Native Apps, or Web Apps? It’s the great debate of the mobile world right now, essentially fueling the platform wars from behind the scenes. Palm took the Web App route with the Pre and webOS, though with the SDK just now available to all its a bit too early to gauge that decision. The iPhone began its life with Web Apps, only to later open up native support and become the apotheosis of how app development and distribution can be done. Even Google, who will try to jam just about anything into the cloud, is putting a lot of weight behind running things locally on their Android platform.
Still, Google VP of Engineering Vic Gundotra says Web Apps are the way.
During a panel at Mobilebeat 2009, Gundotra put his full support behind the Web:
“We believe the web has won and over the next several years, the browser, for economic reasons almost, will become the platform that matters and certainly that’s where Google is investing.”
For developers currently accustomed to developing in Objective-C or J2ME, this might sound like crazy talk; in its current state, the only time we turn to the mobile web for app development is for the most basic of services. Twitter client? Sure. Complex 3D games? Yeah, probably not.
But that could very well change. With the advancement of HTML5 and Web App-centric SDK’s like Palm’s Mojo, the limitations are dwindling. Inch by inch, function by function, handset functionality is becoming web accessible. Already, some handsets are allowing applications to tie into their accelerometers and GPS receivers. As mobile broadband speeds increase and APIs are opened up, what’s really to keep us from piping those complex 3D games from afar?
To think that applications are always going to be completely locally ran is short-sighted; once the functionality of web-based applications is on par with that of native applications, the line blurs. Once the consumer can’t tell the difference between something running on their handset and something coming off the web, they stop caring. If the consumer is no longer resisting, the advantages of web apps, such as instant deployment and simplified cross compatibility, are too great to pass up.
Vic Gundotra acknowledged Apple’s move from Web App to Native, implying that Apple was simply a bit ahead of the times:
“I think Steve [Jobs] really did understand that, over the long term, it would be the web, and I think that’s how things will play out.”
We’d agree with Vic here. It’s not going to happen tomorrow, nor even in the next year – but the short list of reasons to develop natively is growing shorter each day.
[Via FinancialTimes]
It is interesting to think what would have happened with the App Store if Apple had been able to get Flash working on iPhones when they were initially released. It probably wouldn’t have been anywhere near as popular as it is now if they had.
Just look at a browser like Skyfire. Having Flash on your phone really opens up a world of possibilities.
It’s coming.
As an iPhone user I’d really prefer to have Flash capabilities in my phone. Its quite frustrating at times when you can’t watch / surf certain videos and websites which are Flash enabled.
They are certainly not talking about Flash. They are talking about HTML 5, which may turn out to be the Flash killer. Adobe missed this boat – I sat on the plane with one of their executives, and I understand why after that flight: Militant Arrogance and stupidity.
Um, it’s not Adobe’s reluctant to support Flash on iPhone – it’s Apple’s arrogance.
Do you know that anyone can decompile your flash file and read your source code? Do you know there is a thing called “crossdomain.xml”? Do you know flash (or actionscript as a language) does not support database connection? As a developer, I would not use flash even it was on iPhone.
I’m not sold on flash being entirely an Apple issue. I think the carriers will still try a mobile video play, and the bandwidth increase would grind AT&T’s already stressed network into the dirt.
Very interesting as we are just making decisions on mobile integration for http://smibs.com
I see the mobile browser interfaces still as the major roadblock but in the long run I agree with Google’s point.
this sounds so logical and almost obvious to me… I always had a hard time justifying developing anything on the device if it can be pulled from the cloud. Cross platform, centralized updates and fixes. And yes, it’s a matter of time so unless you need to hit the mobile market now, put your bets on web apps.
@MB – I agree to a certain extent but being able to pull stuff from the cloud nearly any time means having rock-steady unlimited data packages at affordable prices for the masses. We’re still a long way from that in the United States, and many of the developing countries (which do provide a lot of revenue for handset makers) aren’t even close. So, yes, this is undoubtedly the future – the question is how far away is it?
Good points. It all comes down to where your market is and what kind of app you’re building.
My thought was based on things that you’d normally find on the web today but it I think of games, yes, there’s a huge industry running on native platforms (sony, nintendo, etc) that the web will take a very long time to displace.
And from an audience POV, yes. Canada (where I live) is much worst that the US for mobile data plans but that’s changing fast (I am writing this now over free wi-fi at a coffee shop) — but yes, if you go out of NA or Asia affordable and widespread connectivity is probably a decade or so away
One big thing missing on iPhone to let the web app rule. Multitask. This is a pain in the ass when you get a link from email or any other application and you loose the context when you click. Native apps needs to add a browser in their app to solve that. Weird!
That’s an iPhone specific problem and blackberrys don’t suffer from that even though these are native apps as well.
