WebKit was originally a fork of KHTML and then was completely rewritten.
Please remember that WebKit changes have been rolling back into KHTML/KJS for the past 4 years.
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
WebKit was originally a fork of KHTML and then was completely rewritten.
Please remember that WebKit changes have been rolling back into KHTML/KJS for the past 4 years.
If Steve Jobs where the one that said this you sheep would be all over it talking about how Apple was ahead of its time etc,but since it's google....
BAAAAAAAA
Steve Jobs would never have said it. In fact, Steve used to mock Sun at NeXT and Apple when this crap was brought up.
Sure once you push the critical mass of functionality provided by the hardware and OS into the browser, this can happen.
But that is a decade off at least.
In between now and then Apple will not just sit still. They will evolve iPhone OS and App Store to accommodate the Chrome OS/Web OS model. This will certainly be a relatively simple technological step. Contrast that with the huge steps Google, Palm and Nokia must take to get their web based OSes and web applications to achieve the type of functionality and performance we see today from a native application on IPhone OS. HTML5? That addresses a mere fraction of the capabilities of the devices you are running these OSes on. You'll need HTML 6, 7, 8 and 9 (etc.) just to get to the point where we have a cross-platform web-standards-based Doom - nevermind a state-of-the art 3D handheld game, nevermind what a state-of-the-art handheld game will look like in the ten years.
But that is a decade off at least.
In between now and then Apple will not just sit still. They will evolve iPhone OS and App Store to accommodate the Chrome OS/Web OS model. This will certainly be a relatively simple technological step. Contrast that with the huge steps Google, Palm and Nokia must take to get their web based OSes and web applications to achieve the type of functionality and performance we see today from a native application on IPhone OS. HTML5? That addresses a mere fraction of the capabilities of the devices you are running these OSes on. You'll need HTML 6, 7, 8 and 9 (etc.) just to get to the point where we have a cross-platform web-standards-based Doom - nevermind a state-of-the art 3D handheld game, nevermind what a state-of-the-art handheld game will look like in the ten years.
Speaking of ignorant, how many apps are available in those other app stores that have been around for 10 years compared to the iPhone app store?
Uh huh. I could rest my case there on ignorance, but let's go further...
The only thing holding those other stores back are the companies that created them and their lack of vision and their crappy SDKs, the reason Apple has matched or beaten them all already in such a short time.
I would argue the exact opposite that you stated you is true, Google has little choice here, because if any of those app stores had as much as the iTunes app store in them, there would be no need for Google talking up "the cloud!" LOL
And the only reason Apple's iPhone is missing some features some of those phones have is THE CLOUD itself! So many people own iPhones, that ATT's cloud cannot support some of these niche features currently. That's a CLOUD issue, not an Apple issue.
The irony here is that when the iPhone only had web apps, all people did was complain!
LOL
So finally, I'd argue that everything you just said is a load of ignorant crap since the CLOUD is the reason for the vast majority of criticism of the iPhone and is not related to a problem with the app store concept. It's ATT (i.e., the cloud itself), that is preventing tethering and many iPhone features, not Apple's app store. Google is just offering a cheap crappy alternative to the poor companies like Palm who couldn't even write their own iTunes sync app. Too funny!
The very reason Apple is selling so many iPhones is because of the app store, not despite it, and the problems related to web apps, and these are not things that Google has control of nor can easily solve.
You and Google are both living on a cloud somewhere if you believe Google's view of the cloud.
My only question is one that almost no one is asking:
What will happen to ISP rates when everything is online? Do you think ISPs are going to provide double or triple the bandwidth for the same price as now? get real. If people start uploading and downloading at exponentially higher rates because everything is done online, the ISPs will respond with equally higher rates. Nowhere will this be more acutely felt than in America, home of some of the highest internet rates in the world and the slowest speeds.
