Underexposed

I have proof from an expert that the iPhone interface really is better. Who's the expert? My 3-year-old son.

Over the years, I've seen countless newbies struggle to use the latest gadget, computer, or software. I like new technology, but it's been work hauling myself up learning curves.

But I'm convinced that after years stuck with only modest tweaks to the WIMP interface--windows, icons, menus, pointing device--real change is upon us. That's chiefly because the pointing devices now can be your own fingers.

Levi types random words on the iPhone's notepad application.

(Credit: Stephen Shankland/CNET News)

Within moments of his first crack at an iPhone, my son, Levi, had figured out how to flip from one photo to another by flicking his finger across the screen. He understood with no coaching how to steer the simulated steel ball around the holes in the Labyrinth game by tilting the phone. He loves to type nonsense words on the notepad application using the virtual keyboard, deleting them once they've been read. In the three months since I got the iPhone 3G, Levi has learned to take photos, browse them, change the phone's wallpaper, and, unfortunately, turn off Wi-Fi and switch on airplane mode.

My proudest moment came when Levi issued his first tweet, borrowing my account: "Eesfrrgjlphdvlksxnjjktwsdvnjmmkbvvnn." Though it was largely a matter of chance, of course, he could do it because he likes the cute bluebird icon of the Twitterific application, and touching it with his finger triggers entertaining interactions.

And I was intrigued when Levi tried unsuccessfully to use the phone's accelerometer to play JellyCar, trying to spur the car by tipping the iPhone so the car would "roll" downhill faster. Note to JellyCar developers: your user interface needs work.

Levi's first tweet

(Credit: CNET News)

As a parent, of course, it's tempting to assume that Levi's accomplishments are the result of his astounding intelligence. But of course much of the credit has to go to Apple and others who've advanced the state of the interface art.

"Human beings are a lot more programmed to manipulate things with our hands and fingers," said Dan Saffer, a founder of Kicker Studio and author of Designing Gestural Interfaces. "I was at a party with a Microsoft Surface table. There was an infant playing with it, not even a year old, pushing photos around and squealing. It's amazing how much it makes sophisticated computing power accessible to a hugely wide segment of the population."

Keyboards and mice aren't being replaced--they offer speed and precision for typing words, entering data, navigating documents, and issuing commands. But they are becoming just one of a host of mechanisms.

Touch screens, available on some Hewlett-Packard computers, are a big part of the revolution, letting people interact more directly rather than relying on a mouse, joystick, or other indirect pointing device. Multitouch sensors, which can detect multiple fingers simultaneously, add more sophistication, such as the ability to shrink a photo by making a pinching gesture on a trackpad. Newer Apple laptops offer more extensive use of multitouch, though at this stage only through the trackpad rather than a touch screen.

Levi took this picture of his feet with my iPhone while he was sitting in his car seat.

(Credit: Levi)

Computing devices also are getting ears and eyes. Speech recognition is available in rough form to power phone search on various phones with services from Yahoo, Google, Microsoft, and Vlingo. FluidTunes lets you control your iTunes library by waving your hands in front of a Mac's video camera.

Intuitive, physical interfaces aren't just for kids. I was stunned to see my technophobic mother-in-law gleefully bowling with a Wii last year. Sure, she couldn't have installed Nintendo's still-popular gaming device if we'd paid her, but using it was as easy as tossing a pebble in a pond.

"There are Wii bowling tournaments now for elders. It takes a sport they love, but there's no weight of the ball anymore. They can play it in a wheelchair. It's a huge hit at nursing homes," Saffer said.

It's not just that devices are easier to use when you can touch the interface, he said. It's that it's easier to learn by watching others use them.

"One interesting thing about touch screens is there's this whole realm of observation you don't have with standard computer setups, where the icons are smaller, and it's hard to tell what people are doing by watching," he said. "You can learn how to use an iPhone by watching people flip through it for a second. You can get it in a way you can't with a standard phone, where you're watching people push buttons to get through menus."

