The Mac Sucks!

Today I read CNET’s coverage of the Opera 10 “Unite” service. To save you going to the link, the Opera developers decided that it would be a lot better for you to share your content with the world via your web browser rather than a third-party service like YouTube or Flickr. It does this by running a web server on your PC which other users access across the Internet. When I read this, I was immediately struck by several thoughts and questions:

• Why? Maybe if this was 1995, I could see the point of Opera Unite. Today, with well-known and (arguably) good services like YouTube, Flickr, and others out there, what’s the point of serving your media yourself? Why take the risk of opening your personal computer to the Internet, regardless of how careful and secure you are? If people really wanted to share their own content directly from their own PCs, wouldn’t a “real” web server like Apache make more sense? Maybe within a corporate intranet there might be some value to this, but even there, most companies have central servers for doing this sort of thing and would probably prefer not to have end user desktops doing ths sharing.
• What about security? CNET makes a point of mentioning that security experts are already concerned about Opera Unite. They have plenty of good reason. The only real protection in Opera Unite is based on passwords, and certain ways of sharing information put those passwords “in the clear” where anyone can read them. When asked if someone could get to resources on your PC that you didn’t explicitly share, an Opera spokeswoman said “Definitely not — unless they’re a hacker.” I wasn’t too worried that my friends and family would go looking for unauthorized data on my PC. It’s the hackers I’m worried about. Those are the ones who commit blackmail, fraud, and identity theft with your data. Your friends and family, not so much.
• What about that “whole new class of social services”? Opera Unite is supposedly extensible, allowing third parties to create plug-in modules to add functionality to it. Even if Opera itself is bulletproof, what are the odds all those third-party modules will be?

The Opera web browser itself, I like. It’s relatively light on resources, reasonably quick, seems to do a good job displaying web pages, etc. I’m not against people using it. Opera Unite, on the other hand, concerns me. It’s basically turning everyone who uses it into a web server administrator, without the requisite training in security. If it offers something you think you really need, then by all means go ahead and use it. But if you want my advice, I’d steer clear until it’s had a chance to prove itself in the real world.

If you really need your own web presence, look into third party web hosting. Some providers, like Godaddy.com, are relatively inexpensive and can provide the tools and expertise you need to get your content online without having to risk opening your personal computer up to the world. If you need absolute control over your content and want to host it on a box that belongs to you, at least invest in a separate machine to serve that content so that your personal data isn’t at risk if that machine is compromised.

A little while back, Apple caused a stir in the technical community by alleging that its Safari 4 browser was the fastest browser available for Macintosh and Windows. Since then, a lot of articles have been written comparing the Safari browser to various others. The June 2009 PC World issue (on page 10 for those following along at home) is an article entitled “Browser Speed Showdown: Chrome is Golden” in which the page rendering speed of the Google Chrome 2 Beta, Mozilla Firefox 3.0.7, Microsoft Internet Explorer 8, and Apple Safari 4 Beta are compared.

In the comparison, the page load time for each browser is timed for several popular web sites, such as amazon.com, apple.com, and others. An average page load time is computed as well. The results of the comparison appear below:

According to the article, for each test they cleared the browser’s cache and then loaded each page 10 times per site, per browser, to factor out fluctuations in network traffic and to build a sample size large enough to identify trends. In addition, they threw out the two best and two worst scores for each test to reduce the influence of fluctuations and provide more consistent results. The measurements did not rely on the browser’s indication that it was finished rendering the page, but waited until all visual elements were loaded and ready to use.

The fastest result for each site appears in bold in the table, and the fastest overall average also appears in bold. As you can see, for the web sites tested, the Chrome 2 Beta beat all the other browsers in every test except one. It was also interesting to note that Firefox and Safari, which are normally considered “faster” browsers by many users, actually tied for last place in this particular test.

However, as PC World noted in the conclusion to its article “Many users won’t notice any performance difference in the browsers we evaluated. With fast broadband service, you likely spend little time waiting for pages to load anyway… All four of the browser we tested are pretty fast, so you should focus on which one seems most compatible.” It might also be useful to examine the feature sets of the browsers to find one that works most like the way you do, since page rendering speed isn’t that different among them.

Stephen Shankland of CNET posted “Safari 4 a big step up, but not as far as rivals” on Tuesday. In this article, Shankland outlines the major new features of Apple’s Safari 4 browser for Windows. He says that “a big user interface overhaul makes Safari look polished rather than clunky on Windows, builds in better search abilities, and makes good use of the fact that people often visit the same sites over and over.” At the same time, however, he laments that “the lack of something like the extensions architecture that Firefox pioneered still means Safari 4 is only better than Safari 3, not the competition.”

