the developer.
This. The way networking is done on BB devices is positively archaic unless you pay RIM $2K for BIS access. ;deviceside=fuck you RIM.
This is a fair point, but its a fair point that completely misses the point. If RIM wants to attract the huge numbers of developers that are flocking to iOS (and, presumably, doing all of their dev work in OS X), they should be doing everything possible to make the switch from iOS development to Playbook development as smooth as possible. And there is no way to do that without addressing the question of Playbook development on OS X.
Maybe they want to attract the huger-still number of developers who aren't flocking to iOS. It's obvious they won't get the rock star developers, but there's a lot more non-rock stars. How is the experience on Windows?
The question is: do they want to compete with Apple for developers? Given Apple's huge success, I can't see how the answer to that question could be anything but "yes".
If RIM's strategy doesn't have some component that involves trying to woo iOS developers to their platform, I predict they will fail.
At the risk of sounding xkcd-esque, what is your definition of "fail"? I agree that RIM attempting to duplicate Apple's consumer-focused strategy is unlikely to yield Apple-like results in terms of market share and immediate profit, but let's remember that that's not where RIM started out. There's always a lot of money to be made in B2B, which just happens not to be the competition's strong point.
Finally, the whole smartphone area is ten years old, consumer-focused part of it less than five years old: less than mature, and way less than settled. In another ten years, this all might have turned out to be a silly bubble as Moore's law erases any effective differences between smartphones. Will Apple "fail" when its percentage profit per device and market share falls to current Macbook values?
That same thought occurred to me when I was writing it. The conclusion I came to was that if RIM is going to offer an SDK for Mac (or any platform for that matter), they need to do it in an integrated way, just like their competitors. I have no problem with NOT offering an SDK for a specific OS, but if you make the decision to do it, don't do it half assed.
I think it's less important how "good" an SDK is unto itself, and more important how good it is relative to it's competition. In this case you have to compare the Playbook SDK to Apple's and Google's because when developers are making decision about which platform to work with, they're going to be choosing between them.
Judging by the fact that the dev kit shipped with an ISO for running the simulator, I suspect you would need some kind of VM package in Windows as well. However Windows has the free VMWare Player.
The choice of VMWare over VirtualBox is worth complaining about then.
Until RIM makes its own desktop operating system I don't think it's comparable at all. And, once again, iOS is the market leader. They get to be different. RIM doesn't.
Except iPhone OS was also different back when Symbian was the market leader and Blackberry was the #2. People complained then. It still rose.
False dichotomy. Everyone knew that RIM and Nokia weren't even playing the same game. This is clear from the fact that neither had App Stores. You'll also remember that when the iPhone was introduced there was only web development, meaning iPhones had no special restrictions.
Native applications and the App Store were introduced in iPhone 3G / iPhone OS 2.0, when iPhone was still far from being the bestseller it is now: http://en.wikipedia.org/wiki/File:IPhone_sales_per_quarter_s...
What about using the free VirtualBox, like Palm and Android dev kits do?