I think web apps will take off in the next 3 to 5 years. 10 years is too long given the current pace of development.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Google: The Web, Not App Stores, is the Future Of Mobile Development
- Thread starter MacRumors
- Start date
- Sort by reaction score
If the applications require the full potential of the hardware.
In the future a mobile device will have 4 GB of memory , 2 core 1-2 GHz processor in it. To run Word, Powerpoint, a Sales assistance , a stock ticker , etc. ( a very long list of applications ).
Sure some folks will say they are solely interested in apps that fully activitate all of the hardware abilities ( games, some niche audio/video , maybe something that has some kind of "real time" constraints).
The Google representative said "many, many applications...". He didn't say all applications. I only scanned many of the majority of the replies, but what read rings just like many ancient arguments have heard before.
The generic argument generally goes like this......
Some aspect of the OS has to be written in assembler/C , so therefore the whole OS has to be written in assembler/C. Since the OS is written in assembler/C then all applications have to be written in assembler/C. Sometimes that is followed by any customization/modifications to applications has to be written in assembler/C.
Over time folks have added C++ and Objective-C and Java into the slot that C plays in the general framework above. Amusingly if go back two decades this same general argument framework were being used by some against Objective-C/C++/Java. Two decades later, substantially faster hardware with less resources constraints and ta-da .... they are now in the "fast" category and now part of the "low level" mix of languages because those other "high level" languages are just tooooooo slow and tooooo lacking in comprehensive scope to be used for application development.
Google and web apps only have to successful prune off an subset of the application space that makes money. If a mostly portable framework shows up so that folks can cost effectively port their applications to multiple mobile and desktop platforms at substantially lower costs than writing separate versions for each prorprietary SDK with enough performance for the users to get work done..... then it likely will be successful.
Will it be universally used for all apps? No. However, none of these SDKs are going to be universally used for all apps. (whether web / C / java / whatever based. )
A couple of more specific things.
A major contributing reason that iPhone javascript apps sucked when iPhone first appeared is because the javascript implementation sucked. Very similar to the not so modern core that formed the Java runtime before Hotspot and modern technology garbage collectors were inserted. If javascript speed and web apps don't matter at all why is Apple putting effort into creating a faster webcore?
This isn't Lord of the Rings where there is just one "ring" to rule them all.
Second, attempts to recast the "cloud computing" as network plumbing is puzzling. Again really two different spaces of apps if talking apps that don't need any network and those that primarily do ( work in a distributed , client/server fashion). For collaboration applications .... how can they have high utility and not be connected? For a collaborative applications where it is an exchange of ideas why wouldn't that lead itself to a cloud ( really shared resources located elsewhere on the internet.). For instance you haven't really have a shared file system if the devices aren't hooked together.
SETI@home/BIONIC has a multipeta flop computational resources available to it. Is that useful for all applications? No. Is there a viable niche that does derive utility from massively distributed applications. Yes.
In the multibillion dollar business application market distributed, client-server architecture are not rare. It is somewhat myopic to app architectures that dominated early 80s non-networked PCs and say that is the only architecture that matters in this next century.
Google is still now a days very popular for having mobile's prob solution and day by day it's popularity is being increased.If someone can solve his problem without installing an application which one he will choose?I would choose the easiest way.So web is going to dominate app stores as i think.
Pretty sure mobile doesn't necessarily mean cellular. for example, the ipod touch belongs in the mobile space, and has done well without needing cellular service.
Speaking of predicting the future!!!...
Apple rips off the iPod concept from a British man who invented in 1979!!!!
http://gizmodo.com/5315766/suspiciously-prescient-man-files-patent-for-ipod+like-device-in-1979
Apple rips off the iPod concept from a British man who invented in 1979!!!!
http://gizmodo.com/5315766/suspiciously-prescient-man-files-patent-for-ipod+like-device-in-1979
What a load of crap!
These are the words of a man desperate to catch up to the brilliant model that Apple has created.
Web based apps will obviously become more useful and thus used, but will not in my opinion ever completely replace their stand alone counterparts.
The idea that you need to be connected to the internet to do your work or whatever is simply laughable.
These are the words of a man desperate to catch up to the brilliant model that Apple has created.
Web based apps will obviously become more useful and thus used, but will not in my opinion ever completely replace their stand alone counterparts.
The idea that you need to be connected to the internet to do your work or whatever is simply laughable.
Apple rips off the iPod concept from a British man who invented in 1979!!!!
http://gizmodo.com/5315766/suspiciously-prescient-man-files-patent-for-ipod+like-device-in-1979
And..?
