Google is now building their own web browser named Chrome, based on the WebKit rendering engine and a beta version for Windows is currently available. They've put up some great comics that do a great job of introducing the new browser and some videos explaining some of the changes they've made.
I've always found Google applications on my Mac to be a bit out of place. Maybe it's part of the design they are going for, but even today, when I use their online apps, like Gmail, or desktop applications like Google Earth, they work well, but beautiful they are not. They are blocky, text heavy, lowest common denominator type of UIs. It's always felt to me like they didn't quite "get it" when it comes to blending into the platform and learning to be a "good citizen" on the user's platform of choice.
Today, it's clear to me why I've felt this way: Google isn't interested at all in "being a citizen" or part of a platform, they are interested in being the platform. If you look at the way Chrome is designed, it's not so much designed to be a good browser, as much as it is a good operating system for web applications. Google's desire is very much the same as Microsoft's, except abstracted a little higher up the stack. They want to own the platform upon which web applications are built, just like Microsoft wants to own the platform upon which desktop applications are built. This game of disintermediation seems to never end, but this time, what can Microsoft do? Or anyone else for that matter?
This is not to say that Google's success criteria for Chrome is market share. I think what they are trying to do is have a more direct hand in guiding and shaping the web app platform and raising it to a level that best fits their desires and needs. Google will be successful if in the future developers see no downside to developing a web app versus a traditional desktop application, but in-fact see a sizable upside to taking the web app route. For end users success will be when the "Omnibar" becomes the default interaction mechanism, the place they go to first and installers become a thing of the past.
Maybe that's why I think Google's stuff looks kind of basic. They are to the current web platform what command line terminals were to the earlier personal computer platform. The basics, from which great things are built.
Not the first company to do this...looking at you Adobe...
Adobe's been crapping on both Mac and Windows for ages, ignoring UI guidelines on both platforms for the sake of 'uniformity'.
Adobe is an interesting case. I'm not so sure Adobe's app UI design stems from a desire to be a platform like with Google. For Adobe I think it's more about engineering expediency as much as anything else. That said, they have definite platform plays with Flash, Air etc.
"If you look at the way Chrome is designed, it's not so much designed to be a good browser, as much as it is a good operating system for web applications"
No need to divine their intentions from the look: they say as much explicitly.
The comparison of a company's intentions to Microsoft's is very Godwin's Law 2.0. There is a massive difference bewteen creating a platform allowing anyone to write apps that function properly in the browser and trying to lock a platform down for the purposes of selling the tools to do so or extracting licensing fees for use of the platform itself.
Google is doing itself a favour but at no cost to anyone.
If you look at the way Chrome is designed, it's not so much designed to be a good browser, as much as it is a good operating system for web applications. Google's desire is very much the same as Microsoft's, except abstracted a little higher up the stack. They want to own the platform upon which web applications are built, just like Microsoft wants to own the platform upon which desktop applications are built. This game of disintermediation seems to never end, but this time, what can Microsoft do? Or anyone else for that matter?
I don't think I can agree with you here, at least not entirely. You are plain wrong in that Chrome is not designed to be a good browser. Quite frankly having process separation between websites is something that should have been implemented a while ago (not a long while ago, but Apple could have done it with Safari any time in the last year or so). We've had stuff more complicated then HTML (Java, Flash, scripting) for a long time now, and bugs in that stuff can definitely bring down a browser in various ways. There have been band aids, like browsers that maintain state even after a crash (and that should remain), but process separation for an interpreter running third party code is hardly a revolutionary concept.
Additionally, your contention of "wants to own the platform" is harmed by the fact that Google is using the very same, core web standards that every browser except IE use (and even IE slowly seems to be working towards them). While Chrome might be able to provide a superior user experience and thus get people to use the browser, absolutely nothing it runs can't be run the same in any other WebKit based application, or assuming decent coding under any other standard implementation be it Opera or Gecko. That's the great thing about web standards, and why so many of us have been pushing for them for so long! When the basic core is the same, individual groups are free to build whatever innovative stuff they can think of on top.
Clearly, Google has an interest in there being robust support for web applications out there, but for now there is zero evidence that there is anything proprietary involved in their plans for the client side. This is just the opposite. It appears Google is perfectly confident about being able to compete for some fraction of the application market if web support is strong enough, and that's definitely a potential threat to Microsoft in some ways (since Office is such a cash printer). But it's pretty far from there to "own the whole stack." On the contrary, as long as their basic needs are met Google isn't affected by anything at all from hardware straight up to web browser, that's the whole point for them.
Google gives the platform away for free to make money.
Microsoft sells the platform to make money.
Anything Google can do to drive people to their "platform" means analytics for them and advertising revenue as well. Chrome is one of those drivers. Give the browser away, give cell phones away, give their "office type" apps away, anything to drive traffic and revenue.
It wasn't too long ago that someone gave browsers away to destroy another browser company. Turn about is fair play, right?
Interesting times we find ourselves in.
Dad
The cost of web applications becoming the default delivery mechanism or platform for software is to hurt those who build or support predominately desktop based software bound to a specific OS.
