There’s More Than a Browser War

When Google released its new (and first) browser a few days ago (Chrome), many praised that move or welcomed this new player into the arena, but many others simply were a bit surprised and wondered if a new browser was really needed when this market already features IE, Firefox, Opera, Safari, Konqueror and a plethora of smaller ones. While IE is still leader, those who aren’t satisfied with it have a good choice of alternatives, especially in Firefox and Opera. So fasten your seat belt to join me in a ride which will attempt to explain why this browser war could be a threat to Google’s very foundation and why Chrome is maybe the most important move Google could have done to protect itself.

A (quick) dissection of bigG
Given all news we hear, you might think Google does and IS many things, but this is not the bare truth. Google’s cash cow is essentially its search engine and, more specifically, its “AdWords” advertising division. That’s where Google gets most of its money, the same way Microsoft has its cash cows in Windows and Office. But unlike Microsoft, Google has no ways to channel itself and its products to users in a direct way. It’s all about typing that “www.” magical sequence in your browser: when you don’t, you’re out of Google’s reach. That’s important, even in browser wars.

might say that I’m crazy: Google does a LOT of things as we hear in the news everyday. But that’s not completely true. Google is ACTUALLY investing into many things, but as any of us can see, its revenues from all these investments are limited. True, bigG has a lot of things in place (GMail, Google Maps, Google Docs – or whatever it’s named, YouTube, Analytics… you name it!), invests in green energy, builds fiber and wi-fi networks, etc. True. But let’s tell ourselves the bare truth: none of those really brings much revenue to Google. In fact, most of those services (including, according to what I know, the wi-fi networks they built or they’re building) aren’t meant for users to pay to use them. A few of them (GMail, YouTube) can be linked to their advertising division but to be honest it’s only GMail itself that Google could actually think to sell users. It might get some cash from Government projects (for example, building networks or renting them) but that’s only a tiny part of their revenues.

If you think I’m just misinformed, let’s look at Google 2008 Q1 financial summary they publish at this address:

  • total revenues for 2008 Q1 were $5.19 billions.
  • Google-owned websites represent 66% of such revenues;
  • Google network (AdSense) represents 33% of total revenues;

The sum is easy: 99% of Google revenues come from either advertising on their own websites or from other sites running ads from Google’s network. There are no significant revenues from building wi-fi networks or other fancy stuff we read about everyday.

Many people are talking about using YouTube to channel advertising (and I mean, inside videos, not textual one) but honestly this is mostly buzz. Years after its acquisition, Google is still not able to channel advertising inside videos and I doubt it will ever be able. Sometimes some friend (of Google, not mine, of course) journalist writes an article hinting that Google could use YouTube to sell advertising but there’s no clear sign of it and for a reason: YouTube is not streaming something you would absolutely want to see but just something that you MIGHT want to see. That is, it’s not streaming EXCLUSIVE contents, something you can see on YT only, so it’s very unlikely that you could swallow advertising in order to see contents. If there was advertising, those little kids would bring their cellphone movies somewhere else, where they could just show them to their friends without the need to learn why you would need to buy a new Ford.

Other key service for Google is Maps. As you might know, Maps is free unless you perform an insane number of requests per month. Free like in beer. Sure, that’s a GREAT service (and I mean GREAT, though Microsoft’s LiveMaps is very good too) but do you think Google could sell it? If, all of sudden, Google started to ask money to put those maps on your website, in most cases users would simply remove them. The great bulk of Maps-enabled websites are just pages showing where webmaster is now, though there are a few of very hi-end websites using Maps.

And what about green energy or renewable resources? Sun or wind? What about fiber or wi-fi networks? What about it’s massive scanning of ancient books to build a digital bibliotheque? What about subway paper advertising projects? Sometimes you hear something about Google and then… puff… nothing happens for months or years.

I will tell you a secret: reason why Google is not charging for all those services and it’s investing into basically anything is bigG knows a way to make money out of them without the need for you to pay. That secret is called “stock price”.