The first computer as made in Greece eons ago. (An abacus for you slow people)
Brilliant model? Um......app store was in play long before apple joined that space.
Did you think Altavista was incompetent at the time? Since the AppStore wasn't the first mobile application store, couldn't you argue it's wild success is because of marketing?
You're in dangerous territory when you start talking about what we need. I didn't need keyless entry on my car until I had it. I didn't need a cell phone until I got one. I didn't need Chrome's V8 JS engine until I tried it. Now I need all three of those just as much as you needed Google.
I haven't a clue what you do on a computer, but the cloud is brilliant and a blessing for me. I happen to like editing a document on my laptop and being able to view it on my mobile device without syncing. I badly want to be able to edit Google Docs/Spreadsheets on my BlackBerry, and I've got every reason to believe this is coming soon.
As an aside, to all in the 'Google owns you' club: I couldn't care less. Their interest in protecting the data I throw on Google Docs is far greater than my own. If they want to see what my answers are for end-of-chapter questions or how close I'm coming to my monthly budget, great. If they want to start betraying users, users will begin to leave in droves...and never return. Profits don't have much of a chance to grow with a diminishing user base.
I guess you're stating no belief in the free market? Either that or a very strong belief in collusion? Please clarify why the prices will go up.
I live in a building that only has two non-wireless internet providers to choose from, but there are at least 4 that are prevalent in my area. I get the feeling they all want my business with the specials they run: 10Mbps downstream for $30/mo for a year isn't too bad; especially considering what I paid in the 90s for 56k. Speeds are capped at 20Mbps with this provider as well. Since fiber optic rollouts are picking up speed, and prices have dropped for faster speeds consistently since the 90's, I see why no one is asking it.
First, the original abacus were not invented in Greece. Several of the civilizations that predate which is typically identified as Greek civilization had them. There are Greek variants, but not the creators.
Second, the somewhat archaic definition of computer would probably be called a calculator today. There are no "programs" on an abacus. It is computation device. It only does math.
In the more modern era before iPod Diamond/Rio , Compaq( sold off concept) , and I think Digital (the R&D labs in California) had players before Apple did. After the SonyWalkman, this was really a step in using the now more appropriate mass market technology available. Apple threw in some incremental refinements on control but the basic concept was all there. ( iTunes also in part, bought off the shelf and not invented internally. Selling stuff online....been to amazon.com among others. Again a control subvariant of putting a mini browser in the app; but web store was already done. )
guys, guess what? the people working at Google are some of the brightest guys in IT, generally they know what they are doing (same goes for MS, IBM, Sun/Oracle & Apple).
If this announcement does not make sense to you, you should rather check my facts than dismissing outright.
enough Google kool-aid for now...
If this announcement does not make sense to you, you should rather check my facts than dismissing outright.
- Remember we are talking about strategy, not of finished products. Thinking on a timescale of 3-5 years, the current "state of the art" is pretty irrelevant, the general movement is important. So, the mere fact that the Apple AppStore is successful today, does not imply it will be as dominant tomorrow. As Google Chrome already is, I would expect Chrome OS to be a PoC mainly aimed at early adopters. Google will show what is possible today and this will shape tomorrow's marketplace. Google is and advertising company, they don't have a direct interest in building operating systems - however, they have a strong interest in increased internet usage.
- "But I don't want my apps to be in the cloud" - several posters have already pointed out the possibilities of offline storage (HTML5, Google Gears). IMHO it is reasonable to assume these technologies will undergo improvement/refinement as they become more widely used.
- "Javascript sucks" - well, IMO it doesn't (it just uses a somewhat different programming paradigma than what most developers are used to - seems like many programmers don't even try to learn it seriously). But look and behold, there are several ways to get around using js directly: ever heard of GWT or Objective-J (which is a superset of js)?
- "Javascript performance sucks". Several points on this one: First: expect js performance to improve further (it already has gone a very long way in the last 1-2 years). With the wide adoption of serious JIT compilers for Javascript, js performance will be able to benefit from years of experience in building JIT compilers (yes, .NET/JAVA code runs pretty much at native speed). Granted, you will experience worse performance than when using those pre-compiled languages, as you have no possibility to distribute js bytecode - otoh your language is very dynamic/flexible. Second - perhaps the task should be done on the server instead? Third, no reason to use js, if you don't like it: in Google's Dream World you would just use Google Native Client. It allows you to execute compiled C++ code more or less directly (very thin "hypervisor") and its security model rocks.