What will happen to ISP rates when everything is online? Do you think ISPs are going to provide double or triple the bandwidth for the same price as now? get real. If people start uploading and downloading at exponentially higher rates because everything is done online, the ISPs will respond with equally higher rates. Nowhere will this be more acutely felt than in America, home of some of the highest internet rates in the world and the slowest speeds.
I think iphone apps store will be copied and copied...which will make it frustrating for developers to convert all their apps to all the different platforms so i think developers will gravitate towards developing towards their favorite platform or the one with the largest market.
side note...anyone know a good place to share this... i just saw it and thought it might be news...
http://news.cnet.com/8301-13860_3-10286732-56.html?tag=mncol;txt
side note...anyone know a good place to share this... i just saw it and thought it might be news...
http://news.cnet.com/8301-13860_3-10286732-56.html?tag=mncol;txt
What in the world are you talking about? Infrastructure?
The U.S. has the western world's WORST cellular infrastructure, i.e. "the cloud" and that's not going to change very quickly or any time soon.
You can watch live TV in parts of Europe or Japan on your cell phone, but not the U.S.
The bandwidth just isn't there yet and won't be anytime soon.
Web apps were already tried on the iPhone and all people did was complain!
And some of those weren't half bad and STILL people disliked the slowness.
Apple already proved people prefer to own their music and movies over renting via the web.
I predict the same thing will ultimately prove true with web apps.
Everyone keeps trying these subscription or rental web models.
Who's succeeded and made a fortune so far?
Google only has its head in the clouds because none of the other app stores have as many cool apps as Apple's iTunes and those companies have been bested by Apple in a big way because most of them lack the software skills that Apple has or the ease of use.
It's not a bad business strategy though for Google because there is little risk to lose tons of money, since there are tons of apps missing from these other app stores that people might want.
But, it's hardly the future. It's the slow future on the cheap & a last resort for some of these phones which might actually end up failing. But, Google will still make some money, so I don't blame them for trying this. It's easy money.
I'd more likely predict Palm or RIM or both will be bankrupt before Google's Web App Cloud theory ever becomes a reality!
Point taken.
Exactly.
So why have these kind of applications when you can have applications developed in C/C++/ObjC/Java?
Interesting point.
The question is one of robustness and the data requirements for that robustness. In someways it could be compared to purchasing a song and DL'ing it once to your portable device and played from there verses streaming the same song each time that you want to hear it.
While it remains to be seen, I would venture to say that the HTML5 based applications will not have the functionality that a stand alone application will have.
Even Microsoft has already indicated that the web based version of Office 2010 will not have the functionality of the desktop versions. Some folks need the capability of the stand alone applications as they are more full featured.
Personally, I want to be able to turn off all WiFi, 3G, and other wireless connectivity and still be able to use the full features of my apps.
And Apple is the only company that will evolve? The Pre, for example, runs a Linux based OS, has comparable hardware to the 3Gs and has already been hacked to run things like Doom and a PS1 emulator. The current SDK for the Pre doesn't give developers very deep access into the device but Palm has hinted that more access will come down the road.
Lethal
There's something about native apps that I just trust more. I won't trust web apps totally until network coverage is EVERYWHERE.....so uh, maybe by the time I'm 80
People keep saying this.
YOU DO NOT NEED CONNECTIVITY TO HAVE A WEB BASED APPLICATION.
You can have an application that is built on web standards, runs over the web, but also runs locally and synchs up when required.
My guess is also that the software makers are behind the cloud push because it will end piracy almost completely and it will end virtually all of their "hard copy" production costs.
I don't like the idea of web apps. I don't have a fast connection, and there are times I don't even get a connection on my phone. So no. I don't want to use apps OTA. I want a powerful mobile device to run it's own native apps. This might have been useful in the old mobile days, but now we have powerful phones with lots (and increasing) storage space.
since when do direct competitors provide confirmation?
At least partially supporting Gundotra's viewpoint were fellow panelists from Palm and Nokia.