Enter the world of “stock price”
I’ll keep this short: do you feel sometimes confused about what you hear about Google? Me too. And everyone is. Sometimes Google is set to go to space, sometimes it’s going down into deep oceans, sometimes has satellites, sometimes is building something, sometimes is bulldozing something to build something else, sometimes it’s scanning all books on Earth, sometimes they just say they’re making a lot of money.

It’s a mess. And that’s how it’s meant to be.

For all public companies there’s a value which is more “valuable” than all others: stock price. As long the price of your stocks is going up, everything is fine. You might invest billions into making a better ice to sell penguins, that’s not important as long as that number is going up. And that number is going up only when people think you’re about to do something which could be really “Wow!” and they want be part of it to make money out of it as you will probably do. So people will buy your stocks, stock price will go up and actual value of your company will be bigger. That means that your company will have more money to do what it wants and you will be, essentially, richer.

By announcing something you will eventually do, people will see that on you’re on the move and will be eager to buy your stocks not to miss a chance to make some money. And, as the old saying goes, the less analysts understand about what you’re doing, the more they will recommend their customers to buy your stocks. It’s the Enron way.

Be aware I’m not stating that Google is going bankrupt. Refrain from accusing me of that nonsense. But the scheme is the same for all public companies, everywhere in the World.

In a summary, I’m stating that the core business of Google is their advertising division. The bulk of its other activities is to keep stock price up and running and a very few of them are real investments which bigG thinks it could get real money from. And, while this is not public, of course Google has a financial division whose task is to invest money into financial instruments to get revenues from. Remember about this as it will be important later.

The only thing Google is not talking about is its search engines and how they can make MORE money out of it. I wonder…

The “website-as-application” paradigm
I was a bit surprised when the comics used to launch Chrome used the “website-as-application” paradigm to justify Chrome itself. I was surprised because that’s the heart of my article too and I found that funny.

As a summary, Google stated that websites are often not a collection of pages, but most likely they represent an application. I agree. Today a smaller part of websites are simply collections of HTML markup. Most of them, at least most of those developed after 2000, can be better considered applications as they usually perform complex operations that pure HTML pages wouldn’t be able to do, most of them are connected to databases and/or other external services. So I agree about this, I agree 100%.

The thing here is if HTML/Javascript is the best way to create such applications or if there are other technologies which would be more suitable to be used for that. And if there are, why is Google heading its own way?

To understand my point of view, you have to remember what Google really is: a collection of services for business and individuals, all of them served via its magical “www.”. When you don’t visit Google, bigG is dead like a dodo. It’s simply not part of your disconnected life, unlike many of its competitors. When you’re on, Google is there and shines. When you’re off…

This is the heart of Google’s bet, whose status we will examine later.

To go back to web applications (or like many call them, RIAs, Rich Internet Applications), even the newbie knows that HTML+Javascript is definitely NOT the best way to implement them. Rather, it’s the worst way to do it, but there’s a very important reason why developers keep getting headaches from HTML instead of moving.

HTML is just markup and everything else which turned it into a foundation for RIAs can be considered an hack to its original goal. HTML is so popular because it’s historically been so simple to implement but RIAs today have nothing simple in them, nor is it handy to keep hacking HTML structure in order to support new concepts. The result is what we have in front of our eyes: loads of painfully slow javascript-based applications which require petabytes of memory to do almost anything. How many times were we forced to close our tabs because there’s a damn website eating up all my memory, or all my CPU or both of them? How many times you have to wait for tens of seconds to see anything on a webpage just because those scripts are sucking the hell out of your dual core machine? It’s simply inconceivable that machines which can handle millions of polygons when playing your favorite game are then unable to cope with a couple of javascript-based web pages.