Of course, immersion helps, too. Levi's parents spend altogether too much time punching at keyboards and staring at screens, so he's got plenty of examples to emulate his elders. As a camera buff, I'm delighted when Levi pretends to take pictures--he made a toy camera out of Lego once.

A view of Microsoft's Surface device at the Sheraton Gateway Hotel shows a list of entertainment options for hotel guests.

(Credit: Stefanie Olsen/CNET News)

But I vacillate between pleasure that he's learning how to use technology and fear that he'll grow up ignorant of the non-electronic world. I'm prone to inordinate "screen time," a term heavily freighted with negative baggage in our household, and Levi's childhood will be far more digitally immersed than mine.

And perhaps worse, there's the prospect of losing my status as resident guru. There are plenty of more technically proficient people in my orbit, but none of them live in my house, and Levi doesn't ask any of them to read his typed nonsense words.

Most families come to some sort of reckoning when their son beats their dad in basketball. Ours will come when my wife asks Levi for technical support.

It looks like Mark Hamburg, an Adobe Systems Photoshop and Lightroom programming guru, will be leading work to give Microsoft Windows a better user interface.

And given the dramatic user interface differences between earlier and later Adobe projects that Hamburg worked on, that raises some very intriguing possibilities.

Adobe Photoshop Lightroom is used to edit and catalog photos, chiefly the raw images that come from higher-end digital cameras. Compare its design, deliberately imbued with 'personality' and 'elegance,' to that of Photoshop below.

(Credit: Adobe Systems)

Microsoft and Adobe Systems confirmed Hamburg's move on Monday, but at the time, Microsoft wouldn't share details beyond saying Hamburg would work on "user experience" for the company. However, Chicago photographer and Photoshop consultant Jeff Schewe, who caught a plane to California to attend Hamburg's going-away party, shared a lot more on his blog.

"He was heavily recruited by Microsoft and given an unbeatable opportunity to work outside his normal digital imaging field," Schewe said. "Mark was invited by (Microsoft Chief Technology Officer) David Vaskevitch to come lead a team working on the future of operating system user experience at Microsoft."

Adobe Photoshop's interface has well over a decade's worth of accumulated menus, panels, and dialog boxes.

(Credit: James Martin/CNET Networks)

Schewe also quoted Hamburg about the change: "Given that I find the current Windows experience really annoying and yet I keep having to deal with it, this opportunity was a little too interesting to turn down. I can't imagine doing serious imaging anywhere other than Adobe, but I needed to do something other than imaging for a while."

Hamburg's baby: Adobe Photoshop Lightroom
So what does Hamburg's move portend? It's way too soon to say Microsoft will be better able to counter the widespread opinion that Apple's Mac OS X is superior, but Hamburg's Adobe work sheds some light on the new possibilities.

Hamburg joined Adobe to work on version 2.0 of Photoshop in 1990. After Photoshop 7 was released, he turned his attention to lead Shadowland, the project that became Photoshop Lightroom. That software, which is used to edit and catalog photos, is a major break from Photoshop when it comes to user interface.

Where Photoshop has a seemingly endless list of menus, submenus, dialog boxes, and configurable panels, Lightroom adapts to the task at hand.

Central is the photo in the middle, as large as possible. Adjustment panes can be pulled out from all four sides based on various tasks. The software shifts appearance according to modes for managing catalogs, developing an individual photo, showing slideshows, printing, and creating photo galleries for the Web.

Overhauling user interfaces can be tough, though. Short-term pain caused by unfamiliarity can challenge the long-term benefits of a clean-slate design.

Adobe is proceeding cautiously with a Photoshop interface overhaul. And Microsoft has had trouble with its "ribbon," which presents a task-based interface across the top of Microsoft Office 2007 programs. It's been tough for many users to adjust to the ribbon, and Microsoft is trying ways to make it easier to find the commands they want to perform.

