Related
Titanium claims it can do the same app on average 70% faster than native XCode.
What's been everyone else's experience in terms of difference in speed of development （between native XCode and PhoneGap or titanium) ?
Let's say an app like Kik Messenger or Badoo ....
Typically, a good XCode developer can do it in 4-5 weeks, assuming graphics and backend are in place.
What would it take for an experienced Titanium (HTML5) person to achieve this? (roughly)
Time to market depends on quality of specifications, process and people, much more than the underlying technology or framework.
Coding a real application with Appcelerator Titanium is not that easy, and runtime performances are SLOWER than native code because it's using a javascript engine as a bridge. Especially with a big TableView, it's much more slower, and the feeling is just not the same. But once you have purged the memory leaks, the feeling is nevertheless incredibly better than with HTML5.
You should be interested in Titanium or PhoneGap(now known as Cordova) if you plan to distribute your application on other devices or if you really don't like Objective C.
If not, keep it with the Native Xcode.
I would add that Cordova will not make any UI, but let you access camera, accelerometer or GPS with javascript inside HTML5 code. You would probably use Sencha Touch or jqueryMobile with Cordova.
In my experience, if the app is not a simple template app then you would be better advised to create a native app for each platform.
As Rob says, trying to overcome the lowest-common denominator situation and overcoming limitations in cross-platform "solutions" usually means it takes longer to code than doing it natively in the first place.
You might even hit a problem which causes you to abandon ship and start from scratch as native apps. So if you decide to go a PhoneGap or Titanium route then make sure you research fully before starting and that you won't have future requirements not covered by them.
If you are an iOS developer and you are developing it only for iOS device, then it is better to code using XCode. If you are more into Javascript and developing for both android and iOS then you should use Titanium or Phonegap. Between Titanium and Phonegap, I found it easier to code using Titanium(and yes fast as well). But I am not sure how much worth is using Titanium. http://usingimho.wordpress.com/2011/06/14/why-you-should-stay-away-from-appcelerators-titanium/
I'm actually performing a fairly intensive survey of all the major cross-platform mobile development kits right now. I started by making a sample application from scratch in IOS that uses a few simple device features, and then reimplemented that as an Adroid app. Both of those took about a day to complete (the android took maybe half a day longer). Since I've never written an android app before, I think that's a good baseline in terms of comparing development time between the various other frameworks I'm testing out.
I'll update this comment in a few weeks with a blog post when I'm done, but for the moment I've been finding that these cross-platform kits are vastly more difficult to use and take a lot more time, even for the simplest applications. and despite this, there's still quite a bit of custom per-device code that has to be written for UI and fundamental idiosyncratic differences between how device services function, so you don't really get the value of a true "single code base" that you may have been expecting.
I think the main value in these may turn out not to be anything related to development time or code reuse, but instead only as a way for non-app-developers to create simple prototypes that can later be handed over to the "real" mobile developers to be built out into true native apps afterwards... Not really all that useful in my opinion, but maybe my thoughts will change as I delve into this further.
Appcelerator is not HTML5, it is a native app built in a higher level language of JavaScript. It abstracts the complexity of common elements away and provides huge value, ping me offline to know more. I run our California business.
I have developed an app for the Android and it's working well, finally, and thanks to all the help from StackOverflow!!
Now I am being asked to make it work on the iPhone. I looked at iPhone a while ago but not recently.
What does everyone think? Should I take the time to learn Objective C and iPhone and port the app or forget it?
Are there any books that cross-reference functions so that you can look up how to do something in iPhone that you already have on Android?
From my experience in school, if you have already been able to create a working smartphone app in at least one mobile OS such as android, it wont take long before you can understand objective C and cocoa framework stuff. The only problem with that is you may probably need an apple developer license to use XCode.
So, I would say go for it since you also get paid, and also here is a link to iphone development guide for android developers : http://integratingstuff.com/2011/02/27/starting-iphone-development-as-an-android-developer/
Probably, it's better to get a partner who develop to iOS than doing it yourself.