That might seem a good go for Google because they used the same arguments to promote Chrome but the bare fact here is HTML+Javascript is not a good way to code RIAs, nor what we will be using in near future. That platform is slow, prone to errors, hacks and exploitation; it’s not secure enough, it’s not safe and fault-tolerant enough, it’s not able to cope with innovation (if it were, we would be using HTML to watch our favorite YouTube movies and instead we’re using Flash…), it’s not able to evolve fast enough (ditto), it’s not able to provide a complex infrastructure modern RIAs need, whatever extension you think and whatever script you load. Plus, the more complex you want it to be, the more painfully slow it will become. That platform wasn’t simply meant to do that and, as it’s clear, it won’t be able to last long. And no, Google won’t be able to native-compile Javascript and make it blazing fast even if they claim that. There are technical reasons for that and the only way to do that is to switch to a run-time based environment, the same way Flash, Silverlight and Java do (and that’s not actually “native” yet as they run VMs).

As it’s obvious, we have better and modern platforms which have been designed and developed with modern RIAs on mind. We basically have three of them now: Flash (the most popular), Silverlight and Java. It’s easy to understand that Google could develop its own technology for that but they didn’t.

And, by releasing Chrome, they promise to make HTML+JavaScript a platform suitable for modern RIAs, to make it blazing fast while still reducing resources it uses, to make it secure, to keep it simple (even when many webpages today loads hundreds and even thousands of lines of JS-based code). bigG is basically trying to defend JavaScript against anything and anyone and, while to common people their promise seems impossible to keep up, they’re just asking for an act of faith: trust them against all odds.

And to emphasize that the key problem is JavaScript, Chrome didn’t even care to implement a new rendering engine since they took an open-source one which is used by other browsers (WebKit). Rendering is not the real problem: most efforts will be dedicated to JS and to improve its performance. By doing this, Google is just sending out this message: “Keep coding JS because we will make JS blazing fast and secure and able to cope with everything you need. Keep investing into JS.”.

It’s just that: they want to defend JS. But why is that?

Google’s bet
Since Google was founded, a few years before the first dotcom crash, its bet was about the expansion of the Internet. For years, Google has been considered the successful model of a Internet-only company, based on extremely innovative services and away from traditional model of “selling a tangible product”. It has been considered the successful model of dotcom and of so-called New Economy.

At that time, the Internet and companies related to it were growing at the speed of light and many predicted that in a few years the network would have been a necessity just like connection to power is. Everyone was going to be connected to a very very fast Internet backbone, probably by spreading fiber connection almost everywhere, especially in Western countries. It looked so sure that Google wasn’t the only company betting on that (remember Sun’s “The network is the computer”?). Pervasive and fast connections: that was the key to Internet-only companies. The bet was you could be like TV: basically reaching anyone 24hrs per day.

Many predicted the end of traditional computers which could have been replaced by dumb (or almost dumb) terminals, whose only task was to connect to the Internet when turned on and download a basic OS which should have provided basic functionalities. Everything else users could need would come from the Internet. Many predicted end of traditional OSes and, of course, many predicted end of Windows era (how many times people and analysts predicted that Microsoft era was going to end?), because the OS needed by those terminals was going to be much lighter and, most of all, it would be possible to download it from the ‘net once the terminal was on. Guess what?

Today, after knowing what happened, it’s easy to say that bet didn’t materialize (yet). We still have Microsoft; we still use Windows. In essence, it’s not that Windows proved to be superior to anything else, it’s just that bet didn’t keep up. Fiber-based networks, after initial development, faced huge costs and while Google tried to cooperate and also sponsored some projects to build such networks (I believe they also built themselves a fiber network in the US), it soon became clear that building those networks was going to cost huge money even in metropolitan areas and even in biggest cities, let alone building them in less populated areas. Not to mention how incredibly SLOW was to pose such cables. Today fiber projects are still alive but they represent a tiny part of the whole and fibers are usually considered good for small tracks in huge metropolitan areas or to secure connections among Governments and institutional places.

xDSL connection are good but those also face a problem in less populated areas. Moreover, those connection face a lower bandwitdh peak and they tend to saturate very quickly. They’re still considered one of the best way to get connected but they’re not considered a way to keep up the “pervasive Internet” bet.

