Unfortunately, adoption of this great technology is limited by crippleware license. You have to pay even for just developing applications. Of course, you can use starter edition, but it's useless because of size limit.

The same story is repeated again and again. Tool vendors think more about profits than about growth and adoption. Sun limited usage of server side Java on Linux which allowed PHP to occupy linux server side niche. When it understood that it made a mistake, it was too late. I am afraid, we will be developing mobile applications in JS in several years' time.

The license fees are extremely reasonable for anyone doing professional development. You'll recoup the cost in one or two days of saved work.

Where exactly do you expect them to get the funds to develop this technology if they don't charge a reasonable fee to use it?

Problem is, with such high fees they have discouraged the adoption of all the students/starters/tester around the world.

Why corporation like microsoft are offering os, ides, tools and also devices free to develop to those categories of people? 'Cause if you make them learn maybe they will make/support the adoption of your platform in the professional field in which they will work.

This is why Xamarin fees are nonsense. It's ok to pay high fees in the professional field, but you must offer low fees for new entries. And not, indie costs (299*2/3) are not enough low, and the starter pack is just useless.

Partially agree. When I choose a technology for my project, I have to prove to my supervisor that technology is a good fit. In case of Mono.Touch, its usage leads to too many risks. What if Mono.Touch is bought by evil company like Oracle (see what happened to MySQL) or decides to abuse its user base by imposing crazy fees? What if the company fails, and who is going to maintain the solution in this case etc? In case of current license, the risks related to these questions are just too high.

Don't you work for JetBrains? Couldn't the same thing be said for your tools? I love them but they're not cheap.

The situation is very different here. The core IntelliJ is open source. There're plenty of open source alternatives. So if the license is changed, there're a plenty of options to migrate project to alternative IDEs.

In case of Xamarin, there's no such choice. Only Xamarin can compile C# to iOS, and if the license changes, the whole project implemented with the tool is locked in. The only option is to rewrite the whole project with alternative technology.

I think, LGPL open source core SDK + paid IDE + paid additional libraries would be a much better choice which mitigates the risk. But Xamarin for some reason decided to choose a different path.

The starter edition worked for a long while for me to get a good feel of it, and I had no problem putting down the $ since it's a great platform.

> I am afraid, we will be developing mobile applications in JS in several years' time.

Yeah, in QtQuick! (unlike Xamarin, Digia lets you use all the development tools out of the gate with the paid-license-only ones really enterprise centric (like the Oracle dbms client) and the foss version is lgpl so you can use it in your apps no problem (as long as you don't modify).

Also, C++11 has async! http://en.cppreference.com/w/cpp/thread/async

The only barrier to qtquick in mobile is that Apple won't let Qt include v8 for the same reasons Chrome can't. So you are stuck on qtquick1 on ios.

> Also, C++11 has async!

You may not have fully understood the C# async/await feature.

The C# compiler rewrites apparent linear code that awaits on a future (Task in .NET) so that all the code that comes after the await goes into a continuation of that future. It can even rewrite a loop with an await in it into a chain of continuations, all the while taking care of hairy details like propagating exceptions.

It's pretty ironic that Mono was a reaction to MSFT's licensing issues.

Now we'll need the Mono equivalent for the Xamarin iOS api's etc.

Mono-ception.

They are more concerned about making money like MSFT than making world a better place.

And there is nothing wrong with that. Xamarin staff have given the world so much; I'd say billion+ dollar industry was built on the back of their labor. About time Miguel et al profited.

I agree that it's nothing wrong in making money. All the work should be paid.

However, their license makes their technology a niche product, when it could have been the main way mobile applications are developed.

> could have been the main way mobile applications are developed

is this so bad? I see at a competitive advantage to use their platform.

I'd rather ship 1.0 and 2.0 quickly to learn from customers than focus on the one true way to build mobile applications.

> However, their license makes their technology a niche product

Does not seem very niche to me.

The price starts at $300 per person per platform per year, and that applies to you right now should you decide to build anything that involves more than a few kilobytes of code.