Focus on a platform and let your products run to all users.
Unless you are using a framework that supports both iOS and Android (something like the Corona SDK) you won't have much actual code that will transfer over. Ideas, algorithms, logic, graphics, designs, etc will all transfer over just fine. Those are the hardest parts (IMO) of software dev.
Objective C (the language iOS apps are written in) is not that hard of a language to learn if you already know C-based languages (like Java). There are a few concepts that are different, but for the most part, it's not that bad. The biggest challenge for developing on iOS is buying a Mac. You can program for Android on Windows or Linux boxes, but iOS apps can only be developed on an Apple. Unless there is something that has happened in the Hacintosh arena that allows for iOS development on other platforms, you're stuck buying new hardware. BUT if you already have a Mac, download XCode and go to town!
Like Haphazard said, if there is enough money in it to make it worth your time, do it.
If you are getting paid, go for it! (Also, it could be a great learning experience.)
When I had to make the same decision, I considered the following criteria:
how much money is in the app on the other platform ?
how many times will this happen in the future, or is this going to be the only app? (how big is the benefit of learning the other platform for the future)
how much insider know-how is in the app that one is willing to reveal to another programmer porting this app (in my case I do mostly device handling apps, which is not really all that common)
what is the opportunity cost of spending time on porting an app instead of developing another profitable on the initial platform
If you have any possibility, you may look into similar apps and see how they are doing on the two platforms...
Good luck, whatever you are going to do...
I just learned about this and have not tested it yet, but one thing that you could do depending on the app you have you could take a look at PhoneGap. It looks pretty promising, though it may not work for your case with your initial application already made. But in the future this could help.
Unfortunately you will either have to re-write the app from scratch for iOS, or hand over the job to an experienced iOS developer. You can fairly easily port over the logic and computations in your app from Java to Objective C, but the user interface is the area where you cannot really re-use anything (except maybe icons), and the user interface tends to be a large portion of most apps.
As an Android developer who has ported several of my apps to iOS, I can say that this transition is a hard one. Firstly, you need to buy an iPhone and a Mac (if you don't already have these), since you cannot develop apps for iOS without the Apple hardware. Secondly, you need to learn how to use XCode and Objective C or Swift. And thirdly, since XCode ONLY allows creating the user interface graphically (as opposed to Android which lets you hand-edit the XML), there are many hidden things which can cause you to come unstuck. (UPDATE: Using the new SwiftUI approach to user interface design really helps with this last point, and in my opinion makes the transition from Android to iOS easier).
Finally the Apple and XCode environments seem rather alien to someone who is used to Windows and Android Studio. There are things like the Home and End keys having completely different behaviors to Windows which is frustrating. Also you have to use a combination of key shortcuts and mouse movements to hook up user interface elements to your code. Also, there are big annoyances such as the pop-up keyboard on iOS not pushing the content out of the way automatically like it does with Android. This is probably because Android is an OS designed for multiple screen sizes, whereas iOS is design for a limited number of screen configurations, but it makes iOS feel inferior and harder to work with from a developers point of view. (UPDATE: Using a ScrollView in SwiftUI solves the keyboard obscuring problem).
I'm really worried that when I write iPhone app with Sencha Touch and put it in PhoneGap container the user experience would downgrade.
I particularly see the bottlenecks in:
fluency of the screen transitions (animations)
fluency of scrolling
Please have in mind that there are lot of 3G iPhones runnin iOS 4.x that made them very slow. I'm discarding the support for the original iPhone.
I, being a trained UI professional, can spot the ST app just by touching few things in it.
Does the shift from Safari to PhoneGap container increases the performance?
Do you have any experience with it?
I haven't used ST or PhoneGap, but I have used an app built with them. I can say definitely experience about the apps was worst in my 3GS. If you're planning a demo, prototype or test, you're safe with them. However if you want to make an app with competitive UI/UX, you should not.
