Slashdot Mirror
Google Chrome: the New Web Platform?
snydeq writes "The Chrome dev team is working toward a vision of Web apps that offers a clean break from traditional websites, writes Fatal Exception's Neil McAllister, in response to Google's new Field Guide for Web Applications. 'When you add it up, it starts to look as though, for all the noise Google makes about Web standards, Chrome is moving further and further apart from competing browsers, just by virtue of its technological advantages. In that sense, maybe Chrome isn't just a Web browser; maybe Chrome itself is the platform — or is becoming one.'"
Embrace, extend, extinguish. Or is there any other way this can be interpreted?
It's pretty obvious Google are trying to make a "web platform" with Dart and NaCL. Most people spend a lot of time using the web, probably most of the time in front of the computer is web-time. When I use the web, which is a lot I want a better experience, I want native speed, I want real apps and games delivered on the web. If Google can give me that, more power to them. So far their technologies is open source, so I see little wrong from their doings. I don't like installing crap on my computer, phone or "pad", if apps can be delivered over the web, all the better.
Seeing the web as a glorified publishing tool as it is today is old school. Google should have WOW ported to NaCL, that would give it a boost.
You know what is ironic?
IE 9 and 10 are fully open standards compliant with XAML the only semi proprietary thing in it for Metro hooks. Oddly I am considering going to IE 10 when it comes out because it is the the most standards compliant, secure, and best browser and I am afraid to get locked into an eco system.
I feel I just walked into the twilight for even saying this! But I feel it is exactly 10 years ago all over again with IE 6 and that fear of the web being closed off. Those were some dark days.
Companies like Apple have proved that once they are in a dominate position can quickly turn into the bad guys. 10 years ago I never would have imagined Apple doing anything as crazy as they are today. Google just might pull it off as IE and FF usage is declining.
http://saveie6.com/
Almost everything in the "Field Guide" is supported across browsers --- which is good and proper, but it's not pointing the way to Chrome-only applications like the Infoworld article suggests.
Apparently in the Infoworld article, Chrome's "technical advantages" are NaCl and Dart (not mentioned in the "Field Guide"). NaCl is bad for the Web in multiple ways. It ties Web apps to specific processor architectures. (PNaCl is going nowhere because LLVM bitcode is not actually architecture-independent.) Worse, it creates a huge new set of Web APIs ("Pepper") for NaCl applications that mostly duplicate the functionality of the standards-based APIs we already have. This is a lot of unnecessary bloat, complexity and attack surface, plus a lot of extra standards work that would have to be done if NaCl were to be come a real cross-browser standard (which it won't, because no other browser vendor has shown interest in implementing it). The performance advantages of NaCl are overrated; C-to-JS compilers like Emscripten are rapidly improving, the JS language and implementations are rapidly improving, and for a lot of modern apps you want to be offloading to the GPU anyway.
Dart is unnecessary and will simply be overtaken by improvements to Javascript.
Apps are not the future. They are the past.
Webapps or just web pages, as we used to call them, are the future of software. You just enter an address or click a link and you get to the most up to date "app". No installation, no updates, no permissions, no specific OS or hardware necessary. It works everywhere by everyone and all the time with no hassles.
The reason apps made a comeback is because you can charge for apps. An app is a defined thing and an installation is a chargeable privilege. So thank Apple and all the me-too followers for burdening us with software deployment and management just as we were about to escape those unnecessary activities.
Apps as platform is not driven by mobile OSes, browsers or other modern technology. It is driven by capitalism.
Google has proprietary CSS and Javascript as "enchancements". Yes webkit is open source but the CSS 3 implementations are not w3c standards. THey propose replacing HTTP with SPDY and already violate RFC implementations of http that can flood routers that are not configured properly in order to make it appear faster. Now everyone is doing it.
To me Chrome feels a lot like IE 5 or 5.5 where cool AJAX was introduced and IE at the time was a great browser that was faster and sleeker. However, proprietarness crept in at those releases just like it is with Chrome.
IE did its work in the corporate market with tie in. Chrome is doing it in the consumer market. Oddly, IE 10 is one of the most standards compliant browsers that is being developed. It is the total opposite of 10 years ago but Chrome will be stuck with many webmasters a decade from now who will wine like they do today with supporting IE 6.
http://saveie6.com/
Gotta agree, native networked apps have some big advantages - fast local processing, local gfx elements, cached local data, richer GUI etc. But they have real disadvantages too - they have to be written for specific architectures & platforms, and consumers have to locate (and update) them through a whole different channel (an app store) than the site they communicate with (which is usually a company website).
But a NaCl (or similar) app could work just as well as a mobile app does. They have all the same advantages (fast, rich local GUI, etc), with the added advantage of being downloaded on-demand directly from the relevant website. And NaCl's disadvantages (platform-specificity and security issues) are no worse than existing native apps.
Heck, for mobile platforms (well, Android) you might as well use standard APKs as your native client, with just a streamlined installer (click a link, show a permissions dialog, then auto d/l+install+start), and bypass the app stores altogether. Next time you click that link, it takes you directly to that respective app (as with Maps and YouTube links), which then displays the content you wanted.
Why would anyone engrave "Elbereth"?
They are both companies. Their natural goal is to make as much money as possible.
This is a horribly dangerous misconception.
The fact is that for MANY companies money is not the primary goal, once they have enough money to "coast".
Larry Ellison. Steve Jobs. Scott McNealy. Bill Gates.
These are all guys who did insane things with the companies they ran, insane at least with the expectation you'd be making money.
No, for the top players it is NOT about making money, it's about the Game. There's not even one Game, each is playing his own and wanting to "win" by whatever internal definition they have.
Until you understand this basic truth you will never understand the computer/technology market, or be able to predict anything about it - and that is VERY dangerous if you have a career that intersects it in any way.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
No one wants to use Dart because it doesn't provide any benefits that couldn't also be provided by extending JavaScript, which is something Mozilla, MS, and Google are all working on. This is why no one else wants to implement it, and Google knew no one else wanted to implement it before it was even announced. I'm not even sure if Google intends for Dart to be used, as opposed to using it to try to push specific features through TC39.
No one wants to use Native Client because it will tie the web to specific CPU architectures. Comparing this to ActiveX is appropriate in some way, because it puts additional restrictions on what devices can access the web. If Native Client had come of age at the same time as JavaScript, real smartphone platforms would be probably still not exist, since websites would require x86 processors to run. Intel has only recently announced x86 chips that can provide decent performance while fitting the power profile of a mobile device, and only after getting their ass kicked for many years as the mobile market has continued to grow.
Alternately, different people with different viewpoints (pro linux, pro microsoft, pro apple, pro richard stallman, etc) will mod the most well written posts that support their viewpoints. Given enough moderators, both methods should end up with roughly the same result.
This only works if moderators only up moderate though. Somewhere a couple years ago CmdrTaco pointed out that something like less than 5% of all moderation points are spent down moderating posts, so I think this works overall. Moderation on Slashdot has been pretty consistently high quality for the last decade, so whatever moderators are doing, it seems to work ok!
moox. for a new generation.