Hamburg's goals: "elegance," "personality"
Some possibilities can be gleaned from Hamburg himself. He discussed some of his Lightroom design goals in a 2007 blog posting.

"We wanted Lightroom to seem elegant, to exhibit grace, to show an attention to style beyond the utilitarian aspect that dominated Adobe's products up to that time. We wanted a richer UI experience," Hamburg said.

And Adobe wanted to give Lightroom a deliberate personality--even if that means some feathers are ruffled.

"One of the goals in Lightroom was to consciously think about the product personality we were trying to create with the expectation that a less accidental personality would induce a stronger emotional reaction in users. That stronger reaction can be both positive and negative," he said. "The second part of this goal was to have enough passionate users to outweigh the detractors."

Finally, he said Adobe wanted to balance power and complexity, adding the latter only when it significantly increased the former.

Designing a user interface for a product with as limited a range of abilities as Lightroom is a very different task than a user interface for an entire operating system, though. But even if Windows doesn't directly copy Lightroom, for example, by changing its look to suit the task at hand, I for one would welcome a version of Windows with elegance, personality, and power.

KDE programmers released a significantly revamped version of its Linux graphical interfaces software on Friday, incorporating several features that also appear in Windows Vista and Mac OS X.

KDE features a new start menu reminiscent of Windows Vista and a new System Settings interface similar to Apple Mac OS X. Click the above image for more screenshots of KDE 4.0.

(Credit: KDE)

Among new features in KDE 4.0 are a start menu on steroids called Kickoff, new ways of viewing widgets and applications, a revamped file browser, and a new look to some entertainment applications that I hope will help pioneer a new user interface technology.

Unfortunately for KDE fans, the upgrade to version 4.0 comes at an awkward time, just a few months before Ubuntu's planned release in April of its "Hardy Heron" version of Linux. This will be the second version of Ubuntu for which its backer, Canonical, offers long-term support. Because Canonical wasn't confident that there would be good developer support for the previous KDE 3.5 and expected KDE 4.0 not to be mature enough, Canonical decided to support just GNOME.

But there still are plenty of other Linux distributions, and KDE 4 will work fine on Ubuntu (the version is called Kubuntu) even if commercial support is absent. And let's face it--Linux on the desktop has appealed more to programmers and technically savvy do-it-yourselfers than to mainstream computer users.

KDE (K Desktop Environment), is one of the two major interfaces for Linux, the other being GNOME (GNU Network Object Model Environment). Both open-source projects include software ranging from low-level components such as buttons and drop-down menus to higher-level applications such as file browsers, games, and a console for those who want a command-line interface. The software handles many basic user interface tasks such as managing windows on the screen and letting users launch programs and switch between them.

One of the significant new features is Kickoff, the revamped start menu. Instead of offering just a hierarchical list of applications, Kickoff offers several other ways to get at programs you might want, including a search bar a la Windows Vista, a list of favorite programs, and a list of recently used programs and documents. It also provides quick access to hard drives, USB drives, and other storage devices.

KDE 4 also has been reworked to take advantage of new glitzy interface possibilities. Windows can be made transparent--a feature for which I personally see almost no utility, but I'll keep an open mind. But there are more useful options, too, such as the ability to quickly show all running widgets or to show all running applications in miniature, features that users of Mac OS X's Dashboard and Expose will recognize.

Perhaps more significant in the long run is some work to make KDE more resolution-independent. Most operating systems and accompanying software assume computer screens have a resolution of something like 96 pixels per inch, but hardware companies are capable of producing much finer resolution.

Theoretically, that could help produce higher-quality text that's less pixilated and easier to read and photos with more detail, but in practice you risk running software that's unusable because of with microscopic type and icons.

Some KDE applications, including the KMines minesweeper game and KPat solitaire card game, now have vector graphics, which scale to any size independent of pixel resolution. It's a small but welcome step.

Another new feature is Dolphin, a new file browser that among other things can present thumbnails of images and let users add captions and star ratings.

