It may come as a surprise that Google has never before attempted to write a browser, but it certainly was a surprise this week when Google offered a new browser called Chrome as a beta. Google gives the impression that by talking about browsers it is actually talking about Microsoft, and specifically the shortcomings of Internet Explorer, but without ever actually mentioning it.
And it says that the reason Google has entered the browser market now, with a free downloadable piece of software, is because when most browsers were designed they were being used to view web pages, whereas now they are used to watch and upload videos, chat with other people, read email and play games ‘ in other words to run applications.
The logic then is fairly simple. It is the underlying architecture of browsers which is at fault, and that no amount of tinkering with an existing browser is going to solve all the problems on the internet ‘ a fresh start is far more likely to absorb all the changes that have happened in browser usage.
Of course Google has perhaps started to fear that its entire future wealth is based on advertising and video which is delivered to a browser, and in all instances it not only does not control that browser, but also no software below that, such as the operating system. It is relying entirely on software written by other companies,’ mostly companies that wish it would go away.
In the past Microsoft has been able to insert an inferior piece of software between the PC user and the browser or operating system, and at least make users try out its various offerings such as media player, DRM or office applications, before some customers discard these and try their own applications. What can Google build out towards from a Google browser platform?
Might it offer mail, video and chat applications to all browser users, just as if it was Microsoft? Maybe it can attempt to get between the user and many Microsoft apps which they never realized there were alternatives to. And perhaps this is the start of shifting many of its own applications from being entirely web resident, to being partially and optionally PC resident. For that having a browser on the device may be a big help.
Perhaps then Google might go further and write downwards to the hardware, and create its own operating system, something that Netscape was always expected to do, which is why Microsoft went and crushed it with a zero priced browser.
But the truth is that Google doesn’t want the hassle of owning and updating an operating system and the browser is perhaps as low down the application stack that it needs to go, because increasingly (Google argues) browser based apps make up more and more of what PC and handset users use. It might also make it possible for Google to add things like video accelerators, and browser plug ins.
So what exactly can Google hope to achieve? Parity with Internet Explorer? Market leadership? We doubt either of those two are possible in the short term but we are humbled by what Adobe achieved in online video formats over Windows Media formats, with Flash Video, taking a market lead in 18 months. The key to that was having an application with an updater on the device, which could prompt and remind people to upgrade and adopt the format, and it looks as if Google has that on a decent percentage of devices ‘ even if the application is just the Google Search bar.
The other critical factor will be whether or not this is something that developers can get behind. If applications which sit on top of the browser, or plug into, it are easy to develop and deploy, and if there is a revenue stream to be had from them, then quite rapidly Google might create a genuine eco-system of browser based innovation and attract developers ‘ but we think that’s unlikely.
But such is the power of Google as a brand, that people are going to prefer this browser, quite simply because it will take a certain amount of grief out of the browsing process.
Google says that the modern requirements for a browser means that a new architecture is required. Browers needs to be more stable, because you may be reading and sending email, and it needs to load pages faster, which Google says implies a lot more use of Javascript, and the browser needs to be more secure, with a simpler interface.
One problem is that browsers execute applications in single threads, so Google wants to change this because at present, if say a Javascript application gets locked, so does the browser. And sometimes that can lock your whole machine. Google considered making its browser multi-threaded, but instead went with creating each browsing tab as an entirely new process, with its own memory, so it is more like a modern operation system working in a partition, than a multi-threaded application. All of the tabs will operate independently, so if one process freezes, the others are all still working.
When something goes wrong with one browsing tab, then the rest of the browser and all the other applications can still function. So instead of one chunk of memory, with multiple pages, operating as one process, the entire thing is duplicated. Google also says by sandboxing the tab then any malware or phishing activity is not only stopped from spreading, but it can be closed by simply closing one browser tab.
Google will also provide tools for looking at which processes are hogging all the memory, and again allow selective closing of browser tabs. The browser has already been tested using Google robots, loading millions of pages, and it has been tried on all web pages that are high up its own rankings.
The rendering engine within Chrome is the same WebKit which is the open source version used in Android, and in Apple’s Safari and on of one of Nokia’s offerings, which Google says it chose because it uses memory efficiently and because it is easily adapted to embedded devices.
This seems to indicate a cooling of Google’s once close relationship with open source project Mozilla, on which Firefox, the most credible open source rival to Internet Explorer (and now going mobile), is built. As with its go-it-alone approach to a mobile Linux platform and to mobile Java, Google is looking to set standards rather than support existing efforts, and appears happy to split the market.
All of which implies that Chrome is naturally destined to browse on the handset, as much as the PC. If that’s the case, this is a major bid for Google owning a piece of the ‘home page’ market on portable devices, devices which run its own operating environment. We suspect that this move is as much about that, as Google genuinely wanting to evade Microsoft influence on the PC.
Google says that it has added a Javascript virtual machine for safety and program independence, built for larger programs than Javascript is usually associated with and it has added stuff like hidden classes, so that anything within the same class is treated the same but without anyone telling the browser what classes of objects it is dealing with. It works it out for itself.
Each tab with have its own controls and URL box, and Chrome will offer suggestions to go to pages you have visited before, but only things you have explicitly typed before, so there will be no long crazy URLs that clog your URL bar which have come from some strange databases with auto-generate URLs.
Any pop ups that a browsing session initiates will be attached only to the tab they popped up in and will be confined there says Google and if a popped up tab has something you want in it, you can drag it outside the tab and it will create its own fresh tab.
Google reckons that Malware and phishing are stopped because it has stripped away all of the rights of any tabbed browsers and put these browsers in a sandbox. Breaking out of that sandbox will become every hackers dream we imagine. Tabs cannot write files to hard drives, cannot read files from sensitive areas of a PC, and the tab is prevented from telling Windows to run a new piece of executable code in the start up menu. All good things, if true and unbreakable.
Google confesses to have a saying which is ‘launch early and iterate,’ which roughly translated means let the public test your products for you and effectively help you build them. We downloaded the browser and made it crash twice, so Google we’re glad that we were able to help.
Google-ists (staff and aficionados) spend a lot of time working inside their browser, but perhaps the real world, or at least the enterprise real world, spends time hopping from application to application (mostly Microsoft applications) which run on the PC and where the data is safely stored on their own device. If, in a future world, more people spend time in their browser, Chrome is going to be a natural place for them to be, if these architectural descriptions work out, and if the bugs can be pushed out of what must be an inordinately complex piece of software.
We worry about all those banking sites built around IE extensions only, which makes running a business or buying things on the web almost impossible with any other browser than IE. Will Microsoft be powerful enough to shut Chrome out of the market? We would expect not, but it will be a long hard road to browser respectability.
Today Microsoft only has 70% plus of the browser market, where once it had a significantly higher market share post-Netscape, with Firefox now holding the next largest share. Google has clearly looked at this, seen that a viable market share CAN be created, because it has helped Mozilla achieve this, and now it wants that share for its own front window to display its other wares. The likelihood is that the net result of what is a commercial move will be that people get a better browser (eventually) and that our own digital media community has a new potential partner it can leverage to get space on the desktop and on future portable devices.