The electric scooter is still the height of Silicon Valley chic. These boys' toys litter the pathways that snake between futuristic buildings at Google's world headquarters. The company's elite band of software engineers use them to zip from the lab to lunch, or from the conference hall to the volleyball court.
The billionaire founders of the company, Larry Page and Sergey Brin, strived hard to cultivate a fun and liberating atmosphere, part university campus and part kindergarten playground, splashed with Google's primary colours. It is a creative hothouse, where world-changing new ideas are baked in the California sun.
The question exercising them now, though, is whether it has all gone a little too far. Are the engineers out of control?
Larry Page says it is the curse of the growing company. "Every time we double in size, the company completely changes - and we've done that a lot of times now. We are also pretty disorganised; that's part of our culture. We are a little over the edge right now in terms of our disorganisation but in six months we will probably have fixed it."
Google has always been fiercely defensive of its concepts of "20 per cent time", where engineers are encouraged to spend a day a week away from their formal projects to follow brainwaves of their own. And across the company, it shoots for a 70-20-10 split of engineering resources, where the 70 is focused on the software behind search results and the 20 is work on related projects such as Google maps. That leaves 10 per cent of resources devoted to what? At the height of dot.commery, it would have been called "thinking outside the box".
Only things have got out of kilter. An audit about a month ago discovered that work on wild new projects squeezed the core effort of improving search results down to barely 60 per cent. It is time to put some engineers back in their box.
There will still be room for creative chaos, says Eric Schmidt, the chief executive, but there needs to be a root-and-branch look at the management structure of the company. "You have to get a balance," he says. "If you have too much management you don't have innovation, and if you have too little you get a little craziness sometimes."
More than 100 engineers will be moved to new projects over the next few months and Mr Schmidt plans to introduce scientific measurement of the precise productivity of Google's engineers. This is a company based on complex mathematical algorithms, after all.
Mr Schmidt, a computer scientist and one-time chief technology officer at Sun Microsystems, is the only one of the decision-making triumvirate with previous executive experience. He will impose the new management control. Mr Brin and Mr Page, with the titles of president, focus these days on the new products and technology.
Mr Page sees no reason to upset this delicately balanced triumvirate, which was formed in 2001 when Google's venture capital backers forced the founding pair to accept a chief executive to keep an eye on the bottom line.
Mr Page says: "If you look at the tech companies that are very successful, most have pretty stable managements, usually a visionary leadership led by the founders or people who came in close to the start. I don't think that is an accident. It is hard to predict what will happen but we're very happy with the situation now."
It is difficult to credit that Google, a company with net profits of $2bn (£1bn) last year and a market value of $120bn, did not even exist a decade ago. Back then, Mr Page was still toiling in the Bill Gates computer science building at Stanford University, trying to download the entire worldwide web in the hope of cataloguing it and testing his new mathematical formulae for searching the contents.
His own day has changed a lot as the company has expanded. "I use a lot of e-mail," Mr Page says. "We still do our Friday executive meeting but the room we used to do it in got too small. Now we videocast it for people in other time zones and sometimes the executives travel. But I still prefer to manage by just wandering around meeting people."
Mr Schmidt's plan to tighten management control will be music to ears on Wall Street, where investors have a stormy relationship with Google. Since the company's unconventional flotation in August 2004, it's shares have swung about wildly. They are now almost one-fifth below their peak at the start of this year after a string of recent upsets. These include the accidental publication of internal profit guesstimates, downbeat comments from the finance director that Google's exponential growth cannot last for ever and a whopping great miss on the profit line of January's quarterly results. And behind it all, the nagging feeling that, while advertising revenues are cascading into the company at a faster rate than expected, Google's spending is also rising much more steeply than anyone foresaw - with nothing very obvious to show for it.
It is in this context that Messrs Page, Brin and Schmidt promise to be more open about Google's future plans, although they continue to infuriate analysts by refusing to give formal profit forecasts.
Google's engineers are working on so many big or whacky ideas - from free wireless internet to architectural software that allows users of Google Earth to draw their own fictional buildings - that it is hard for investors to see where the company is heading. Before, Google might have said: "Neither can we." Now it aims to be a little more helpful about which product it is prioritising.
If there was a theme to Google's annual "teach-in" for the world's press here this week, it was that the company is trying to re-emphasise "search". It is not trying to be the new Microsoft, it is not turning itself into a telecoms company, it will not create its own content for the Web. Of course it will bump up against Microsoft and Yahoo! and others. Increasingly so. A new suit of software products for personal computers which includes calendars and a media player certainly takes it deeper on to Microsoft turf. But at the end of the day, Google is just trying to help everyone find stuff.
Mr Page talks most effusively about the innovations in search. These include the new Google Co-Op, where users can rate the quality of information on the website or add their own favourite sites to help Google skew search results towards their known personal interests. There is also a constant battle to stay ahead of rogue webmasters who try to push their sites selling Viagra or pornography further up Google's search results. Many former Google employees have defected to this dark side of the Web but Mr Page says Google is not in a defensive mode.
This refocus on search has a new urgency. Google's secret internal analysis is worrying - it no longer has as vast a lead over rivals when it comes to the relevance of search results. Google captured millions of dollars of advertising revenue, almost by accident, because it has invented the world's best-ever search engine. Now Yahoo! and Microsoft are hungry and they're coming after it. To stay ahead, Google will have to invent new and better ways to search the Web for words, pictures and for video.
If it can continue to lead the pack, the possibilities are limitless, Mr Schmidt said. "More search, more users, more advertisers, more innovators. We don't see today any limits to the growth of this model."
More than ever, Google needs its engineers to be as creative as possible, so that means keeping the atmosphere in Mountain View free and fun. It means a strong defence of cherished institutions, such as the staff restaurant which overflows with choice and where all the meals are free for employees. Mr Page pours himself a mango lassi. "I mean, we pay them too, right? It's not that important to be collecting money from your employees."
Actually, Mr Schmidt has a great test. "Is Google a place where people want to work? Computer scientists fresh out of college and full of excitement? If the answer is ever not 'yes', then we have done something wrong."Reuse content