Ten years back, the so-called “Web Services” were already supposed to crush standalone apps.
It is not the case seemingly.
I have used many of them : Business Object “online” is shoddy compared to the standalone Exe.
Outlook 2007 already exists as a Web Service and, surprisingly, the layout is nearly identical to the one on your hard disk. But it runs noticeably slower, and it lacks a few details.
Besides, Don’t forget that Web services will have a bad impact on network bandwidth.
Conclusion : Web services were a failure. No matter : let’s change the name to “Cloud computing”? With such a cool name, may be it’s gonna work this time.
I agree with the article. We have tried both ways of developing for mobile devices (starting with Nokia 3510i), so I can compare both approaches.
The web way is for sure much more flexible and less expensive to develop for.
I also think because the number of platform will increase in the future, web will be the only way.
tomas zeman
Why does everyone assume we all want to store our personal data in the cloud?
Ok, facebook and idiots apart. I can’t see in my lifetime, that I would want to give others easy access to my data. I want my data stored locally or where I can keep an eye on it.
Security, in the cloud will have to get a whole decade or two better before I trust web apps for anything meaningful.
So, would you store it on your phone instead? choose your bet here: either store it on the phone and risk getting stolen or store it on the cloud and risk getting stolen and published by techcrunch. What’s worse?
Do you use online banking much?
One more note for caution, what happens to personal data when the company managing it goes out of business?
Does it end up being sold on ebay along with the hard drives, or ’sold’ on to lots of other companies, or worse still criminal networks.
You can have your web apps but they must absolutely be able to store data locally, with that absolutely being the default action.
Let me know when, for example, a web app has full access to the iPhone’s address book.
I’d much rather write web apps, but there are things crucial to the functionality I want to offer users that native apps can do but web apps can’t yet.
In this debate, one must keep in mind that Google is trying to stack the deck.
For them, eyeballs online translates, nearly directly, to revenue from ad sense and ad words. Where google doesn’t make money, is when users are on Desktop or embedded applications.
This is why Chrome is a Web-Based-Operating system. This is why they want mobile web-apps to win in the end…and this is why you should take their “Technical Analysis” with a little bit of salt.
Exactly. Google would gladly burn money to “steer” the market to a browser-centric mobile world. That’s the only reason they are doing Android and Chrome OS. Feel manipulated yet?
Yes indeed. All about mobile applications and sites that function well on mobile platforms.
SMACK IT.
but the short list of reasons to develop natively is growing shorter each day.
I’d say the long list of reasons grows shorter. Still a veeery long way to go.
Google are wrong! Internet computing yes … HTML apps in the long term.. nah!
HTML as a publishing platform, beautiful. HTML as an OS ? Nah..
Why? As a developer you spend half or your time just making sure you data display correctly! What a waste of time! Developers put up with HTML for apps not because its good, but because its ubiquitous and cheap.
But the “ubiquitous” nature of any platform is not the reserve of html. Flash on desktops is almost 95%. Silverlight can get there as well. Only takes 2 mins of anyones time to install. And what do you get? A far more powerful framework than whatever html 5 (when its finally ratified!) will provide.
Do you think the Aviary photoshop type apps written forFlash could be made in Html? HTML is a publishing platform that has been bastardized into a RIA platform. Its not designed for it, and it takes a gang of phd heads to create any apps in html that could compete with the traditional desktop paradigm.
Silverlight and Flash are the future of applications on the web. When mobile phone processors start getting faster and Flash/SL are ported it’s game over for HTML as a RIA.
In fact, the only thing that could stop that happeining is Apple and google Android deliberately blocking Flash and Silverlight. That just seems to be there strategy. Nothing to do with html5 being better for ria’s. Apple and Google just dont want to lose control to MS and Adobe…
Agreed. At least somebody gets it! Several years from now HTML5 capable browsers will have reasonable penetration but it could still be more time consuming and require more resources to create the same things (or close) to rich apps developed in Flash platform or silverlight. The dev tools are far superior than html + javascript, etc. Plus, Adobe has new API’s and functionality in development and can get new plugin versions up to reasonable penetration in 18 months. The cycles are much faster.
Ask a Mac or Linux user how fast Adobe is at getting working plugins for any of their products. Then ask them about 64-bit support.
That, my friend, is why proprietary plugins like Flash/Silverlight are not the answer.
Flash continually crashes Safari on my Macs, and I want this on my iPhone? Also, which Flash shall I develop for? The common framework, ala Java, is sadly not common.
Market share, and device penetration will be as important in the mobile space as in the desktop. I see web applications working incredibly well for certain experiences and lacking in many others. There will be plenty of optimized, web-like, networked applications created for dominant platforms, because they perform well.
