Dave’s Lab

Thoughts on the Reaction to Google Chrome

Sep 10 2008

It has been a week since Google released their addition to the web browser market, Google Chrome. In case you’ve been in a coma for the last seven days or turned off the RSS feeds for any tech site on the internet (I wouldn’t blame you), Chrome is based on WebKit, the same spectacular rendering engine behind Safari, Adobe AIR, the Google Android browser, the iPhone, and a few other projects. Also, much like WebKit, Google is open sourcing almost all of the work that they’ve done on the browser, including new additions to the framework like their Javascript virtual machine called V8. They have also included a way for outside developers to contribute to the project via Google Chromium. This is to be commended in my view, because many other open source projects sponsored by big companies, do not allow developers to contribute to the product.

Google has released this product only for Windows users, for now, but they intend to release versions for Linux and Mac. This is to be expected of course, since WebKit is, after all, largely developed by Apple and we’ve seen ports of the engine ported to Linux in some upcoming projects. The real problem right now is that they’re trying to finalize UI (user interface) and integration designs, which, quite frankly, I don’t mind them taking their sweet time on. Too often, open source projects get way too excited about the idea of the project, rather than the people who will be using it, and, as a result, user interface in open source projects falls to the wayside. To be clear, this is purely my opinion, but I strongly believe that software is inspiring and worth spending my life studying, using, and making because of people, not in spite of them.

Now, I know that this post, so far, is off topic from my stated mission of helping you with your computer problems. I promise that I am working on a way to make many, many more tips available, in shorter write-ups, so you’ll see more content coming at you soon. However, I was so interested by Google’s efforts in Chrome and what it could mean for the rest of us, and was so frustrated by journalists’ coverage of the event, that I felt compelled to put my thoughts out there. So, here goes.

Following the advice of Henry McCracken, formerly of PC World and now founder of Technologizer, you can put almost anything in a list, so why not my contentions? Here are my problems with most news analyses of Google Chrome.

• People have been discussing the introduction of this new browser as if it was just simply that, a new browser. However, I think we should all at least objectively recognize that this product is a little bit more. When we discuss Chrome in the context of Firefox, IE, Safari, and Opera, the discussion should not mirror the tone of comparing cars. For Google’s engineers, this was not a move from a Ford to a Honda. This was about moving from gas powered cars to hybrids. Though you may be able to sell this product with claims of being faster or better, most users will not care about what’s under the hood of their car (or their browser, for that matter). This remains true even if the technology in this new thing they use is revolutionary. The inherent problem I have with the coverage of Google Chrome is that almost every analyst and member of the technorati has reported on it as if they were a normal, unconcerned user. Now, I can respect this since they are trying to reach a broad audience, but I also am disappointed at their superficial concern.
• One thing people have been poking fun at is that Google Chrome is in “beta”, meaning that it is unfinished. Critics have looked to the “beta” status of mainstream Google products like Gmail and Google Calendar, which have been so for years, and thusly expect a fully ready product release out of the Google team. However, ladies and gentlemen, this is not just another instance of Google abusing the term “beta”, but rather this is an actual unfinished product. I do not mean that they intend to add on new features to it as time goes on, but that it has not yet reached a level of completion that they find fully satisfactory, i.e. the real meaning of the word “beta.” How can I assert this? Google has said so! They are certainly excited about this product and have put a lot of energy into it, but they have said it is not fully complete as they want it to be. Furthermore, the product announcement (the video, not the comic) was done by geeks. Google did not bring out Marissa Mayer or Larry or Sergey or Eric to make this announcement, but rather the humble, regular engineers who were working on the product. The whole tone of the event was geared toward making a huge first step, but not giving you something finished. It was an announcement for developers, by developers, in my opinion.
• To be fair, in a counter point against this last one, Google has put links to the download page for Chrome on its front page, which might indicate that they want it to receive mass-adoption. However, I am inclined to think this is just to get people using the thing, before they start pushing it in all areas.
• For the love of Pete, folks, this is not a “new” browser! It uses WebKit, the same engine as Safari, so web developers don’t have to “worry about writing code for yet another browser.” Take it from me, folks, I am a web developer and I am glad to see it. Furthermore, even if you don’t develop for Safari, WebKit is so standards-compliant, that you won’t have to worry most of the time.
• Many people have wondered out loud if Google intends to farm and harvest your information if you use Chrome. I must say that after the whole EULA fiasco, this is certainly a valid concern. However, I must say that if Google did that, the tech elite of this country would freak and since the technorati are the alpha consumers and early adopters, a wall of fire through whom many new tech products must pass before making it to the average person, it would be suicide for Google to do that. Is that reasoning fool-proof? No, but it seems to be a fair assumption that if the journalists and bloggers and media crowd don’t like a product, it will not get adopted. (PS I’d love to hear examples to the contrary in the comments.) A more reasonable approach would be to use Google Chrome as a sort of funnel that draws people into their other services like search, email, advertising, etc. Almost every product Google has made and released does exactly that, and it has been a great strategy so far. A good example of this principle in action was KeyHole, which you know as Google Earth. It used to be horribly expensive, but then Google bought the company and released it for free. Why? Because now everyone had an awesome, free product that directed you straight to Google’s search and maps. Furthermore, I have heard from a Microsoft executive that when Google pays hardware manufacturers like Dell to release their products with Google as the default search engine, they make as much as $30 per person per month! That’s just off the ads in their searches!
• Finally, the most underrated aspect of Chrome has been the revolutionary new Javascript technology in it, with multi-process design and a new virtual machine called V8. I could quite honestly write a book on why these two things are so critical for the future of Javascript, but I’ll give you a small idea. V8 is just a simple C++ application. It can be embedded in any other C++ application and can be ported to any platform. Sound familiar? Languages like Java, Python, Perl, and Smalltalk especially have adapted this concept. Imagine if all of the web developers out there could take this blazing Javascript VM and start to write desktop applications…
• Update: Mozilla is NOT going to adopt WebKit! These rumors are beyond ridiculous. Does anyone in the press have any coding experience whatsoever? I think not. For a great argument in my favor, read here.