A revamped Systems Settings interface resembles Mac OS X's approach, with different options split into related categories.

Cosmetically, KDE has new artwork, including graphical elements such as buttons and window frames, called Oxygen.

And under the covers, there are other changes. A new Phonon library provides audio support to programs, KHTML is available for Web page rendering (it's used by Apple's Safari, too), Trolltech's QT 4 user interface components require less memory, and a package called Solid helps manage hardware details such as power management, wireless networking, removable storage devices, and Bluetooth networking. And for those whose computers have multicore processors, the ThreadWeaver library is designed to make it easier for software to take advantage of hardware abilities.

Update 8 a.m. PST January 14: Sorry, I ran out of invitations, but you can request them from Photophlow's home page. Update 8 a.m. PST January 11: I added links to a couple of helpful videos.

For a Web 2.0 powerhouse, Flickr feels awfully Web 1.0. At least that was my conclusion after spending a few hours in the chat rooms of Photophlow, a start-up that grafts a highly interactive experience on top of Yahoo's photo-sharing Web site.

Flickr deserves credit for pioneering what can be done with photos on the Internet beyond merely displaying photos and albums. Flickr advantages include tags that let members sort and search photos, groups for finding like-minded photographers and sharing photos, and maps to sift through pictures geographically.

Photophlow presents a live chat about Flickr photos that members select.

(Credit: Photophlow)

But Photophlow, which presents a chat room interface to the act of browsing Flickr, makes all those interfaces seem static. For me, the site felt like wandering through a museum with a group of new acquaintances, commenting on pictures as we went from room to room. And some of the rooms featured our own pictures.

The site is invitation-only right now so that Oortle, the start-up behind Photophlow, can keep up with growth. I ran out of invitations, but you also can request one at the site, which is how I got in.

I'm not the only person who's favorably impressed.

'A comfy coffee lounge'
"It really changes the way I use Flickr," said Alex Almeida, who publishes the Phat Photographer blog, who described Photophlow with a different metaphor. With its instant interaction, "it really is like a comfy coffee lounge with a big shoe box of photos where people can chat comfortably and pull any of those photos out of the shoebox and discuss them."

... Read more

There are two dominant software projects that provide Linux with a graphical user interface, but only one of them will get long-term support in Ubuntu's next version of the open-source operating system.

GNOME, the default user interface for Ubuntu, will receive the support, but KDE won't. The reason, according to Canonical, which sponsors Ubuntu and is trying to make a business of selling the support contracts, is simply that KDE is at an awkward transitional period between two versions, the old-line 3.5 and the imminent and significantly different 4.0.

Developer interest is focused on KDE 4.0, but it's not mature enough yet to use in the next KDE-based variation of Ubuntu, called Kubuntu, Scott James Remnant, leader of the Ubuntu Desktop team, said in an explanation to a Kubuntu mailing list. But most Kubuntu developers adding features "upstream" of today's products are focused on KDE 4.0, meaning that it's risky to release a long-term support version based on 3.5.

"Given the attention being paid to KDE 4, it is difficult to believe that this will not be the preferred release in three years' time," Remnant said. "The KDE upstream position appears clear: KDE 4 is the focus of developer attention; KDE 3.5 will be supported as long as KDE 4 isn't suitable for support."

Even though I'm among those who prefer KDE overall, I think Canonical's decision is sensible under the circumstances. And maybe, if we're lucky, this choice will be one small step toward moving beyond the problem that there have to be different Ubuntu flavors with different user interfaces in the first place. But more on that later.

Ubuntu 8.04, aka "Hardy Heron" and due in April 2008, will become the second version of Ubuntu Linux to receive Canonical's long-term support (LTS) designation. Most Ubuntu versions are supported for 18 months, but LTS products are supported for three years for desktop machines and five years for servers.