Wi-Fi and other wireless are spreading fast but while they reach good speeds (in Europe, they’re starting to sell 7.2mbit/s connections via HDSPA protocol), radio signals aren’t dependable as connection tends to drop where there are many interferences. Besides, there’s still a big question mark about such technology as studies are starting to hint they could be dangerous for people’s health and even bring cancer in some cases. While this is not proved yet, many are focusing on such researches which could affect mobile phones as well. Anyway, it’s proved that mobile phones can cause headaches and other problems to some people.

The reason why I talked about Internet situation is that this directly affects companies like Google, whose bet is still unfulfilled as of today. In 2008, if you don’t type those magical “www.”, Google is out of reach.

And that’s why two important events happened during 2008, both of which signal that bigG is aware of threat and that it conceded that its bet not only didn’t materialize, but that won’t materialize anytime soon and that could be a threat to Google itself.

First one was the Gears release: don’t underestimate its meaning. Gears might look like a side technology but it actually is a key technology for Google. For first time in its history, Google concedes and admits that there needs to be (and there will be need to be) “disconnected” applications which cannot work on the Internet only and whose data cannot reside on anyone’s server. For the first time in its history, Google concedes that traditional OSes are going nowhere soon: they will just stay and they will be part of future applications for at least an amount of time which is so long that Google must develop a strategy to cope with that (so it won’t be 1, 2 or 4 years… but longer).

Don’t get me wrong: Gears is great stuff. From a developer perspective, that’s an amazing technology and would be a MUST if… if developers didn’t have BETTER technologies to achieve the same goals. At its heart, Gears is a technology to build DESKTOP applications which could work disconnected from the Internet and access the network when present. It’s a good technology because it tends to blend differences between desktop and web applications, in a way in which developers can work in both ways. But Gears could also be used to built desktop applications which don’t need to connect to the Internet. The bare truth is that’s a technology to build HTML+JavaScript based desktop applications when you have technologies like Flash, Silverlight and Java which could do much more, in a easier way and with better results.

Couldn’t Google develop a technology like Flash or Silverlight itself too? Of course it could but they didn’t. They’re still trying to make developers swallow an inferior technology like JS, mixed to a presentation technology based on a lot of hacks like HTML is, instead of moving to something else. So you will be facing the daunting task of building complex and huge JS-based applications, hard to code and hard to maintain, instead of moving to something better, easier and more powerful. I think any developer would say his/her prayers when assigned to be part of an effort to build a Office-like applications using JS and HTML while Flash is moving to integrate PDF into its runtime and Silverlight will integrate (and already had) XPS into itself and even Google wasn’t able to get even closer to a full Office suite with its pile of cash and legions of developers. Why the hell is that?

The second notable event was the announcement of Android technology. This is Google’s answer to the problem I told you about earlier: if you don’t type “www.”, Google is dead in your life. While many hinted for a long time that Google was going to release its own OS for desktop PCs, bigG ended up releasing its OS for mobile phones. And this is perfectly understandable! They probably learned a lesson while watching Jobs struggle for 8 years with his OS X doll and still requiring iPods to save his ass. 8 years and he still owns … what? 2% market-share?

And I’m sure Google carefully watched IBM and loads of companies (including Google itself) pouring tons of money into Linux just to have a 0.something percent share. Schmidt correctly understood that’s a battle you cannot win until you will be able to download your OS at boot-time from the Internet. But yet, they couldn’t afford to let users stay out-of-reach until they typed “www.” and they needed a solution for that. The mobile phone market is still a virgin one, where Windows doesn’t rule and where other people are in trouble (Symbian anyone?). It’s a market where billions of terminals get sold every year and where people tend to replace their phones every 1 or 2 years at max. And best of all, a mobile phone is even more “personal” than a personal computer! Being on those phones means being able to channel their services to users even when they aren’t connected to the Internet. It means having the ability to get revenues from services you sell BUT ALSO from software you could make. And best of all, I always have my phone near me. For example, I’ve been 15 days in Kenya this August and I didn’t touch my notebook during my holidays but, guess what?, I always had my phone with me.