- But wait, even with nativeclient I cannot use my "local" devices (printer, etc.). Remember Google's "failed" 3D socializing thingie called Lively. It just used a platform-specific plugin to make direct use of your local graphics card (via DirectX on Windows). Noone knows how much "security" Google put into the APIs they developed Lively on - but probably they have a working model for using local platform-specific hardware/APIs securely.
- video-tag, 3D effects implemented in CSS, etc. - you see where the journey is going. Adobe will have a hard time once these technologies are fully embraced by the web devs.
- Developers, Developers, Developers... (sry, had to bring this one up). In the end it's developers deciding what platform they want to code for. Write your application just for the iPhone (using a language/framework that is common on no other mobile platform) and profit from the nice Apple GUI and the popularity of the AppStore. Or write once & run everywhere? If Apple messes up future iPhone releases, lots of developers will have a very hard time recovering from the blow - diversification in the platforms your product can/could run on with minimal effort means a lot when designing your business plan (yes, even IT companies have one). Market size will probably more than make up for the polish lost when developing "one-size-fits-all" applications. And frankly, CSS is one of the best and most flexible technologies available to customize your product to different screen-sizes, GUI layouts, ... without changing one line of "real" code. If you read slashdot today, you might have noticed GNOME devs are seriously considering using a CSS-based theme-engine for v3.
enough Google kool-aid for now...
And there you have it - connecting the dots, fusing things together, implementing the conceptual hybrid of pre-existent components into a unique, innovative, smoothly integrated and polished reality - this is what Apple has successfully done, repeatedly, and will likely continue to do, by redefining and refining the user experience, making complicated operations and devices simpler and easier to use. This will only inspire others to keep innovation flowing, all while consumers continue to reap the benefits.
you are confusing apples web apps thinking thats what all web apps for other platforms are going to be like. put down your iphone for one second and maybe try another device for a change so you actually know what you're talking about.
do you even know what a web app is? since when do you have to connect to the internet for every web app? adobe air apps are web apps. you don't need to connect to the internet to use many of them that aren't actually internet based. why do people here talk about stuff they know ABSOLUTELY NOTHING about?
The mobile web and cloud computing
Cloud computing is very attractive for large developers because of the profit potential. Subscriptions to apps in the cloud that replace existing native apps are necessary for these legacy software developers to continue the revenue stream when they have run out of revolutionary software innovations.
Mobile computing will likely take better advantage of networked resources available in the future. This will probably include an offloading of some tasks to hardware resources to enhance computing power and expand storage and improve connectivity and distribution of media among other things. The iPhone and App Store have already started down this road.
Cloud computing in the future will have remote apps competing with local native and hybrid apps that will involve a great deal of local resources such as home computers and televisions. Applications will span many of these different types of resources while pulling remote data as needed. Apps that rely on dependable high bandwidth connections will only grow as fast as the internet infrastructure allows.
It is unlikely that all functions will be moved to remote physical locations due to the inherently poor performance compared to native apps. I believe native apps and hybrid apps that use both remote and local resources will dominate for the near future. Pure cloud computing where everything is remote including the OS will remain a niche market.
Cloud computing is very attractive for large developers because of the profit potential. Subscriptions to apps in the cloud that replace existing native apps are necessary for these legacy software developers to continue the revenue stream when they have run out of revolutionary software innovations.
Mobile computing will likely take better advantage of networked resources available in the future. This will probably include an offloading of some tasks to hardware resources to enhance computing power and expand storage and improve connectivity and distribution of media among other things. The iPhone and App Store have already started down this road.
Cloud computing in the future will have remote apps competing with local native and hybrid apps that will involve a great deal of local resources such as home computers and televisions. Applications will span many of these different types of resources while pulling remote data as needed. Apps that rely on dependable high bandwidth connections will only grow as fast as the internet infrastructure allows.
It is unlikely that all functions will be moved to remote physical locations due to the inherently poor performance compared to native apps. I believe native apps and hybrid apps that use both remote and local resources will dominate for the near future. Pure cloud computing where everything is remote including the OS will remain a niche market.
Just coming in late to this thread, my sentiment is that Web apps like cloud computing will only really catch on when we can get a connection to the web anywhere at anytime. For now I'll stick to native apps and my ability to use them all the time as well as on a device like the iPod touch.
Wirelessly posted (iPhone: Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_0 like Mac OS X; en-us) AppleWebKit/528.18 (KHTML, like Gecko) Version/4.0 Mobile/7A341 Safari/528.16)
This was Apple's opinion at the initial release of the iPhone. I think that things may head that direction eventually but for now nothing can beat having the applications locally. To really make it reasonable for all applications to be on the Web, you'd have to have a much faster connection that what is currently available and be connected 100% of the time, even on airplanes (and some people don't want to pay a fee for wifi on an airplane just to play scrabble). Until this sort of thing can be worked out, local apps are here to stay.