GNOME-based Ubuntu more popular
GNOME is dominant among Ubuntu users, accounting for about two-thirds of Ubuntu downloads, according to Canonical Chief Executive Mark Shuttleworth.

The remaining third using KDE are a sizable minority, though, and Shuttleworth has taken pains to reassure them that KDE is a priority. Notably, in 2006, Shuttleworth became the first KDE "patron". He's since been joined by four other patron-level KDE sponsors.

Canonical's commercial interests aren't always aligned with community programming-project priorities, Remnant said.

"LTS' is a commercial-support commitment provided by Canonical, who shoulders the financial and administrative burden of doing so; as such, it is entirely their decision as to whether or not they provide that support for a particular release," Remnant said. "It is difficult for this decision to be made by the community because the community's stake in Kubuntu is one of personal achievement and pride, whereas Canonical's is financial and of commercial commitments."

One Kubuntu community member, Juan Carlos Torres, said on his blog that he isn't terribly happy with the decision, but he urged programmers to channel their energies into improving Kubuntu based on KDE 4.0.

"Kubuntu doesn't have the manpower to aggressively maintain two KDE versions. With this, we can focus our efforts on KDE 4 (and migrating KDE 3 utilities to KDE 4)," he said. "As Kubuntu shifts its gears towards KDE 4, we need as many hands as we can get."

KDE 4.0 is due to be released January 18 at the Google campus in Mountain View, Calif.

Wasted energy
I see this GNOME-KDE desktop interface split as a terrible waste of energy. It's based more a historical licensing artifact rather than on some engineering breakthrough.

KDE had the early lead among Linux users, compared to Unix interface predecessors such as CDE (Common Desktop Environment), but Miguel de Icaza, among others, started the GNOME project because of open-source licensing concerns regarding a collection of KDE user interface elements called Qt. By the time Qt developer Trolltech liberalized the license terms, GNOME had taken root, with support from companies such as Red Hat.

Now Linux users are stuck not just with two user interfaces, but often two sets of accompanying control panels, music players, modem-dialing utilities, command-line interface consoles, Web browsers, and more.

That's a lot of duplicative work for programmers, but there are other repercussions. Software companies have to decide whether to build their software using Qt or the GNOME analog, GTK+. Linux distributions that ship with both are bulkier, and running both takes up more memory, as multiple libraries are loaded into RAM. New Linux users are faced with confusing inconsistencies.

Open-source fans have long argued that there's healthy competition between KDE and GNOME. That's probably true, to an extent, but I don't see the differences as particularly profound; even my allegiance to KDE is pretty thin. Frankly, the more interesting rival is XFCE (used in an Ubuntu variant called Xubuntu), a spartan, utilitarian interface that forsakes glitz in favor of working on machines without vast quantities of memory, and the latest processors and graphics chips.

But the real competition here is with Microsoft Windows and Apple Mac OS X. All this overlapping work on KDE and GNOME could be put to better use, matching or beating the innovation and performance of proprietary operating-system interfaces.

Different sound frequencies indicate where on the screen a user is blowing.

(Credit: Georgia Tech)

Perhaps huffing at your computer might get you somewhere if research at the Georgia Institute of Technology comes to fruition.

Shwetak Patel and Gregory Abowd from Georgia Tech have published a paper that describes how to use a computer microphone to determine where on a screen a person is blowing. The technique, which they call BLUI for Blowable and Localized User Interaction, can distinguish between the different sounds air makes depending on where the breath is directed.

"BLUI supports blowing at a laptop or computer screen to directly control specific parts of an interactive application, such as blowing at a button to activate it," the researchers said in their paper (click for PDF). The technique requires a period of "training" to calibrate the system--blowing on each region of the screen for 3 to 5 seconds.

The hands-free user interface approach could be useful for situations where a person's hands are busy, or for people who can't control computers with their hands or arms in situations where speech control is impractical. Although speech "is reasonable for complicated or command-based tasks, it is not well-suited for direct, low-level controls such as scrolling, button pressing, or selection," the researchers said.