I don't doubt that web apps will _eventually_ be central, but they is _eventually_. I would not bet any time soon. Here we have three panelists who are rivals (one general and two specific) trying to diminish the mightiness of the iphone/app effort. One may like or not like iPhone/App, but it has by any measure been a staggering success. The market was read and the product delivered. Palm and Nokia look pretty silly here considering that they are actively competing with iPhone/App (and losing).
At least partially supporting Gundotra's viewpoint were fellow panelists from Palm and Nokia.
I don't doubt that web apps will _eventually_ be central, but they is _eventually_. I would not bet any time soon. Here we have three panelists who are rivals (one general and two specific) trying to diminish the mightiness of the iphone/app effort. One may like or not like iPhone/App, but it has by any measure been a staggering success. The market was read and the product delivered. Palm and Nokia look pretty silly here considering that they are actively competing with iPhone/App (and losing).
Okay, what you forget, is that some folks expect full functionality with their apps. More than likely this can only be done when connected to the web and will not be able to be done when in stand alone or offline mode.
I hope that makes sense?
Additionally, I am not holding my breath until web based applications will have the same capability as stand alone apps do now. We shall see.
Yeah, and you want to talk about malware on a mac or data theft. Just consider the recent Java vulnerabilities that existed, although Apple should have addressed sooner they were still around. Multiply that times how ever many platforms for web applications are being developed and you can see the concern.
How many apps are available for OS X compared to Windows NT, which have been around for consumers for 10 years? You’re not going to tell me quantity doesn’t matter, but quality, are you?
I think the only thing you wouldn’t be able to access while offline would be your non-downloaded data. See Gmail with Gears.
Sure data is one part of it.
Speaking of which, I am glad that I can clone my internal HD to an external at FW800 speeds. It takes around 3.5 to 4 hours for me to clone about 460GBs (I have a 5400 internal 2.5 inch HD). I would hate to think how long that would take at 100Mbps (FTTH) or 50Mbps (FIOS) or even slower cable and ADSL.
FWIW, with FTTH, assuming that you get a throughput of around 90Mbps due to error correction, handshaking, etc., it would take at least 12 hours to transfer that same amount of data. Yikes! I would hate to see sync'ing times.
As for the capability of HTML5, we'll see what the final result is. My guess is that the HTML5 APIs will not be as extensive, robust or powerful as what can be done with native applications.
Google is right. It's going to take a long time for network speeds to make it possible, but it'll happen.
What we need is an operating system that integrates with "the cloud" seamlessly, so in the future our kids won't know the difference between local and network. Remember the kids at the 80s cafe in Back To The Future 2? "You have to use your hands? This is for babies!" Kind of irrelevant, yeah, but it was a cool movie.
Google has made its intentions obvious. Microsoft will be heading in the same direction. They're constantly trying it. Remember Active Desktop? What a mess!
We'll all have dumb terminals soon, mobile and desktop. Stop posting your ideas, Google already owns you.
What we need is an operating system that integrates with "the cloud" seamlessly, so in the future our kids won't know the difference between local and network. Remember the kids at the 80s cafe in Back To The Future 2? "You have to use your hands? This is for babies!" Kind of irrelevant, yeah, but it was a cool movie.
Google has made its intentions obvious. Microsoft will be heading in the same direction. They're constantly trying it. Remember Active Desktop? What a mess!
We'll all have dumb terminals soon, mobile and desktop. Stop posting your ideas, Google already owns you.
Don't you remember Steve Jobs saying developing web apps was "sweet" back when the iPhone was launched? And this Apple announcement:
Not Gonna Happen: Cached Web Apps are really just installed apps
There are a myriad of reasons this model has failed each time it has been attempted.
Reasons I can readily think of:
(1) Standards: Unless one company controls the implementation of all web browsers, all the implementation will vary. It's irrelevant if a specific implementation can pass the compliance tests. Developers will always end up exploiting some feature that certain implementations do much better than others. Maybe Google is planning such that Chrome will be the only browser, but I seriously doubt this will occur.