(and even you want to make prototypes, it should be better having some papers)
For your question, the speed. It's definitely not comparable. HTML + CSS is feature-rich, easy to use framework. Of course, it's slow as much as it's easy. Most of HTML based UI tools uses just UIWebView which is a part of native framework. In fact UIWebView is Mobile Safari itself. So the performance of the tools never be better than Mobile Safari. If you want to check performance in animations, just visit http://www.chromeexperiments.com/ with Mobile Safari. I checked none of showcase is running smoothly even many of them does not require strong graphical power.
Native apps are compiled and optimized with cutting edge technologies from professional researches over decades. And there are a lot of options to tweak and tune the code for performance. However a few of them are applied to HTML. Because HTML should guarantee feature-rich, easy-to-use framework always. And most of optimizations (which makes performance improvement) are trade off between feature and simplicity.
However in iOS 4.3 Mobile Safari's performance is improved. But I don't believe it's meaningful for apps with shining UIs.
I saw a considerable graphic framework with JavaScript. In fact, it was game framework with scripting in JavaScript. So it has no relation with HTML or CSS. (I forgot the name of it, however it was incomplete product)
PS.
And there is another big reason for you. The UI behavioral inconsistency. The frameworks mimics native UI of iOS but incomplete. So it feels uncomfortable like imitated copycat brands.
However you have no need to care about it if you don't want native UI.
Edit
It's been a long time after I answered this question, but I realized that I also have to mention about GC. JavaScript is GC based language. It means, it has unpredictable GC time which makes main thread stops. This makes UI struggles. On native implementation, you have control to use GC or not.
This wouldn't be a problem on Android. Android always had those struggling because of GC on Java. Consequently, users will not feel any difference. But on iOS, your HTML5 based app never provide better experience than competitor's native app.
There are many workarounds for this GC time issue. Such as incremental-GC, realtime-GC and so on. But actually, there's no real solution. Because the primitive problem is you don't have control.
It turns out that putting a PhoneGap wrapper around any webapp (including all ST2 apps) can actually significantly decrease their performance. This is because the UIWebView which it uses is actually not the same as iOS's Safari browser which has had a number of enhancements to make it more perforant. One reason behind this is that as of iOS 4.3, Safari make use of the "Nitro" JavaScript engine which is pretty fast, this is not available on the UIWebView which is used by PhoneGap (although as of iOS5 it's available for webapps running in full screen mode).
A few people have experimented with the performance to see what the impact is, one looking at the performance of the Facebook App and another at canvas performance.
It turns out that the embedded UI browser maybe worse than the Andriod one as some tests have shown.
I found the performance of ST apps to suck even outside of PhoneGap.
At time of writing I would say: Good for mockup and quick app with no high demanding graphics.
But things could change over time as both project (PhoneGap and Sencha Touch) are improving day after day.
And hardware is becoming more and more powerful (iPhone 4, 5, ..., dual arm cpu, ...) so there will be a time where ST and PhoneGap based applications will have performance close to native apps, and this might come after than expected.
So keep an eye on these projects and keep on developing "basic" applications with them to test their performance.
I built a business app with Sencha Touch with and without PhoneGap. Any major performance impact is from Sencha Touch (1.x or 2), not PhoneGap. As Eonil explains, this is because PhoneGap is really using the native browser. I've noticed ST performance "good enough" on iPhone 4 (and I'd argue 3GS with iOS 4.3+; depends on your users' expectations). On Android, only really recent devices are adequate for ST 1.1, but ST 2 has focused on Android performance. I'm sure Sencha will continue to innovate and drive performance because their business depends on it.
If you do use Sencha Touch (or any JavaScript), be sure to use the minified (-debug) version.
I've also built trivial native apps on iOS and Android, and indeed there is no comparison; native is far faster.
I've made several app with ST2, and wrapped for native.