Of course, the resolution isn't as fine as a mouse pointer.

The accuracy was 100 percent when dividing a laptop screen in to a nine-rectangle region. It dropped to 96 percent for 16 regions, 80 percent for 25 regions, and 62 percent for 36 regions.

The technique also could be used for games such as a basic one in which users blow out virtual candles, shown in the YouTube video above.

(Via John Nack.)

Adobe Systems wants to transform its flagship Photoshop software with an interface customized to the task at hand, a potentially radical revamp for software whose power today is hidden behind hundreds of menu options.

A new user interface will help Photoshop become "everything you need, nothing you don't," said Photoshop product manager John Nack, describing aspirations for the Photoshop overhaul on his blog Monday.

"We must make Photoshop dramatically more configurable," Nack said. "Presenting the same user experience to a photographer as we do to a radiologist, as to a Web designer, as to a prepress guy, is kind of absurd...With the power of customizability, we can present solutions via task-oriented workspaces," Nack said.

Adobe Photoshop Lightroom offers a very task-specific interface, with editing modules changing according to broad categories of work.

(Credit: Adobe Systems)

In comparison, Photoshop today is unwelcoming and unhelpful. "Today, if a user walks up to Photoshop and says, 'What do I do?' the app kind of shrugs, stubs out a cigarette, and says, 'I dunno--you tell me.' That's not real cool, and we can do better," Nack said.

A new Photoshop approach could let new users get started faster, help Adobe phase out old features, and energize Adobe programmers, he suggested. But such changes are fraught with peril, too: users can be confused or alienated, automated work processes can be broken, and some strong points can be weakened.

Photoshop's general-purpose value
One skeptic is Mark Rolston, chief creative officer for design consultancy Frog Design and a Photoshop user since before it was Photoshop 1.0. Photoshop fundamentally is an all-purpose tool, and tailoring it to be more task-specific could undermine that general usefulness, he said.

"Its generalized approach of being a toolset is the one thing that's made it popular...You can't be task-specific in a professional application like this," Rolston said.

But Nack said in an interview that Adobe sees the value of Photoshop as a general-purpose tool and won't throw the baby out with the bathwater.

"Can Photoshop be totally general and totally focused at the same time? I think it can, through modularity and extensibility," Nack said.

Modularity raises another risk: multiple versions of Photoshop with different features. With its CS3 versions released this year, Adobe has opened that door by reserving some high-end or specialized abilities for the more expensive Extended edition.

There, too, though, Adobe is aware of possible problems. "There's a great deal of value in Photoshop being an industry standard," Nack said. "We therefore have to be careful about taking steps that would balkanize Photoshop...We don't want to get into a state of where people can't readily exchange files because they're using a dozen splintered versions--Photoshop for Web, Photoshop for Medical, Photoshop for Basket Weaving, etc."

Don't expect faster performance along with a cleaner design, though, Nack said. "I think the benefit will be more in users' perception than in saved clock cycles," he said.

The Lightroom lesson
One interface model Adobe no doubt factors into its deliberations is its new Photoshop Lightroom software released earlier this year. Lightroom, which is tailored specifically for editing, cataloging, presenting and printing raw images from higher-end digital cameras, has a very task-specific interface.

For example, Lightroom presents four major panes for different broad categories of work, and optional panels surrounding a central image present different options according to which mode the user selects. But Lightroom is a much more focused tool.

"Lightroom has shown that presenting just the tools needed at any moment can help users manage a complex workflow," Nack said. "With Photoshop, we'll find a way to offer that approach without losing the generality that has let people push the application in so many unexpected ways."

Another major experiment in user interface has been Microsoft's Office 2007, which added a "ribbon" that presents different options to the user according to what tasks are possible. Microsoft Word and the Office packages suffer a similar plight as Photoshop: although most users probably only use a tiny fraction of the software's features, the collective user base needs all of them.