Web will win, if winning means that most capabilities should be cross-platform. It will not win in the performance, gaming, and entertainment space as easily. Custom experiences are still important.
Mobile apps will win as soon as google decides to update the android browser to make it support geolocation and offline browsing properly (without requiring gears) and offer the option to make web apps first-class citizens on the phone
so your phone and all your aps become useless in a dead zone or in the absence of a network?
why would i want to be dependant on the network for everything?
No the don’ẗ. Read up on HTML5.
With Google Gears in HTML5, Web apps are stored on your phone so they work offline, and you don’t need to load the web app each time you use it. It loads instantly like a native app.
Google Gears and HTML5 trying to replicate something that you can already do natively? Progress? Google’s only intention is to keep your eyeballs on their ads. That’s it.
I agree with Sanjay — Google just wants open-standards tech to catch up with far better proprietary tech (WPF, Silverlight, Flash/Flex, most native mobile app platforms, etc), so that no one else will make any money.
On the desktop, web apps never have come close to native apps for usability or features. The same is more true on mobile.
Just because Google wishes things were a certain way, doesn’t make them that way.
What Google is saying, we need standards for HTML5 that for example, brings geolocation, storage, camera, compass, tilt detector, video playback hardware acceleration and all that into the browser as standards.
No reason also that HTML5 doesn’t add native code features as well and that web apps are actually downloaded, streamed and stored on your phones internal memory. And that it then synchronizes your local and online databases by default and checks for new versions of web apps automatically if it needs to redownload it.
What about Objective-J?
you guys profiled some company that does cloud based 3D rendering a few week ago and showed that it could be used on a mobile device… now if I could just remember the company name…
You can develop native iPhone and Android applications (and Desktop, too) using Appcelerator Titanium today — all using HTML and Javascript.
Yup, the web will be the app store.
For a start? Better discoverability.
still need micro-payments across the web to go mainstream
Exactly! This is so true. Money talks!
You can develop native iPhone, Android, Palm, Nokia, RIM, and Windows Mobile using moblyng.com’s technology – all using HTML and Javascript
Vic Gundotra acknowledged Apple’s move from Web App to Native, implying that Apple was simply a bit ahead of the times:
I think you meant from native to Web! :)
Ok, if web apps are going to eliminate the need for native apps, why the heck most people install desktop apps to use Twitter? Check out the numbers!
And I mean both on a PC and on phones. People love native apps by their usability, which is almost unmatched by most web apps, specially on a small device.
As much as HTML5 will improve the overall experience, it will take many many years until we can utilize its full potential.
apple has a whole different perspective because half of their devices aren’t phones, they are ipod touches. people love playing games offline.
It really comes down to the type of app you are trying to build. Graphics intense apps or apps that leverage native device features that are not exposed to the web environment may still need to be native but many standard data display apps can can easily be web without any noticeable difference to the user. One aspect that is important for mobile apps is that the web app needs to be standalone on the device with all of its resources local and local caching of cloud data so the UI is responsive and it can function in poor or absent wireless connectivity. The only negative is that browser support is not quite there yet across all mobile platforms but that is coming and apart from this I think web apps are the future.
With technologies like Silverlight, you can write once and deploy to: Azure Cloud, XBOX 360, Windows Mobile 7, dekstop and web.
Sorry to all the open-source/HTML 5 fanboys…Google is not getting there any faster. In less than a year, you will see Silverlight everywhere in Microsoft based products.
Google is wrong here, imo.
The biggest advantage that web apps had over installed apps was portability. No matter what computer you were using, your data was there in the cloud rather than stored locally.
When it comes to mobile apps, this is a non-issue. Whether the app is installed on the device or launched from a browser on the device, the data is still with you wherever you go.
Take Facebook for example. All your data is stored in the cloud. But the best Facebook app for the iPhone is the installed app, not the iPhone-skinned web app.
Why? Because native apps provide a much better user experience. I have a hard time believing that HTML will ever achieve the response time and input capabilities of a native app.
I think this battle is far from over.
One issue is battery life. Accessing the web is the biggest drain on a mobile’s battery. Locally stored apps will always have an advantage because the code and, most likely, graphics resources, are already on the device and do not have to be downloaded each time the app is run. As people have mentioned, without the ability for web apps to be run locally from the device (without needing to download code and graphics), web apps will always be a bigger drain on battery.
Next is CPU utilization. Unlike web apps, the code in native apps is already compiled, is ready-to-run, and does not have to be interpreted at run-time. Apps using interpreted languages like JavaScript will always be a bigger drain on battery than natively compiled apps.
Have to ask people who own the web app-based Palm Pre… how’s your battery life with heavy app usage?
It’s the Mobile World now!!!! Mobile apps will overcome PC Apps in near future.