(2) The cloud is a joke: The cloud is like all shared services from public transportation to public bathrooms. The cloud won't always flush, some one will take a dump on the ring, the train will break and all traffic will have to go on the bus. The bus driver will wake up tomorrow morning with a hangover, and you'll be late because they had to call in a substitute or replacement for his route. Do you really want to trust your personal info & the viability of your business in the Cloud?
(3) Locally Cached Web Apps are locally installed apps if they can function without connectivity. There can be no difference, if the app can provide its functionality without the presence of the cloud. Sure, it may sync up with home base when connectivity returns, but if the app is to have all of its functionality, it must exist in its entirety on the local device, and you have once again proven that locally installed apps are a necessity.
(4) Innovation: there will always be a need for a manufacturer, to get ahead if their competition, to have more advanced features for his device than the next guy. To keep development secret, that means he has to develop his extensions for the "standard" in secret. Which now means, for the new feature, there is no "standard" except what he has written. Attempting to say that the cloud is the end all/be all, automatically implies that all the apps we will ever need, are already written, and we all know the truth about that one.
(5) Write once, run everywhere: this is just for those among us who can't see the obvious: write once, run everywhere is basically what I mentioned as #1 above, is impossible, unless you own all the implementations of the hosting software (the web browser). Ie., it's a pipe dream. It can actually be idealized and implemented, but the implementation layers end up so far removed from the base device, that you are burning gobs of performance to be "compatible", and you are cornering yourself in your feature set. There will always be the need to run directly on the hardware, and a need to implement new features. Let's call "write once, run everywhere" DOA. Not to mention, canned GUI design usually results in a universally dumb implementation, even using native lower level APIs of specific machines, and will overlook a devices interface gestalt vs. another's model (ie., touch vs. mouse).
(6) Google is just participating in the FUD wars. If it was really the way of the future, it would be happening that way right now, and each step they made would be furthering that goal --- it wouldn't need hammering into our heads day in and day out by high level Google executives.
There are a myriad of reasons this model has failed each time it has been attempted.
Reasons I can readily think of:
(1) Standards: Unless one company controls the implementation of all web browsers, all the implementation will vary. It's irrelevant if a specific implementation can pass the compliance tests. Developers will always end up exploiting some feature that certain implementations do much better than others. Maybe Google is planning such that Chrome will be the only browser, but I seriously doubt this will occur.
(2) The cloud is a joke: The cloud is like all shared services from public transportation to public bathrooms. The cloud won't always flush, some one will take a dump on the ring, the train will break and all traffic will have to go on the bus. The bus driver will wake up tomorrow morning with a hangover, and you'll be late because they had to call in a substitute or replacement for his route. Do you really want to trust your personal info & the viability of your business in the Cloud?
(3) Locally Cached Web Apps are locally installed apps if they can function without connectivity. There can be no difference, if the app can provide its functionality without the presence of the cloud. Sure, it may sync up with home base when connectivity returns, but if the app is to have all of its functionality, it must exist in its entirety on the local device, and you have once again proven that locally installed apps are a necessity.
(4) Innovation: there will always be a need for a manufacturer, to get ahead if their competition, to have more advanced features for his device than the next guy. To keep development secret, that means he has to develop his extensions for the "standard" in secret. Which now means, for the new feature, there is no "standard" except what he has written. Attempting to say that the cloud is the end all/be all, automatically implies that all the apps we will ever need, are already written, and we all know the truth about that one.
(5) Write once, run everywhere: this is just for those among us who can't see the obvious: write once, run everywhere is basically what I mentioned as #1 above, is impossible, unless you own all the implementations of the hosting software (the web browser). Ie., it's a pipe dream. It can actually be idealized and implemented, but the implementation layers end up so far removed from the base device, that you are burning gobs of performance to be "compatible", and you are cornering yourself in your feature set. There will always be the need to run directly on the hardware, and a need to implement new features. Let's call "write once, run everywhere" DOA. Not to mention, canned GUI design usually results in a universally dumb implementation, even using native lower level APIs of specific machines, and will overlook a devices interface gestalt vs. another's model (ie., touch vs. mouse).