But Android and Gears prove that in 2008, you have to use an OS and/or disconnected applications to channel your company to users and being on the Internet only will not be enough. Please, note that, as was expected, while the whole Google strategy for the Web is based on HTML+JS, Android features full-fledged applications, which will be compiled to native (or almost “native”, given its VM) code.

This confirms my view: Google was looking for a mean to channel itself in a disconnected way and found mobile phones, the perfect environment. Android indirectly acknowledges that HTML+JS is not a good platform for today’s applications and, if you want to get serious, you have to switch to something better.

So why is Google instead promoting JS in Web and Desktop applications? To understand this, we also have to understand why developers keep investing into JS and only marginally into better technologies like Silverlight, Flash, Java, and I mean in RIA context and not in general (where, for example, .NET and Java are still ruling).

Google’s hidden dirty little secret…
… which is not hidden, nor little and neither a secret but it’s very dirty!

Let me ask myself: if Silverlight, Flash and Java RIAs are so much better than traditional one, why isn’t everyone switching to such technologies and they still invest into HTML+JS?

Answer is simple: because modern RIAs cannot be indexed! As you might have noticed, most websites rely on search engines to drive users to their pages and search engines cannot index Flash, Silverlight or Java content (yes, they can index a SWF file but that’s not enough for a RIA). So those websites which don’t rely on search engines (because they’re way too popular to need that), will switch to an advanced web application (you don’t need a search engine to visit Ford, or Coca-Cola, or Ferrari and so on…) like Flash. Those needing users from search engines will stay HTML and try to provide a better experience by using AJAX because they need their pages to be indexed to get more users.

So where’s the dirty secret? Here it is: Google must assure that developers will keep creating RIAs using HTML + JS because that will protect its search engine and, obviously, its advertising division, which is what drives 99% of Google revenues.

We have no doubts that Google is number 1 in search engines and nobody has doubts that it will stay on the top, unless they commit suicide or… unless rules change. Ask Microsoft or Yahoo: it’s not that they cannot create a good search engine: both of them have resources and technologies to do that and I can say both of those search engines aren’t that far from Google perfection. Yet Google is making loads of money and they aren’t.

Google will stay #1 for the very same reason for which Windows will stay #1: because they’ve been the first “good one”. And once you are the first good one, you just need to play conservatively and don’t bet too much in a single move: you will stay on top. That’s what happened to Windows and that’s what will happen to Google SE.

However, should the focus move from HTML to RIAs, the game will reset and positions will blend as Google might be the top player for HTML indexing but things could be different when you start indexing RIAs because then people who control platforms have a very very big advantage. And Google doesn’t have ANY platform to control.

So when a search engine will start indexing RIAs, Google will need to move and do the same but this time other companies won’t allow the company years of free ride: they will all start the very same second. Maybe Google will emerge again but maybe not…

A Google browser to rule them all
So that’s why Google released Chrome: it must prove that HTML+JS is still a good platform to invest and delay any other framework adoption. If they can succeed in pushing HTML+JS further, their near future will be bright because of dominance of position.

But if the market happens to move to different platforms, Google will have to struggle to stay on top and, without any platform to control, it won’t be easy.

It’s evident that Chrome’s release that the browser is all about defending Google current market. For example, many complained about lack of extensibility and minimalist UI but that’s not a surprise! Google doesn’t care at all about you investing your time in “extending” Chrome… it’s not Chrome the problem but rather the application in it. They don’t want you to provide disconnected code, they want you to code Web applications. You want to provide something “fancy”? Do that in JS. You want to provide some kind of utility? Do that in JS. At max, use Gears to provide disconnected functionalities but stay HTML. I won’t mention, as many emphasized, how difficult it will be to have an ad-blocking extension for Chrome and you can easily understand why.

