Slashdot Mirror
An Experiment In BlackBerry Development
ballwall writes "We've all read the stories about how lucrative selling apps on the iPhone can be (or not), but what about other platforms? BlackBerry accounts for twice as many handsets shipped as Apple, according to Gartner, so I decided to find out. I wrote about my experiences developing my first BlackBerry application including sales, platform issues, and a bunch of other things I thought new mobile developers might want to know about."
But, there more are corporate users without the right to install anything...
---- Booth was a patriot ----
RIM needs to open the platform up. Nothing more nothing less.
Interesting read, ballwall, and I truly wish you luck with your efforts.
I'm not much of a programmer, but as a SysAdmin (*nix by preference, win* by necessity) I was struck by some parallels I've observed. I find blackberries to be painful. Making them work as a mobile email device tied to Exchange requires a shiat-ton of ugly third party software.
If a client bothers to ask, (and they don't), I tell them iPhone first, WinMobile second, blackberry distant third.
BES is, IMHO, a steaming pile - java, dot.net, 32-bit only. Feh. Recent iPhones handle active sync nicely and don't bitch about self-signed certs. WinMobile is a bit harder, but install your certs and you're done. blackberries (I refuse to capitalize) give me pain.
My clients pay $$ for BES CALs, the devices get stupid and need to be factory reset often and re-activated, costing my client more $$ for my time.
A colleague says "blackberries are great, they help me spot THOSE people". I tend to agree. I honestly cannot see the attraction when there are better solutions to talk to an Exchange server - previously mentioned iPhones, WinMobile or a laptop with RPC over HTTP(S) all work more simply and more reliably, and I tell my clients so. Nevertheless, I still hear "but I've got to have a blackberry"!
When It Counts.
He's certainly right about one thing: his app has an ass UI. It's RIM's fault, of course. On the Palm, Android, or iPhone platforms even "hello, world!" looks great. On BlackBerry it's impossible to get even a simple app to look good. All apps on BlackBerry that do, in fact, look good are using full-custom drawing engines. See Bloomberg, Facebook, etc. For the small developer, doing your own custom drawing is a huge undertaking assuming you have any visual design talent to speak of.
Whoops, oversight on my part. Total sales stand at 2382 copies as of the data in the article (at an average net of $8.50 per sale I've made just over $20k). Thanks for pointing that out. I'd update the site but I'm afraid to break at the moment.
The thing about Blackberry that the business users love most about it is that it works and does exactly what they want it to do. They have their contacts, their email, their to-do list, their notes and a select few other things. They don't need much else. It's perfect the way it is for most users. Adding new software to it is not an entertaining idea for most users.
At the most, they want some mapping... google maps works quite nicely for me, but essentially, Blackberry already does what it needs to do and while some will, most users don't want anything more.
I've been a PodTrapper user since this past January, and I have recently started looking into BlackBerry Development myself. As his write-up makes clear, Marcus at Versatile Monkey has to be one of the best developers I've had the pleasure with interacting with. I really appreciate the 'insider' view of developing for BlackBerry, and I'm sure his observations will be useful for my own pursuits.
I'm sorta cribbing from his summary, but I did RTFA...
Pro:
Con
Most important lesson IMHO: "Everything is marketing."
His issues with the platform and the resources available on a BB really bring the differences with iPhone OSX and BB into relief. An iPhone is guaranteed to have a particular hardware config, and be very capacious in RAM and drive space, and has very teh shiny widgets and will always have the latest APIs; it also provides a brainless e-commerce platform to sell and install your app, to the point where buying a mobile app could be considered impulsive. You pay for all this with the fact that the Apple overmind decides if you can sell your app or not and takes its cut.
Don't blame me, I voted for Baltar.
A lot about it. That is a pretty long posting with some insight for people completely unfamiliar with the world of blackberry development.
The story gets an upmod for that, though I suppose it could be seen as a really long plug for the guy's product. He mentions the surge from advertising on blogs and even includes graphs showing his trials vs. sales over time with some bumps pointed out.
A decent read overall.
Much more on the "This will be fun... Oooh, money" side.
Well, and a lot of "#@$&* why is there no app to do podcasts on BlackBerry?". (Commuting sucks without podcasts, imo).
The interesting part is most of the costs are up front. I took two weeks off of work to write the initial version, and after that I've been tweaking it nights and one or two weekend days (two at first, maybe half of one now).
In hindsight I should have tracked my time better, but I'd estimate I spent just over 250 hours in code, and I probably spend about 6 hours a week doing support these days (much less earlier), so about 400-450 hours to this point. That works out to be somewhere between $45-50/hour.
But... the majority of the work is done at this point and I'm still bringing in an additional $1k/week with the ongoing sales, so the picture is getting continually better. It'll be interesting to see how long it keeps up. This slashdot article (and the waves it makes) will likely be the last bit of marketing hoorah I can get out of this app, so while I'm hopeful, I'm not optimistic. As I said in the article, marketing is really really important. (Well, until the App World gets some marketing of its own, I suppose).
Regardless of how it turns out, it was still a lot of fun. And my users seriously rock.
-Marcus
I actually have a free application on the BlackBerry App World called HP Printer Fun, which lets users mess with the LED screens on the HP Laser Jet printers (plus some inkjet ones too) for fun.
I've written some other apps as well and the experience is not so great. My gripes are as follows:
On the other hand, the docs are pretty good. The support group at BlackBerry dev site is simple superb. Examples are plenty and the API just freaking makes a massive amount of sense. And for the adventurous, you could use a beta version (might be released by now) of an Eclipse plugin.
I demand a free SSH client now
Alright, here you go: http://www.xk72.com/midpssh/
Oh, it's open source too (GPL). Here's the code: http://www.xk72.com/midpssh/v1.7.3/midpssh.zip
Oh, it's J2ME too, so it can run on any phone that has java.
Why, is a SSH client particularly easy to write?
Not so much that its easy to write one, but that several free ones already exist for the platform.
I demand a free SSH client now, who do I speak to?
Speak to Apple.
Snap to it, it's my right to have a free SSH client on platform ${FOOBAR}
In this case ${FOOBAR} is BSD. Its not a new platform, and it already has numerous free and open sourced options for ssh clients.
Only Apple could give you a device with BSD Unix on it, and then block and make you pay extra money to run things like ssh, telnet, ftp, bash, perl, etc...