(6) Google is just participating in the FUD wars. If it was really the way of the future, it would be happening that way right now, and each step they made would be furthering that goal --- it wouldn't need hammering into our heads day in and day out by high level Google executives.
Ok, sure. But giving access to the device-dependent SDK is in direct contrast to what the Google representative was talking about. That is exactly the point here. No one is arguing that the Pre (or Android device, or any modern device) can compete on the hardware level. But you are not going to be able to showcase that functionality by sticking to web-based development. You need native apps to get the full potential of the hardware.
I dont want an app that only lives up to its full potential when it has web access, thats what I meant. It feels half assed to me
MBP15C2D4GBLED, Yours is a well reasoned post. What you say above is indeed what Google is thinking of with Google gears.
The next question to ask is this.
For example, I have a USB compass. What does it take Google Gears to work with the compass? Developing a specific gear to access the compass is the first level answer. Fine. Now it is looking more like an installed app but it is still a cached gear, downloaded and cached on demand. You can take the Compass to your library computer and the surrounding app, in theory, should continue to work. That is the promise of Cloud computing.
Now, Google Gears is not a standard. It is Google proprietary. In fact, one needs Google Gears if you want to go beyond the standard. For example, in Google Wave, coming end of this year, there is one or two features that are not supported by HTML5. Google pushed that to google gears. Once it gets in the standard, it can be removed from Google gears.
As you say, so far any attempts at having an 'abstract common layer' has resulted in bloat, dumbed down applications and pretty much useless. But let us see if Google and other companies design this with proper separation of concerts and a good software architecture. I have not figured this out yet, but I think that new client devices can be supported with stuff like google gears without having to create the monstrous common abstract api. But there is a challenge about how to expose the data and events of such new devices to the application. It can be done using the DOM model that is in existence today but it has to be done on a carefully chosen architecture.
One major thorn on the side of Browser to Server model of computing is, any asynchronous and unsolicited events generated on the server side need to be communicated to the browser, without the browser asking for that info. ( constant polling ). For example, announcing to the broswer that there is a phone call coming. Here the model has to be reversed. There has to be a persistent TCP communication link with the browser as the server and the 'server' as the client. Streaming embedded media software like Flash etc. essentially fill that role. But setting up point to point TCP connections with a lot of servers in the world is not practical. So the internet needs a Public Mobile me type of service where the browsers/clients register with them and have a persistent connection to it and the servers just post to that intermediate agent. Once the browser is woken up by this intermediate agent, then it can go back to the server and establish a connection.
This kind of public internet wide publish subscribe architecture needs to be in place to really push this new architecture forward. It should be along the same lines as the various Registrars and Name servers ( DNS ) that are around. It can be funded by all the ISPs and major internet companies.
Hi, I am not claiming that it can't be done, but I what I do wonder is how "common" it will be... all the companies involved wth Google want to still differentiate themselves. This is where the old phrase in my parts come from: "Just like trying to herd cats."
I have a feeling that the end result will still be super-dumbed down apps with clunky interfaces, and the really cool stuff will be "packaged". Of course, it can be distributed and managed from the cloud, if necessary, but you will still have apps better suited to say, a Pre rather than a Android device, or iphone vs. a Nokia product, etc. Or say, better for a super-hot SLI gaming box, vs. running it on your PDA/iPhone.
Usually when a feature can work ubiquitously across everything, it's not a new feature. It's just a feature that has been around the block enough times that everyone can do a stand up job of writing a version of it
When I think of the google model, I think of wxwidget apps. They look pretty much the same on everything, but that doesn't mean they look good...