Visual Studio 2012 often crashes, the API documentation is often incomplete and not helpful, there are quite a number of incompatibilities between W8 and WP8, XAML has mediocre performance on ARM and can't be accessed from C++ on WP8, the list goes on and on.
If anything, Microsoft managed to make most Windows developers I know want to work on linux/Android or OSX/iOS instead - the rest are blissfully unaware of what UNIX is in the first place.
Sure these things may improve over time, but right now developers are nothing more than beta testers. For us, the possible profits are not worth the efforts to deal with the mediocre development experience.
We've bailed and moving to java on the server and web and android apps for mobile.
The last backend application I wrote was entirely written in CoffeeScript running on node.js. While it doesn't get the performance of the JVM, the development experience has been some of the most fun I've had in years.
For future backend projects, especially if they're big in scale, I'm looking at either Haskell or the D programming language.
I find coffeescript/node.js to be a bit too trendy to trust and Haskell to be perl for computer scientists.
If I had my way, I'd write everything in RPL
The JVM may be great, but the language itself is mediocre at best!
Maybe if it someday gets type inference, removes the null pointer (and therefore the stupid NullPointerException in "memory safe" code), adds lambda expressions, adds compile-time function evaluation, doesn't try to wrap everything inside a class and a few more improvements especially to limit the god-awful verbosity, I might give it a try again - but by then we'd call it Lisp.
The D language has completely replaced C/C++/Java/C# for me, it's not as portable yet but it's getting there.
You should give Haskell a try, I don't use it much yet but merely understanding concepts such as monads, type classes and applicative functors have completely changed my perspective on software development. It's just like learning Lisp; it's enlightening to say the least.
Agree with type inference, but null pointers are fine (if you know what you are doing and use commons.lang validator). Don't really need lambda, but its coming anyway. Compile time evaluation - no thanks. Verbosity is solved with tooling.
I'm proficient with common lisp and scheme (ive even built my own scheme vm) and I'd rather use java over either to be honest because it is so hard to get anywhere with lisp variants and the only ide worth mentioning (slime/emacs) is unreliable and clunky. It's a ghetto.
I spent 2 months trying to build something in Haskell. Its fine for data processing tasks, math and comp Sci focussed stuff but for productive every day use, its a pain in the arse. It's also write only.
Enlightenment for me was discovering and writing a forth interpreter and compiler. Then discovering RPL on the HP48 series of calculators. Before that it was BBC BASIC (not to be confused with mere BASIC) and ARM assembly.
I imagine java was trendy also when everything was written in C++
Java was born out of a trend and still lives as one to this day, good thing only the corporate world really bought into it!
On iOS we regularly get into the Top 10 iPhone and iPad Sports apps (both top paid and top grossing), and on Android into the Top 20 Sports apps on Google Play.
Our WP7 version has generated approximately 0.5% of what our iOS app has made, or approximately 2% of our Android app.
We also have had a Mac version of our app available since July which has almost passed the total sales of the WP version which has been on the market for over a year (our app is primarily targeted at mobile device users while they're at a sporting event, so lower computer sales are expected).
The jury is still out on how our Windows 8/RT app will sell in relation to the others since it was released just prior to the launch of Win8 but after the majority of our sport's season had completed for the year.
Our app does spread pretty virally at the events. In our completely unscientific polling of users, most of them found out about the app from other users, and I would guess that those WP users probably just assume that it's not available to them (we do give it equal promotion with the rest of the versions on our website).
As far as the release of WP8 goes, we haven't seen any noticeable uptick in sales as a result.
Both our iOS and Android versions have tablet optimized interfaces and we see a significant amount of tablet use for both.
A somewhat interesting side note, our iPad top chart rankings lag behind iPhone by a few hours. Presumably people get the app on their phone while at the event and then go home and install it on their iPad as well.
That's why I started this thread though. See if any indies have something out yet and are making actual money comparable to either iPhone or Android.
BTW, the emulator is horrible.
Speed yes, but stability? I've never had the Android emulator crash, and haven't heard of (m)any cases of it either.
Also, is the WP8 a simulator or an emulator?
I'm running Windows in a VM on OS X 10.8 so I wonder what you could be doing to make the WP8 emulator suck.