Nobody's going to pick up Xamarin for fun, for experimentation, for learning, for scratching a creative itch. You can't download it for your kids or recommend it to your students. They have no product for use outside of companies and that guarantees it will be and die a niche riding on companies not wanting to retrain their .NET developers for iOS and Android.

Earlier this month I got to choose between Xamarin and everything else, I have years of C# experience and I like their IDE for general c#-on-mac stuff.

I chose Haxe instead.

They tried the "make the world a better place too".

Actually Miguel started the Gnome project, wrote Gnumeric, sponsored Evolution, etc.

Even Mono was once to be used in Linux and Gnome.

They were forced off Gnome, by valid license concerns, FUD, zealots and not few indiffrence.

What are they to do? Roll over and die?

I have nothing against Miguel and Mono. I just wanted to say that a product which is going to win the market, should be really open source and really free. People would choose PhoneGap for multi-platform mobile development over Mono.Touch because it has a friendlier license and leads to less risks.

I hate their pricing too and I hope they make their products a little more accessible to the developer who is just starting out in mobile.

Ahhh yes! async was really the missing piece for mobile apps, who are so often stuck waiting for something to happen (disk IO, network IO, the user to take a picture). There are obviously ways to deal with this on every platform ... but async just makes it nice :)

I'm curious about Xamarin - has anyone used it, in addition to something like Phonegap / Appcelerator? How does it compare?

Xamarin tools are built around the idea that the core of every application tend to be platform independent and so could be written in C# (or F#). This gets compiled into native code on some platforms (iOS). Their approach is to build native apps that use the native UI design tools, so for each platform, you have to re-design its UI to follow the UI paradigms of that particular platform.

The result is still compelling: the UI is native, the underlying code is common on all platforms, you get to use 100% of your platform technical capabilities, the app runs as fast as native apps, you get the power of .Net development tools (even Visual Studio if you want) and all the cool features of C# 5.0 (and F#) to build your apps.

Xamarin, through Mono and its mobile development tools is actually building what Microsoft promised but never really delivered: platform independence for .Net (OS/X, iOS, Windows, WinRT, Android, Linux).

Are Xamarin tools really the holy grail of cross-platform app development? What are the hidden gotchas?

What about MonoDevelop for the desktop? The Windows and Mac download packages are at v4 but Linux at v2.6. Are Xamarin slowly abandoning use of GTK# ? Is this because of GTK 2 v 3 fragmentation?

Should we be supporting C++ and QT given that QT now supports iOS and Android? Given that some developers are scared of C++ what about C# / F# with QT QML? Are there any bindings which are maintained and tracking QT updates?

Xamarin doesn't provide its products on Linux. But it doesn't mean MonoDevelop doesn't work there. The latest versions run on it, it's just a matter that we don't have the manpower to keep up with packaging.

On GTK+ 2, it's still the main toolkit used by the IDE, but on windows/mac some bits of the native toolkits are used to improve the user experience.

QT on iOS and Android means you get non-native UIX. So why bother with C++?

Disclaimer: I work for Xamarin.

disclaimer: using Xamarin, using MvvmCross and went to Evolve 2013.

- Write your core business logic API once (in a Portable Class Library) using the ModelView ViewModel pattern.

- Design native user interfaces for iOS, Android, Mac, Windows Phone 7.x, Windows Phone 8.x, Windows Store (RT and Desktop) and Windows WPF.

- Bind your user interfaces against the core business logic.

Multiple platforms, native user interface designed for each platform, sharing one core business library, maximum code reuse (90%+!), using either C# or F# with async/await/linq/lambdas/anonymous types, etc.

You can use Visual Studio or MonoDevelop.

Rewriting your business logic multiple times over in different languages to target multiple platforms does not make sense. I'm curious why would anyone not use Xamarin?

Evolve 2013 Videos: http://xamarin.com/evolve/2013

Recommended ModelView ViewModel Frameworks: https://github.com/slodge/MvvmCross or http://www.reactiveui.net/

Disclaimer, I know the Xamarin founders. That being said, I haven't used Xamarin, but I was impressed to see that rdio uses it.