Alright, so now that I’ve beaten up on almost every journalist whose work I read in my RSS feed, I should offer you some negative things about the coverage of Chrome on Google’s end:

• Google has really made a big mistake by only open-sourcing most of the project. Critical components of the program like the updater and security patches have remained closed. Now, I can understand why Google would do this, especially in the name of security, my utmost concern as well. However, consider the transparency that Mozilla has maintained, even in the light of several embarrassments I might add, and compare that with their phenomenal security record. It is possible to be both transparent and secure.
• Google really, really should have been a lot clearer about who this release is directed towards. It is my contention that they intended power users and alpha dogs and developers to take hold of this new product, but many have argued that they sought full market penetration with this one. I’d love to hear what you think of this in the comments, but we can certainly agree that it is ambiguous.
• The security model of this browser is way too flawed. Yes, they’ve done an excellent thing by introducing real sandboxing of the pages and lower the system privileges of the applications, but they messed up on a couple of basic points. One of these is that there is no way to contain your saved passwords in Chrome with a master password, a feature that Firefox has, though it is admittedly turned off by default.
• Of course, it is certainly annoying that I have to wait for a Mac and Linux version of Chrome, but like I said, I don’t mind.
• (Insert all regular features that most other browsers have that Chrome does not like spellchecking, add-ons, etc.)
• And, most obviously, Google has really done a terrible job in fighting back against the criticism that Chrome is “just another regular browser that I don’t need.” – (journalist shall remain unnamed)

Okay, so you’re now exhausted from reading this long criticisms of Google Chrome, the media response, and Google’s road ahead, but you’re probably still wondering why it is exactly that I am so pumped about the technology under the hood in Chrome. I’ll tell you why. Google Chrome uses a different process per tab and window, such that when you close a tab or window, it is really closed. It’s completely gone and does not leave excess memory thrown about. Furthermore, in an age when we will soon seen processors with almost a hundred cores, one wonders whether or not applications that still rely on a single process or thread will be able to cut it as “fast” in the future. Multi-threaded applications and multi-process architectures are a going to be a key phase in the next evolution of software development. Google, with this release, has started pushing the browser market in that direction. So while, as a browser, Chrome is fairly good, as a technology it is extremely exciting and deserves a different kind of attention.