You can of course compare the two, however they are very different beasts!
Put simply, if your Sencha app performs well in Safari and Android Stock, it will perform in a similar manner once it is wrapped for iOS or Android. Phonegap/Cordova do not effect performance much, these are simply the bridge/container to the native functionality.
I think if you are worried about performance, look at the many tutorials around optimizing ST2 apps. There is lots that can be done to improve transitions and list scrolling.
If you are happy with how your app performs as a web app, then I think you have kind of answered your question already. Webapps are not native, and this masquerading done by some of the bigger frameworks is misleading. Ultimately the user will have expectations that are not always met.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 9 years ago.
Improve this question
I want to build a mobile/tablet application whose core feature will be taking pictures with the camera , viewing pictures and receiving notifications. Also I want to target iphone, ipad and android platforms.
Titanium appcelerator piqued my interest because of its cross platform appeal. However I am concerned because I've read mixed reviews on SO and other sites. The things that worry me are:
Subpar android support
Camera support not fully capable (e.g ios 4.1 HDR capability)
Camera support buggy
The nightmare scenario for me would be to invest time in titanium only to discover later on that its a major PITA and drop it and go "native"
Please share your thoughts and experiences.
I chose Titanium for a serious application, although one that does not use the camera. I think there are a variety of things that could play into your decision...
If your app intends to do "fancy" stuff with the camera, or some real heavy image processing and so on, you're likely better off going native. If on the other hand, you just want to have it take pictures, and then those will be used as-is, or sent to a server, or what not, then Titanium should work just fine. Titanium does have some processing and image manipulation things, but as others have said, if you really want to take advantage of the device's hardware, you probably want to go full native.
It should also be noted, and Appcelerator says this as well, that with a Titanium app, you won't just write a single app that works as-is on all devices. You will need to taylor the UI to each device (or class of device, i.e. iPhone, Android), because they have different UI's, and different standard UI flows and so on.
But, one of the potential advantages to Titanium is if you don't know Objective-C and/or Java, and you do know JavaScript (and in my case, I'm actually using Coffeescript :). Or, if you would enjoy your work much more writing JS than ObjC/Java. This was one of the main reasons for me. I have done some ObjC dev in the past, and don't even mind it, but this project I'm doing is on a very very aggressive schedule, and it was just going to be far more effective for me to use Titanium. I was able to get set up and build an app extremely quickly, and I am not spending any time having to become more deeply familiar with the programming language I'm using, memory management bits (you can't fully ignore this with Titanium, but essentially they're doing it for you). Based on the folks I've talked to, and how much time they spend with memory management, Interface Builder issues (this is mostly the ease of forgetting to setup connections or hook various things up, IB is actually a pretty great tool), and so on, I'm quite glad I'm using Titanium.
While I expect to do an Android version at some point, it's not a priority. But, I'm glad to know that a large chunk of my app code will be re-usable, tested, etc. and that I'll wind up mostly just building/revamping the UI for Android, not rewriting networking code, data management, and so on. Android support will be much better (supposedly) in Titanium 1.5, but you may want to wait for that release to evaluate Android if that's a priority.
Finally, Titanium does have a "module" system, that allows you to wrap native code, exposing it as a JavaScript interface in Titanium. We are about to leverage this to integrate a third party library, and at least for what we need, it looks very easy to use, and has given me a little more confidence that if some particular native feature we need access to comes up, that we'd have a decent chance of integrating that while still using Titanium, but I think it would depend on what the particular native functionality was.
Good luck and enjoy building a mobile app, it's pretty fun!
We have been using Titanium in one of our projects for around 2 months, and frankly speaking our experience with Titanium is too bad.
As per my opinion, below are some major drawbacks of Titanium:
1) First thing is you will not get debugging support at all (We can understand how debugging require in any of the project and in any of the technologies).
2) Titanium is NOT fully supporting all the features of Android/iPhone; beyond some level it will not give you support.
