But just how good is Chrome? How does it differ from IE and from less popular, but still important, browsers like Mozilla’s Firefox and Apple’s Safari?

I’ve been testing Chrome for about a week, trying out all its features and using it side by side with Microsoft’s latest iteration of IE, which came out just last week.

My verdict: Chrome is a smart, innovative browser that, in many common scenarios, will make using the Web faster, easier and less frustrating. But this first version — which is just a beta, or test, release — is rough around the edges and lacks some common browser features Google plans to add later. These omissions include a way to manage bookmarks, a command for emailing links and pages directly from the browser, and even a progress bar to show how much of a Web page has loaded.

Chrome’s interface has some bold changes from the standard browser design. These new features enhance the Web experience, but they will require some adjustment on the part of users. For instance, Chrome does away with most menus and toolbar icons to give maximum screen space for the Web pages themselves. Also, Google has merged the address bar, where you type in Web addresses, with the search box, where you type in search terms. This unified feature is called the Omnibox.

One striking difference in Chrome is how it handles tabs, which display a single Web page. In Chrome, each tab behaves as a separate browser. The bookmarks bar, Omnibox, menus and toolbar icons are located inside the tab, rather than atop the entire browser. The tabs appear at the top of the computer screen. Chrome also groups related tabs. If you open a new tab from a link in a page that’s already open, that new tab appears next to the originating page, rather than at the end of the row of tabs.

Despite Google’s claims that Chrome is fast, it was notably slower in my tests at the common task of launching Web pages than either Firefox or Safari. However, it proved faster than the latest version of IE — also a beta version — called IE8.

Meanwhile, Microsoft hasn’t been sitting still. The second beta version of IE8 is the best edition of Internet Explorer in years. It is packed with new features of its own, some of which are similar to those in Chrome, and some of which, in my view, top Chrome’s features.

Google’s Chrome browser displays thumbnails of a user’s most-visited pages when a new tab is opened, rather than a blank page.

For example, while IE8 also groups related tabs, it assigns a different color to each such tab group and allows you to close them all with one click. It has a “smart” address box of its own, that drops down a list of suggestions as you type, though it retains a separate search box.

IE8 also has breakthrough privacy features that exceed Chrome’s, and includes a new technology called Accelerators, which allows you to take rapid action on any selected word or phrase on a Web page, such as generating a map for a place name, without switching to a new page.

As they develop, each of these browsers has a good chance of besting Firefox 3.0, which I have regarded as the best Web browser for Windows, the only operating system on which Chrome currently runs. But they will have to get faster at loading pages. And, to best Firefox on the Macintosh, Google will have to make good on its promise to produce a Mac version of Chrome, something it says it will do in the coming months. Microsoft has no plans to produce a Mac version of IE8.

Chrome and IE8 are far more advanced than Apple’s Safari. Safari is speedy on both Mac and Windows platforms, but lacks many of the key intelligent features of its newer Google and Microsoft rivals.

Why is Google igniting a new browser war? There are two main reasons, and both involve competing with Microsoft. First, the search giant fears that because its search engine and other major products depend on the browser, Microsoft — with its rival online products — might be able to gain an advantage by altering the design of IE, which has roughly a 75% market share.

Second, and more important, Google sees the Web as a platform for the software programs, or applications, that currently run directly on computer operating systems, notably Microsoft’s Windows. It says current browsers lack the underlying architecture to enable future, more powerful Web applications that will rely more heavily on a common Web programming language called JavaScript. Chrome was designed to be the world’s speediest browser at handling JavaScript.

That move might one day make Chrome a sort of online operating system that competes with Windows. “Think of Chrome as more than a simple Web browser,” Google declares. “It’s a platform for running Web applications.”

Microsoft’s IE8 has an “Accelerator” feature that lets users select any Web text and then map, translate, search or email their selection without leaving the page.

I tested Chrome, and IE8, on a plain-vanilla Lenovo ThinkPad laptop running Windows XP, and equipped with a modest processor and one gigabyte of memory.