Previously, I had assumed that the point of Xamarin was a way for C# developers to make iphone apps. However, my understanding was that rdio uses it so they can reuse the same C# code across their iPhone and Android apps. Rdio raised $17M in funding and could easily hire native iOS developers if they wanted to.

The net result is that rdio is actually one of the better iphone apps that I use, and I was really surprised to learn that it wasn't written in objective-C.

Top .net devs are paid more than iOS devs. Because they are paid high salaries by banks and corporations.

Mid tier developers not so much.

Are there many C# jobs in San Francisco?

I have no idea about the specifics of the San Francisco job market. I live in the UK, which can be biased towards finance.

Not really.

PhoneGap and Appcelerator are for building apps in HTML/CSS/JS. Xamarin lets you write apps in C# (statically compiled to native code, in the case of iOS, last I checked) that interact directly with native OS APIs.

also phonegap is free, while xamarin.android and xamarin.ios have high license costs for developer and platform

https://store.xamarin.com/

Not the whole story... Xamarin have a Starter Edition that is free for both platforms; admittedly with limitations (maximum app size, no linking with 3rd party code other than Xamarin provided assemblies and the iOS SDK). Within those limitation, you can publish to the App Store with the Starter Edition (obviously you need a Mac as well for iOS). The Indie Edition drops both of those limitations for USD299 per annum, which is reasonably affordable from my perspective. (Disclaimer: I have paid for the Business Edition; USD999 per annum, which adds the ability to use Visual Studio and a few other things).

The app size limit is so restrictive their own demos won't even compile. It's a tremendous shame because they automatically exclude so many people.

That's 300 dollars per platform right? So 600 dollars if I wanted to create an android / ios app?

More or less

you get a 10% discount on the total price if you buy multiple licences at the same time

You can see all here: https://store.xamarin.com/#purchase

The cheapest PhoneGap offering with support will cost you at least $249/year. And for that much you get No Guarantee support, which sounds quite odd.

As i see in the link above the support for xamarin is only in the business package so it starts at $999/year (please correct me if i am wrong).

Anyway i was not talking about the support option, but about the chance of deploy a full serious multios app. With phonegap you have that for free, with xamarin you have to pay (299*2) minus discount of 10%. Just this.

wrong appcelerator doesnt use CSS/HTML , just javascript. The UI is native with appcelerator.

I'm using it, and I love it. I actually am getting some very good performance on Android since the language is actually good (related: http://jeffrey.io/damn-garbage-collection)

I was wondering if there are other alternatives out there like Xamarin. I've used their Mono for Android software and was good but sadly, the prices went too high this time so if you want to use Visual Studio you need to add a good amount of money to make it true. :(

Titanium would be the closest alternative that I can think of. Titanium allows for apps to be be developed using JavasScript and then compile natively for Android, iOS, and others.

Titanium, last I checked also do not support all the native API's like Xamarin does. Six months ago I did not see support for the ActionBar for Android apps. Even though the Titanium API is supposed to work cross platform, usually there are functions or settings that only works on either Android or iOS. It's messy in my oppinion, but it is a lot cheaper.

they dont compile JS natively. Your code is still in javascript but get executed in a native shell, instead of using a web view , but there is no js compilation into native code.

Thanks. I didn't realize that it was still running interpreting the javascript after a project was built for a device that made calls to native interfaces. My guess is that it would be close to native performance. I was under the impression that Titanium would compile to java bytecode for android and likewise for iOS, but I was wrong.

As someone not familiar with the async/await stuff, I just looked up the official MS documentation and found the explanation pretty obtuse. As much as I can tell, it's a way to facilitate, umm, "chained futures", and by the way it's suddenly OK for a type returned to kind-of-but-not-really match the method signature.

I'm sure it's powerful but it doesn't come across as obviously amazing like LINQ did.

I don't know the specifics in this article but usually the analogy I would give is that async/await is to asynchronous programming as yield/generators are to writing iterators.

When you can use yield you can use regular control flow like branches, for loops, etc to write iterators in a straightforward manner. If you can't use yield you instead end up needing to represent all that implicit control flow using explicit stacks and things like that. Similarly, async/await let you write asynchronous code very similarly to how you would write a synchronous version of the same thing.