3) Comparing with Android/iPhone SDK, developers will get very less amount of help from the internet and API library (Titanium provides the API library help file).
These are the general issues that end developers face while dealing with Titanium and I suppose sometimes it will be tedious and frustrating work for them.
If the functionality of your application is somewhat like displaying data from the web (like many news, media type apps) then Titanium is the suitable option; otherwise not.
The Android support is not near as good as it is for the iPhone. If you were to just say iPhone I would say you would have luck using Titanium. However, I think trying to build one code base in Appcelerator and also use in your Android environment may not be the best experience.
That said, IMO doing Android / Java code is much easier than doing Objective C / iPhone work.
So worst case I would consider using Titanium for your iPhone version & do Android in Java.
You can give it a shot doing them both in Titanium, but worst case code the Java version.
I just hate objective C and the 'native' Apple development environment so much.
I would recommend against using a cross-platform toolkit when interacting with device hardware is one of the key requirements of your application. I haven't worked with Titanium before, but I find it hard to believe that they will give you the same level of hardware access that you get with native frameworks.
In particular, iOS 4.0 added a mess of new capabilities regarding the camera, including live video frame processing through AVFoundation, and I find it hard to believe that a third-party framework will keep up as these platforms advance. To be honest, it's pretty easy to write an application that interacts with the camera on the iPhone nowadays (count the number of them on the App Store as an indicator of this). I wrote a live camera frame processing application in about six hours the other day.
I can't speak for Android, but I imagine dealing with cameras is fairly trivial using the native APIs there as well.
You're also going to find performance testing and debugging your application to be far easier using the native tools than those supplied through a third party. In particular, Apple's Instruments is an extremely powerful, yet easy to work with, application for tracking down CPU and memory issues within your application.
There's also the community aspect. You'll find far, far more people working on Android and Cocoa Touch than on Titanium (just look at the numbers of questions in the various tags on Stack Overflow to see that). This means many more tutorials and a whole lot more sample code that you can use.
The time you'll spend getting your iPhone and Android build environments set up, and submitting to both stores, will be the same no matter if you go with a native environment or with Titanium.
In the end, even with learning both platforms, I think you'll come out ahead by avoiding a cross-platform solution. Trust me, I've tried to do cross-platform development before for other projects and ended up with lowest-common-denominator products that took much longer to write.
I've developed an Appcelerator-based camera application and was very pleased with it. I think some of the negative reviews come from the fact that it's a bit hard to get set up (more due to Apple's crazy developer registration process).
Once I got started, it was easy to do things like overlays on top of the camera screen. I was really expecting difficulty with that part, but it worked well.
I have spoken with the Appcelerator team in the past, and they are a great group to work with. I've seen them be responsive to other user issues, and I'd trust that if I ran across a real bug, they would address it quickly.
A little late, but my two cents...
I honestly believe you can very quickly prototype an application with Titanium Appcelerator and focus on the critical feature sets to determine if it is the appropriate tool for you.
All developers have there opinions and experience(s) which influence there comments; developers have different ways of learning and different levels of productivity... In the end, it comes down to how are you most productive with the tools available to you.
Since you are stating from the start that you want to deliver a solution an multiple platforms, I think it would be a poor decision on your part to not even spend a week or two investigating cross-platform frameworks and then making the decision based on your personal experience.
There is Titanium Appcelerator and there is also PhoneGap, where PhoneGap might help you is that there is the ability to extend/enhance the underlying framework through writing plugins (I wrote one for iphone ) and there is an android one on my blog also... this can fill missing gaps for you when you move across platforms.
Also since the UI in a phone gap solution is HTML5 Webkit based, it can give you a consistent look and feel across you devices if you like. Frameworks like jQTouch and JQuery Mobile are being used for UX with PhoneGap Application
I reviewed negative feedback for Titanium Appcelerator but I tottaly agree with Aaron Saunders that if you use PhongeGap Development is support HTML5 which can getting easy to make apps for iPhone, iPad and Android mobile.