To gauge Chrome’s speed at loading Web pages, I launched two large groups of typical Web pages simultaneously, each site opening in its own tab. One group included 15 sports sites, the second 19 news sites. In both tests, Chrome’s speed fell in the middle, at 35 and 44 seconds, respectively. IE8 was slower, taking 49 and 75 seconds to open the two groups of sites. But Firefox and Safari were much faster, notching identical speeds of 19 seconds for the 15 sites and 28 seconds for the 19 sites.

Google claims that future, more sophisticated Web applications relying more heavily on JavaScript than today’s sites do would run faster on Chrome. Of course, I couldn’t test any claim about future scenarios, but I did run Chrome on several JavaScript test sites, used by developers. It handily beat the other browsers. However, Google doesn’t claim users would see much difference on current Web application sites.

I also tested Chrome’s compatibility with scores of common Web sites. In general, it did well, rendering the sites properly. But I ran into problems with video. Some video sites refused to recognize Chrome, because its development has been a secret. On others, like Major League Baseball’s site, videos mostly played properly, but sometimes didn’t.

IE8 also has some compatibility issues, for different reasons. It’s the first version of Internet Explorer to hew closely to Web standards. Earlier versions used some nonstandard ways of rendering Web sites, prompting some site designers to adopt techniques that made their pages work in IE, but look odd in Firefox and Safari. Now, ironically, these pages also look strange in IE8. So Microsoft was forced to build in a special Compatibility View button that users must click to see the sites properly.

Chrome is built on three core design principles. The first is its spare user interface: just two menus and a handful of toolbar icons. IE introduced a similar approach in its version 7, but with a difference. Microsoft allows users to restore a traditional menu bar; Google doesn’t. The only toolbar icon you can add in Chrome is a Home button.

The second principle is that a user can type anything into a single place, the Omnibox, and instantly get suggestions on where to go, gleaned from the user’s own browsing history and Google’s rankings of popular sites. Whether you type in a Web address or a search term, the Omnibox is very smart. In my tests, it sometimes came up with the right destination after I typed only one or two letters of the name of a site I often visited.

The Omnibox has another cool feature: Tab-to-Search. If you type in the name of another site that includes its own search feature, like Amazon.com, the Omnibox lets you just press the tab key to search within that site, without opening it first. Chrome, through its Options settings, also lets you change the default search engine used by the Omnibox. Instead of Google’s own search service, you can use Microsoft’s Live search, Yahoo search, or others.

The third big principle behind Chrome is that each tab runs, under the hood, as a separate browser. Tabs can be dragged off the main browser and turned into separate windows. If one tab crashes, the rest of the browser keeps running. But this doesn’t work perfectly. In my tests, all of Chrome died on me when I tried watching an Olympics video on the NBC site.

You can even make a tab a standalone application that runs from the Start Menu, or the desktop, as if it was a separate program.

Chrome has a few other key features. When you open a new tab, you don’t get a blank page, but a set of thumbnails for your most-visited pages, plus lists of recent search engines you’ve used, recently used bookmarks and recently closed tabs.

Like other browsers, Chrome puts up a warning when you try to visit a malicious or phony Web site, and it has a private browsing mode, called Incognito, which allows you to browse without leaving any history on your computer — a feature popularized in Safari.

Chrome also has a pop-up blocker, but it’s annoying because it flashes a notice that a pop-up has been blocked. IE also does this, but unlike in Chrome, the warnings are much less intrusive.

Internet Explorer 8 has some new features Chrome lacks. Its private browsing mode, called InPrivate, is the first I’ve seen that not only leaves no traces on your own computer, but also bars Web sites from collecting some types of information on where you’ve previously been surfing.

While IE8’s address box and search box remain separate, each also offers rapid suggestions; and both are organized better than Chrome’s. For instance, the suggestions that drop down from its address bar are divided neatly into categories drawn from the browser’s own guess, your history and your favorites. One downside: For this to work in Windows XP, you must first install Microsoft’s desktop search product.

Like Chrome, IE8 lets you switch your default search provider, but it also allows you to switch search engines on the fly. When you type in a search term, icons for alternate search engines appear at the bottom of the suggestion list, and you need only click on these to see search results from, say, Google, instead of Microsoft’s own Live search engine.

IE8’s Accelerators feature presents a blue-arrow icon above any text on a Web page that you have selected. Clicking on the icon brings up a list of actions you can take using the selected text, such as posting it to a blog, emailing it, mapping it or searching it. While these actions are set by default to use Microsoft’s own Web services, you can change them to use Google’s, Yahoo’s, or those from other companies.