For those who haven’t looked at Safari 4, the new release goes for a more Windows-native appearance than Safari 3, which tried to bring the Mac OS X look to Windows. Apple has organized tabs at the top of the window where a title bar normally resides, as Google did in Chrome. Safari 4 also introduces a “cover flow” style view of browser history, allowing you to flip through a gallery of pages you’ve recently visited. They’ve also added a Chrome-like feature called “Top Sites” that puts previews of several of your most-visited web sites in a sort of 3-D view. It’s all eye candy, some of it seen first in other browsers, and typical Apple “dazzle ‘em with graphics” stuff.

The big claim to fame for Safari 4 is improved performance. On the Apple web site, Safari 4 is touted as “the world’s fastest browser“. On its web site, Apple claims that Safari “outraces Firefox, Internet Explorer, and Chrome” thanks to faster JavaScript rendering and “the industry’s most advanced rendering technologies”. Lee Matthews of Download Squad too a look at those claims, which are being echoed in major media outlets, and says that “Safari 4 can’t beat Google Chrome“. Matthews shows that on the SunSpider benchmark Apple brags about, that Chrome 2 is faster than Safari 4, and that Safari 4 is only barely quicker than Chrome 1. Using the Google V8 benchmark, both Chrome 2 and Chrome 1 beat Safari 4 (as Matthews says) “like a rented mule, and Firefox 3.1 barely puts up a fight”. Matthews concludes that “based on overall numbers, however, Chrome 2 still wins the Javascript speed title”.

Computerworld echoes Matthews’ findings, saying that “Safari 4 rivals Google Chrome in JavaScript race” but that “Contrary to Apple’s claims, the newest version of Safari is not the world’s fastest browser, benchmark scores show.” Computerworld’s tests show that the Safari 4 beta is “in a virtual dead heat” with the most recent edition of Google’s Chrome.” The Computerworld article indicates that Safari is only the “world’s fastest browser” on Mac OS X, where there’s no Google Chrome for it to compete with.

Microsoft, whose Internet Explorer browser tends to get beaten badly in these benchmarks, naturally says that these JavaScript benchmarks are something akin to drag racing and don’t prove anything. Their contention is that, in the end, it’s the user’s perception of a browser’s speed that matters, not its performance in synthetic benchmarks like SunSpider, Dromaeo, V8, or iBench.

Regardless of which is “the fastest browser in the world” it will be interesting to see how Microsoft, Mozilla, Google, and Apple continue to compete and differentiate their browsers.

According to an article by Ina Fried on Download.com, Microsoft has announced that it won’t be finished with Internet Explorer 8 until 2009.  They plan to offer one more public test version of IE 8 before releasing the final version.  That public test version is expected to be released in the first quarter of 2009, meaning that Microsoft will miss its goal of finishing IE 8 in 2008.

It’s good to see that Microsoft is taking its time getting IE 8 to market, which implies that they’re taking the extra time to squash bugs and collect user feedback.  (In fact, Fried makes a note of this at the end of the article, saying that Microsoft’s “Hachamovitch also called on technical users to download the current beta 2 version and let Microsoft know how that goes.”)  I suppose that’s the advantage of working on a product that doesn’t (at least directly) generate revenue for Microsoft…

BusinessWeek poses the question “Will Google’s Browser Hurt Firefox?” on their site. I would say the short answer is “yes” but that doesn’t tell the whole story.

Google’s “Chrome” browser will get attention if for no other reason than it’s from Google, a company that’s perhaps second only to Apple in its ability to capture the attention of the media when it produces something new. Even that statement short-changes the impact that Chrome is likely to have on the browser marketplace. If you read Google’s description of Chrome, it will minimize (though it can’t eliminate) the potential for browser-based malware attacks to do any significant damage to a system. It will (likely very drastically) improve the performance of Javascript-heavy browsing, such as Google’s own browser-based apps. More perhaps than Firefox and IE, Chrome is in a position to be heavily tested, since Google can rely on its huge database of web sites to find flaws in Chrome – perhaps even before the users see them.

This is not to say that Chrome will be the be-all and end-all of browsers. Its future market share will depend, in large part, on the quality of the software and the browsing experience. In its early days, FireFox was praised for its speed, but jeered for its lack of compatibility with many popular web sites (which were largely optimized for Internet Explorer). It seems to have largely overcome that today. I suspect that Chrome may have similar issues “out of the gate” in that it’s based on WebKit, the same technology behind Safari and Konqueror. We already know from existing reports that Safari’s compatibility with IE-optimized sites is far from perfect. By borrowing the same code base, Chrome will very likely inherit the same issues.

Perhaps the greatest benefit from the development of Chrome, however, won’t be the browser itself. Even if Chrome is a complete failure in the market, there is a very real chance that the work done on its open source core will improve the compatibility of Safari, FireFox, and other open-source browsers. And, as Google suggests in its comic book, Chrome might be taken over by a third party and improved in ways that Google couldn’t dream of. So Chrome’s real benefit is the effect it should have on WebKit-based and open source web browsing. It represents a new way for Mozilla and Microsoft to look at the browser itself. While it’s probably too late to do anything for Internet Explorer 8, it could inspire Microsoft to do good things with IE 9 (or whatever follows IE8).