In the case of "certain classes," the fact is that some stuff is, for example, very latency or bandwidth restricted. Short of us figuring out a way around the light-speed barrier there simply isn't any way to avoid the requirement for a local application (although some of the data might get loaded remotely). It doesn't matter how fast or reliable the pipe is, there will always be a significant penalty to accessing something off a server 1000 miles away versus off your own local storage pool. So many "pro" apps and stuff like complex video games aren't going anywhere anytime in the foreseeable future. At best they might get loaded remotely, but even that is a stretch.
And even taking the case of very favorable-to-webapp classes, like text editors, restrictions seem to still be there. The obvious one being "what happens when your link drops out," but also stuff like "what about all those useful text/application services that Mac OS X provides for everything Cocoa?" A webapp has value in being more ubiquitously available and always being familiar, but a (well written) native app has value in its integration with other applications and parts of the OS, user control, and so on. So it seems like there will still be plenty of "upscale" markets for richer stuff, it's just that the bottom feeders in particular will all get annihilated. Also, worldwide the market is still expanding tremendously, so Google could seize a base of hundreds of millions of users and still leave vast opportunities for other players.
"The cost of web applications becoming the default ... is to hurt those who build ... software bound to a specific OS"
Why is this 'cost' a concern to anyone except that very small minority of people who actually do--and want to continue to--build OS-bound software?
@Simon. David writes apps for an OS. This is his blog. I think it's an appropriate venue to talk about issues important to "that very small minority of people who actually do--and want to continue to--build OS-bound software". Also worth noting that while web apps get richer and more competative with Windows apps everyday, I as a user would be bummed if I lost rich desktop apps like say... ecto and suddenly was stuck with only the web interface to wordpress--which is still a very nice web app in case anyone thinks I'm knocking it.
@Zanon: I agree with David that Google wants to own the platform, and I don't think that's at all unusual. A lot of companies want that. It's a perfectly reasonable business move, and they're at least trying to do it by making things better.
It's true that another browser could do the same things Chrome does, but I personally took David's post a bit differently. I don't think the main goal is really to see Chrome get substantial marketshare (which David says), though I'm sure that would be nice too.
I think his point is that Google would like the web to be a preferred platform because it has eyes into everything on the web, and has significant influence in that context. It's an attempt to bring the game to the home field. Again, I don't think this is an unreasonable approach, and I think we're all likely to see the benefits (except probably Microsoft).
Every single web app that launches is potentially more revenue and influence for Google in some form -- either directly or indirectly -- whereas they'll probably never see a dime from an internal enterprise Windows app.
I also agree with you that certain things are not a good match for web applications, and are unlikely to be for some time. Video editing or any kind of media-intensive application are clear examples. It's also obvious from the interest in the native iPhone SDK that a platform-specific API has a lot of user experience advantages. But there are many apps that the web-based approach is ideal for.
By the way, David -- you have a fantastic site here. I love the design and the total overhaul since the other blog.
Now that each tab is a process will we ever be able to let each tab run as a different user or in a different group so we can assign different levels of protection? I might trust Google apps to write files to my local disk or change aspects of my configuration, but it might make sense to run less trusted sites as less privileged users. (Yeah, I'm so old fashioned).
I explored why Chrome is not an OS in:
Why Google Chrome is not a "Windows Killer"
http://counternotions.com/2008/09/02/chrome/
and why it poses a great threat to AIR which poses as a supra-platform over multiple OSes in:
Google Chrome: Bad news for Adobe
http://counternotions.com/2008/09/03/badnews/
If you look at the way Chrome is designed, it's not so much designed to be a good browser, as much as it is a good operating system for web applications
I couldn't disagree more strongly. Chrome is not trying to replace the operating system, it's trying to get out of its way. By using a shared-nothing separate process for each tab and plugin, it's letting the Kernel and libc do the job they were designed to do.
Most modern browsers (especially FF3) are trying to do an operating system's job — scheduling logically independent processes, micro-managing memory allocations, mapping virtual memory, providing an internal windowing system, providing an internal GUI scripting system.
Chrome does none of these things. It parries them off to the real operating system, where they belong.
"Today, it's clear to me why I've felt this way: Google isn't interested at all in "being a citizen" or part of a platform, they are interested in being the platform."
Yup... Any business that's not out there to 'take the cake' is going to fail. Does Ford want to be just another auto maker? Of course not, they want to be *the* auto maker. A company that doesn't have lofty goals like that but wants to compete in an international market cannot be taken seriously.
"They want to own the platform upon which web applications are built..."
Close, but not quite. They are giving away the source to the whole thing. They don't care about owning the platform, they just want to ensure that an awesome platform EXISTS for their web apps.
http://www.joelonsoftware.com/articles/StrategyLetterV.html
JavaScript-heavy apps are very demanding. Rather than wait for hardware to get faster or for other people to write better JS engines or spend countless hours optimizing every last side of a web app, they figured the best way to increase web app performance was to make a whole new browser. The fact that it will make ALL JS-heavy pages (i.e., every single major popular site out there other than Craigslist) faster is what will make it popular.
It is entirely possible that Apple and/or Mozilla could wrap their own existing UI around Chromium, or at least use V8 as their JS engine.* Whatever. Google just wants all web apps (including their own, natch) to be an order of magnitude faster than they currently are. A rising tide lifts all boats.
* For that matter, Microsoft could, too, but they won't.
there are so many advantages and features with Chrome, such as it's speed, for example; now if only they would take care it's flighty cookie management...