Microsoft also has built in a feature called Web Slices. These are portions of a Web site that a site developer can designate to appear in the IE8 Favorites bar and to constantly update themselves. An example might be bidding on eBay.

Like Chrome, IE8 also displays useful information whenever you create a new tab, including a list of recently closed tabs and a list of Accelerators.

With the emergence of Chrome, consumers have a new and innovative browser choice, and with IE8, the new browser war is sure to be a worthy contest.

Find all of Walt Mossberg’s columns and videos online, free, at the All Things Digital Web site, walt.allthingsd.com. Email him at mossberg@wsj.com.

This interface is generally called “multitouch,” and it involves using one or more fingers on a screen or touchpad to perform special gestures that manipulate lists or objects on a screen — without moving a mouse, pressing buttons, turning scroll wheels or striking keys.

The best-known example of the interface is on Apple’s iPhone and iPod Touch devices. It allows you, for example, to rapidly flip through photos, lists of emails or song titles by merely “flicking” a fingertip on its screen, or to resize a photo by “pinching” the image with two fingers. And, this month, Apple moved some of these multitouch features onto a laptop, its new MacBook Air, where fingertip actions are performed on an oversized touchpad rather than on a screen.

Apple didn’t invent the multitouch concept. Academic and commercial researchers, and small, obscure companies, have been working on it for years. Apple is adapting the concept, adding its own ideas and popularizing it — just as it did in the 1980s with the mouse and the graphical user interface, which had also been invented elsewhere.

Rival companies are scrambling to add multitouch features to laptops and other digital gadgets. Synaptics, a leading supplier of touchpads for laptop makers who compete with Apple, has announced that shortly it will incorporate several multitouch features into its touchpads. Microsoft is producing a coffee-table-size computer called the Surface, meant for hotels, casinos and retail stores, that uses multitouch finger gestures to move around digital objects such as photos, play games and browse through product options. Hewlett-Packard developed a prototype of a similar multitouch coffee-table computer for home use.

Rotating a photo on Apple’s MacBook Air

And, in the back rooms of this month’s Consumer Electronics Show in Las Vegas, various cellphone makers, seeking to emulate the iPhone, were showing off their own unannounced efforts at multitouch. One prominent cellphone maker, Taiwan-based HTC, has already built a phone, called the Touch, which slaps a rudimentary multitouch interface on Microsoft’s Windows Mobile operating system.

The basic touch screen, of course, is an old and familiar thing. People use them every day to get money from ATMs or to check into flights at airport kiosks. Microsoft’s Tablet computers and Palm’s Treo cellphones, among others, use touch screens that employ styluses or fingers to move cursors, select icons and other items, and even to write on the screen.

But multitouch interfaces are potentially much more versatile. They allow you to use your fingers to manipulate virtual objects on a screen as if they were real, sort of the way Tom Cruise’s character did in the 2002 Steven Spielberg science-fiction film, “Minority Report.” For example, Microsoft’s Surface allows users to rearrange groups of digital photos by just dragging them around on the table top as if they were actual paper prints.

Unlike the touch screens on, say, ATMs, multitouch devices are able to distinguish between the press of a single finger and the press of multiple fingers, and to interpret the motions or gestures you make. They take different actions depending on how many fingers they detect and which gestures a user performs.

On Apple’s MacBook Air, the touchpad still allows you to use one finger to move the cursor and click like a mouse can. But, optionally, it can do much more using multitouch gestures.

You can rotate photos by just touching two fingers to the touchpad and moving the images on the screen as you wish. You can quickly move back and forth through a series of Web pages or photos by “swiping,” or placing three fingers on the touchpad and moving them rapidly sideways. And you can shrink or expand a photo, or zoom in and out on a Web page, by pinching the image.

All recent Mac laptops, not just the new Air, have the optional ability to scroll through a screen without using any button or special zone on the touchpad. You just place two fingers, instead of one, anywhere on the touchpad and drag them across its surface.