Has anyone highlighted the cost off titanium.
I was contacted by them today and if you are more than a one man band you have to sign up for a partnership program else you are held liable for breech of contract if you release the app.
The partnership program is £5000 which is far to much for us as a start up company when it's our first application, we are currently looking for a different option now.
I am interested in writing games for the iPhone and the Web. Ideally, there would be one language that I could write my games in and it would work on both platforms. I know this is not the case, so what is the best way to leverage code between iPhone apps (Objective-C/C++) and Flash SWFs (ActionScript)?
This maybe of some help
It uses the NME library which will allows code to mostly be written to the Flash 9 API and create the C++ for XCode to compile and run on the iPhone and Touch. This creates a path to port Flash games to iPhone/Touch.
Unfortunately, Flash and Objective-C are very, very different - and it's unlikely that a Flash player will be available for the iPhone in the near future. The native input methods used in Flash games - the keyboard and mouse - don't lend themselves well to the iPhone. While Apple could make Flash run on the iPhone, most Flash games would be totally unplayable (or feel very unnatural. They'd have to overlay a keyboard probably?). With the success of the App Store and native iPhone games, I think it's very unlikely you'll see Flash support anytime soon.
You might want to consider using a game development tool like Unity instead of Flash in the future. Unity allows you to create both 2D and 3D games, and you can program them in various .NET scripting languages. Once you've created the game, you can cross-compile it for web (their own plugin, not Flash), iPhone, or the desktop.
I know that doesn't help much since you have an existing codebase, but it might be something to consider for the future!
My company is developing a toolchain that allows compiling ActionScript3 to native code for mobile devices.
It now supports Windows Mobile and Symbian, and iPhone supported will be released in a couple of weeks.
Check it out at: http://developer.openplug.com/
BR
Guilhem
Adobe Alchemy looks promising. It is not released yet, but from their website:
Alchemy is a research project that allows users to compile C and C++ code that is targeted to run on the open source ActionScript Virtual Machine (AVM2).
This would allow iPhone apps and Web apps to share non OS-dependent C/C++ code, which is a very exciting prospect.
One option would be building everything in unity. The engine facilitates building the same game project to any of the following platforms:
Webplayer
OS X
Windows
iPhone
Wii
Actually, the iPhone supports Flash technically (see Developer creates Flash for iPhone and Flash Installer Update #2). It is just Apple's crippleware restrictions that prevent installation.
Other than that, there's really not much you can do. Flash/ActionScript and Objective-C are radically different. You can have a central server store data, but that doesn't solve the duplicated logic.
If you're already willing to use ActionScript you could go all the way over to the dark side and switch to Javascript. That's the only common language supported by your clients (web and iPhone).
How comfortable you are with either development environment certainly plays a role here. If you are a die-hard Objective C and a super star Actionscript programmer then doing both shouldn't be much of a problem. It will be lots of work of course, but not a problem.
However, if you are neither or only skilled at Actionscript then I suggest you focus on Flash/Actionscript for the time being. Eventually Flash will be available on the iPhone anyway. When that happens you can already have a number of apps ready to be quickly ported to iPhone. Also keep in mind. There are more portable devices out there than just iPhone. Getting your apps running on other devices might be worth it in the mean time.
Just keep in mind when you're developing your apps now that at one point you also want to run these apps on the iPhone. So make 'm in such a way that they can be controlled with an iPhone as well.
Updating this old QA with new information. The recently released monkey development framework deploys to both iOS and Flash: http://www.monkeycoder.co.nz/
It's so new that I wouldn't necessarily recommend it, but it has a great pedigree: the creator made Blitz3D and BlitzMax before, and those were great game development tools.
That said, I would strongly recommend a combo like Corona for iOS and Flash for web, so that you're using optimal tools for each platform.