- A lot of people answering questions (even if WP itself doesn't have a big developer community, lots of questions are common to Silverlight, C# and .NET platform ones so you have plenty of resources). The documentation is really good, at least comparing with what I've seen on Android and iOS (Android was far worse when I looked into it).
- VS2012 is also pretty good. Of course, you don't get as deep as you would with a plain C compiler, but that's another point. I've heard that it really becomes great when you use all MS tools like TFS for version control and collaboration. With git... well, there are plugins but I think it's better to use the console. Same with unit tests. There is a library for unit testing on Windows Phone and it's good enough, but I expected something more from MS.
- The store is a 5/10. It's good enough, but it can improve a lot. Luckily, they're very open to new ideas and are improving each month.
- Making money: My app is free, made 3000 downloads last month. Not a bad number taking into account that it's difficult to compete with well established apps like Mehdoh, Rowi, Gleek and Peregrine and I don't get so much exposure. But I'm pretty sure that this number could be far higher: on average, I got ~50 downloads per day, but the app was highlighted three days in the spanish Store, about 1200 downloads more.
So, with a bigger user base and more exposure in the store (localization helps a lot, for me it almost tripled downloads of the app) you can make money. Of couse, if nobody downloads your app, you won't make any... But there's less competing apps here, so it's easier to get to the top of the market. You won't get as many benefits as if you were on Android / iOS, but probably there you wouldn't be on the top.
Price is something you have to take into account; price it too high and people won't buy it, which in turn gives you a lower ROI. Then you have developers complaining that their WP app didn't turn a profit and the market is dead when in reality the price was just out of sync with other platforms. Consumer's aren't blind, and regardless of the platform they don't care the reason why your prices are high. All they care about is iOS gets it for 99c and they're saddled with a price significantly higher than that.
The thing is, apps are priced as they are because of risk. In general, apps are worth either 0 or significantly higher than a dollar. If it turns out you don't like it and you don't use it, then it's worth absolutely nothing. On the other hand, if you like it and end up using it all the time, it's worth far far more than $1.
If you already know you like an app because you've used it on another platform, or tried a trial or demo version, spending 1.99 or 2.99 really shouldn't be much of a hurdle.
Also, I suspect those devs started out at the same price point, and when they sold 1-5% of what they do on iOS or Android they decided to raise the price.
I didn't set the standard price for apps. The market did. Is software worth more than 99c? Sure. But that's the standard price for a standard app. $3 is seen as a major hurdle. Some big-name publishers are putting out Pac Man and Galaga etc at $7 on Windows Phone. The reviews section is full of "this is way too expensive, I'm just playing the demo over again". I have paid $15 for an app, but not many people will. If you want to sell, you need to price it at an appetizing rate. $3 is fine if you can justify it to the market and it's a fair price. Telling users "your platform is more expensive to develop for" will get the developer nowhere.
It's basic economics. Valve even did an analysis that is posted occasionally here on HN showing that their profits went up when they reduced prices. I've heard some developers say that their ad revenue on free apps is more than their sales revenue on paid apps, regardless of the platform. Whatever the reason, if you think that dollar or two difference is insignificant, read the reviews of some WP apps with a price differential between marketplaces.
And I don't think a Windows 8 app can be the same as a WP8 app either. At the very least you'll need a whole new UI and design, and most likely quite different code, too. It's not as easy as Microsoft says: "write once, publish to all our platforms with Windows in their name".
So, they are very similar and porting from one env to another usually isn't a huge deal, but issues can arise. See Portable Class Libraries to make your life easier in this regard: http://msdn.microsoft.com/en-us/magazine/hh335063.aspx
TLDR; minor code changes, change screen ratio, add screenshots
Far from impossible, but not a simple 'press button, app upgraded' as with iOS or Android. Plus this is apparently changing even further for the new MS Blue sometime in the future.
You seem to have some strange idea of FUD too.
There are a few things I still need to figure out before going over there tho.
Where to get a free/cheap device? I keep hearing people saying they got free testing devices
What lang I need to write in for it to work on win7phone, win8phone and win8rt/win8
What would be a good test device? No contract, not too expensive but also a viable device.
Anywhere or anyone recommended to do the development?
Never test on a flagship device!