In addition to pinching, the new Synaptics touchpads have a feature called “ChiralMotion,” which achieves rapid scrolling by moving two fingers in a circle anywhere on the touchpad. Another Synaptics feature, called Momentum, lets you emulate a trackball by flicking a single finger across the touchpad. This gesture can move an object on screen, like a virtual bowling ball in a game, with the illusion of inertia.

It’s too early to know if the new multitouch approach will ever be as big as the mouse-driven graphical user interface. But it’s already evident that it offers real advantages on devices where a mouse isn’t possible or convenient to use, or the decades-old interface of menus and folders is too cumbersome.

Email me at mossberg@wsj.com. Find all my columns and videos online, free, at the new All Things Digital Web site, http://walt.allthingsd.com.

New languages have been added to the traditional lists and Arabic, in particular, has been in demand recently. I spent the past few weeks tinkering with four free online services, translating various texts from English to Arabic and vice versa to test their speed and accuracy. I tested Google’s Language Tools and services from Applied Language Solutions, WorldLingo Translations and Systran.

Customers who have been waiting for such services to be perfected will find improvements are slow in coming. Overall, I found the Arabic-English translations rife with syntactic and semantic errors — from the merely too-literal to the laughably bad.

For the purposes of my test, I selected different texts: conversation, news stories, and legal and scientific documents. First, I picked an Associated Press story that started with the sentence: “A wintry storm caked the center of the nation with a thick layer of ice Monday…”

I got a variety of imprecise translations into Arabic (which I’m interpreting below).

Applied Language and WorldLingo offered identical translations, which were slightly better than the other two: “A storm covered the center’s storm from the nation with a thick layer snow Monday.”

Systran: “A stormy storm covered the center for the mother with a thick layer snow Monday.”

Language Tools: “The storm grilled bloc in the middle of the nation with a thick layer of snow Monday.”

The translations would have been nearly impossible to understand were I not fluent in both languages. It’s worse in Arabic than it seems above. Arabic has masculine and feminine nouns, verbs and adjectives that have to agree in a sentence; otherwise, the sentence makes a native speaker wince.

Next, I processed some longer news stories. Only Language Tools didn’t set text limits. WorldLingo and Applied Language each had a 150-word limit. Systran didn’t specify a limit, but it rendered only a short part of the text.

Language Tools came out ahead this time. It was the only one to translate the word “Taliban” from Arabic to English contextually correct, as a movement. The other services translated it literally from the Arabic as “two students.”

The services were better at translating everyday phrases, but even these sometimes came out missing a word, or were scrambled.

In this category, I again found translations by Google’s Language Tools closest to the original texts. Still, there is much room for improvement. Google, for example, translated from Arabic to English the simple question, “Do you speak English?” as “Do they speak English?”

Other services got the pronoun right but botched other parts of the sentence. With the exception of Google, all three services, oddly, attempted to write the Arabic word for “English” in the Roman alphabet (aalaanklyzyh) in the middle of an Arabic sentence.

All the services did a terrible job with metaphors and other figurative uses of the language, whether Arabic or English.

The weakest performance by all the services was the translation of legal and scientific texts. Only Language Tools correctly translated the word “noncompliance” in a legal text, for example. Instead of using the proper word in Arabic, the other services transliterated it phonetically into a meaningless word.

All four services have an interface that is easy to use, with a pull-down menu listing several languages. Each has two text boxes, one for the original language and the other for the desired translation. They also translate entire Web sites, but the translation again tended to be awkwardly verbatim.

Google also has a feature that lets you translate search results free. (It also offers users an option to send in a better translation.) The others require you to become a paid subscriber. English and Arabic results appeared side-by-side.

I also liked WorldLingo and Applied Language’s email-translation feature. After clicking the email button, a window with two text boxes pops up. You enter your name and email address, and the recipient’s name and address. When you send the message with WorldLingo, both recipient and sender see the message in both languages. Neither Google nor Systran has this feature.

Systran has a convenient swap button that lets users easily flip the source and target languages. This saves time when going back-and-forth between two languages. The other services have you use pull-down menus. Systran’s interface also allows prompt translation of a text as soon as it’s pasted in a text box, without the need to click a “translate” button.

Free online translation tools help travelers or those curious about languages, but I found them unreliable for important documents. Use with caution.

Write to Sarmad Ali at sarmad.ali@wsj.com Walt Mossberg is on vacation.