Take Mozilla and Firefox: Google poured lots of money there and has a strong relationship with FF so one might think Google could have “influenced” Mozilla to bring it where Chrome is. But if they moved to release their own browser was because their priorities were different than Mozilla’s.

And their priority was basically to provide a better JS environment to keep developers investing into JS and HTML.

Look at me, look at me !
I’m sure there will be a very few people reaching this point. This article was sooo long and for sure sooo boring. But for those of you who were so brave to read up until here, I want to play the magician and (boldly) anticipate the future with my crystal ball.

Remember when Steve Ballmer said in a interview (and later, in a memo to MS employees) that Microsoft might be slow but they have resources and time to be number 1? Well, you might not like him but he’s no fool (isn’t he, Steve Jobs? How come Exchange support appeared on your iPhones?). So when he (humbly) said that Microsoft lost first search engine war but would emerge in the long run and that they’re going to change the very whole way you look at search engines, I don’t think he was boiling his supper.

In my opinion, in 2009 Microsoft will announce a revamped version of Live search engine, which will be able to index RIAs. They will probably provide an open API that RIAs can use to feed content to index (and will leave that open not only to Silverlight but to Flash and Java too) and then use some kind of token value to pass each other, in a way SE can later tell RIA which content user wants to see. And they for sure embed such new SE platform inside .NET and Silverlight, in a way some indexing can happen in auto-magical way, the very same way ASP.NET controls can be turned into AJAX-enabled ASP.NET controls. Dreams? We will see…

If they succeed to do that, they would open a break in Google dominance and, most of all, will weaken JS position in RIAs development, which could open new possibilities for more advanced platforms and, in case of Microsoft, for their Windows Server systems, which are extremely interconnected with .NET and Silverlight.

If that will actually happen in 2009 (it’s not a matter of “if”, it’s just when but search engines have to move from where they are now to incorporate new stuff), I’m sure Google will sit down and watch what happens but not for long as they cannot afford to loose a inch. I’m sure they’re full of statistics and if those stats will show that RIA websites will increase beyond current limit, they for sure do a couple of things: first, they will release Android VM for Windows (and possibly other OSes) in order to have a platform they can control; second, they will enable (because I’m sure they already considered this option) their search engine to index RIAs.

As for me, I just hope it will happen in 2009 so I will be able to put a nail into coffin of HTML and move on. Hell, I’ve been waiting for this since 1998…

(Notice: the bulk of this article was written before Google released Chrome. That event only confirmed my view and this article has been adapted to include that)


  1. 2008-09-09 6:17 pm
    • 2008-09-09 6:45 pm
      • 2008-09-09 7:41 pm
  2. 2008-09-09 6:34 pm
    • 2008-09-09 11:01 pm
  3. 2008-09-09 6:38 pm
    • 2008-09-09 11:27 pm
      • 2008-09-09 11:32 pm
  4. 2008-09-09 6:42 pm
  5. 2008-09-09 6:47 pm
    • 2008-09-09 7:03 pm
    • 2008-09-09 8:14 pm
    • 2008-09-09 8:17 pm
  6. 2008-09-09 6:55 pm
    • 2008-09-10 2:44 pm
      • 2008-09-10 2:56 pm
        • 2008-09-10 3:10 pm
      • 2008-09-10 3:02 pm
    • 2008-09-10 9:05 pm
  7. 2008-09-09 7:58 pm
  8. 2008-09-09 8:09 pm
  9. 2008-09-09 8:19 pm
  10. 2008-09-09 8:51 pm
  11. 2008-09-09 10:13 pm
  12. 2008-09-09 10:20 pm
  13. 2008-09-10 1:38 am
  14. 2008-09-10 7:34 am
  15. 2008-09-10 9:31 am
  16. 2008-09-10 2:21 pm
  17. 2008-09-10 3:02 pm
  18. 2008-09-10 3:40 pm
  19. 2008-09-11 1:58 pm
  20. 2008-09-14 3:46 am