This was Apple's opinion at the initial release of the iPhone. I think that things may head that direction eventually but for now nothing can beat having the applications locally. To really make it reasonable for all applications to be on the Web, you'd have to have a much faster connection that what is currently available and be connected 100% of the time, even on airplanes (and some people don't want to pay a fee for wifi on an airplane just to play scrabble). Until this sort of thing can be worked out, local apps are here to stay.
What for? Every piece of performance being eeked out for Javascript by Low Level Virtual Machines includes performance gains for C/C++/ObjC/Smalltalk/Eiffel/D/Ada/Fortran, etc.
It's DOA. 3D Power requires OpenCL 1.x/OpenGL 3.x heavily paralleled code offloaded to the GPGPUs.
None of that comes from Javascript/HTML5/CSS. It's all C/C++/ObjC driven.
Google is running out of options to justify it's overhyped valuation, period.
But he did say "We believe the web has won..." and made other supporting statements. The message he sends is quite clear - we can compete with the app store directly by using the browser instead. I don't think that its true today, but if you look at my first post, I did say it was likely possible in the future. I don't personally think it can happen "soon", but we shall see. I'm not saying that you can't get anything up and running doing it this way today. Because you can build some very capable apps. But I am saying that for now you still need native apps if you want to cover all of the same markets that Apple does. After all Google does still have an Android SDK that lives outside of the browser, right?
Sure, I've been around long enough to remember all of those debates. You can take any level of abstraction provided by all of the current software stacks and say that that level will eventually be the one we code everything on top of. And you'd be right in some cases. But what it takes to get there is not just technical feasibility resulting from capable hardware and smart compilers/interpreters, but also the existence of proper APIs and services such that you can leverage the full feature set of the hardware. Those don't all exist yet in the browser, at least not in a standards-compliant form. Will they eventually? Yes, probably, but even when you get there it doesn't preclude the app-store model because people aren't going to just stop wanting to make money. So you still need a place to sell apps, even if they are not native apps.
Well yes, that's the hypothesis. It's not here yet, at least not if you are using what you can achieve using the iPhone SDK vs. what can be built using various web technologies as a means for comparison.
I lose service every now and then. If I can't access an App because of that, then what's the point of even having apps...
Google is, once more time, right wether Apple likes or not.
It doesn't matter how much money Apple spends on marketing to promote its app store, the trend is the web, and almost every single application tends to be a web app.
HTML5 is here, CSS3 is there, web browsers render javascript faster and faster... what's the point of downloading an application when you can download it on your favourite device and share or store in some cloud?
I'm afraid (and happy about it) that the web is the new operating system and Apple will soon change to it because it can't fight Google nor it can't fight what, we, the users do want.
It doesn't matter how much money Apple spends on marketing to promote its app store, the trend is the web, and almost every single application tends to be a web app.
HTML5 is here, CSS3 is there, web browsers render javascript faster and faster... what's the point of downloading an application when you can download it on your favourite device and share or store in some cloud?
I'm afraid (and happy about it) that the web is the new operating system
and Apple will soon change to it because it can't fight Google nor it can't fight what, we, the users do want.
Wow I'm rather shocked at how one sided most of the opinions here are on this. Yes, the app store & native apps are miles ahead of web apps at the current time. But just how many of those apps rely on some kind of network connection? Twitter apps, email, youtube, even games that post scores to the web or offer extra content downloads. All of it coming from the web.
Basically as it is right now the "native" apps are acting as an enhanced layer on top of the web as it is. So yes, I think he's right here, down the road web apps will take over from native apps, there is still a lot of growth to happen as far as the technology goes, but it will happen.
As a developer would you rather build an app that works on one phone/platform, or offer one that works across many?
Basically as it is right now the "native" apps are acting as an enhanced layer on top of the web as it is. So yes, I think he's right here, down the road web apps will take over from native apps, there is still a lot of growth to happen as far as the technology goes, but it will happen.
As a developer would you rather build an app that works on one phone/platform, or offer one that works across many?
If WebAPps are the way forward.. why have Google released native apps for the iPhone?
Anyway, seriously, as mentioned previously in this thread, neither WebApps or Native apps are the future - its both. It depends on the best tool for the job.
My preference is for a native app that integrates with a web service - best of both.
Anyway, seriously, as mentioned previously in this thread, neither WebApps or Native apps are the future - its both. It depends on the best tool for the job.
My preference is for a native app that integrates with a web service - best of both.