Laying the groundwork
Nack knows that coaxing users into alignment with Adobe's vision will be one hard part of the change. But he--along with Photoshop programmers--has been working for months to make the change palatable.

In May 2006 came some cajoling. "If you could take away the pain that comes with a large and growing feature set, yet keep its benefits, would it cool the critics out?" he asked in a May 2006 posting. "We need your permission to take Photoshop in new directions, to add features that will blow people's heads clean off."

Then, a year ago, Nack grew a bit sterner, saying Photoshop users bear some responsibility for the software's sprawling state.

"We can add things, but we can never take them away. When we decided to stop maintaining the archaic, seldom-used 3D Transform filter, we made it optional content (not disabled, just moved). The tech support boards lit up with all kinds of complaints," he said.

Making Photoshop better for users also could make it a better project within Adobe, he indicated. "No one wants to work with--or work on--some shambling, bloated monster of a program."

Adobe coders have been working to make Photoshop to enable the modular, adaptable vision with features such as customizable menus and shortcuts, or workspaces that let users save particular configurations of editing palettes.

Also behind the scenes, Adobe has been working to make Photoshop more modular. Several modules--among them the type engine, 3D tools, the Camera Raw system, and the "Save for Web" process--only load into memory when called upon, he said.

"We're already making the code modular so that people aren't running what they don't need," Nack said. "Now we need to follow up at the user experience level, so that people don't have to wade through anything not geared towards the task at hand."

What do you get when you cross a Firefox with a chameleon?

An open-source Web browser whose user interface is adapted to the look of the operating system it's running on. One change planned for the upcoming Firefox version 3, code-named Gran Paradiso, is this more native appearance.

"The Web browser is an incredibly central piece of the user's operating system, and we don't want the user's initial reaction to be that they have modified their computer to add some type of strange, foreign application," said Mozilla interface designer Alex Faaborg in a blog posting last week. "Mozilla's user experience team literally wants to do a better job of visually integrating with Windows than IE, and a better job of visually integrating with OS X than Safari. I don't know if we will be able to pull that off, but that's the goal."

Firefox will have different looks for Windows XP and Windows Vista, but the much broader diversity of Linux interface options makes it more challenging. Red Hat, Suse and Ubuntu all look different, just to name three popular versions, and as a further complication, each is available with the KDE and GNOME graphical interfaces.

"We still aren't sure what the best way to visually integrate with Linux is, given the number of different distributions," Faaborg said. He also referred those interested in the issue to related posts by lead Firefox engineer Mike Connor and Firefox user experience leader Mike Beltzner, who detailed some of the problems.

The Firefox native-look approach goes counter to one trend.

There was a time when user interface guidelines for operating systems were rigorously set. Buttons and menus and scroll bars had to look and behave in a certain way so computer users would know what to expect and have an easier time figuring out how to accomplish what they wanted.

But user interfaces today are exploding in diversity. Years ago, software such as media players forsook a traditional appearance in favor of an interface that looks like a car radio. Followed suit are a profusion of smaller programs called widgets and gadgets such as clocks or weather monitors. And rich Internet applications, which run in Web browsers, are designed to look the same across operating systems.

Software that's adapted for multiple operating systems always faces something of an identity crisis. Should the software look the same from one operating system to the next, providing a familiar look regardless of where it's running, or should it fit in with the local system?

Faaborg said he believes people will imprint more on what Firefox can do than on how exactly it looks.

"I personally think Firefox has in the past established its identity through interactions as opposed to the visual design of the interface itself," he said, citing for example people's recognition of the tabbed browser windows in Firefox 1.0. And users similarly might identify in Firefox 3 with a feature that lets them navigate to a Web page by typing some part of its name in the location bar, with Firefox suggesting full links based on bookmarks and previous pages visited.

"When you think about the difference between Firefox 2 and 3, or the difference between Firefox and other Web browsers, I think it is streamlined interactions like this, or one-click bookmarking, that are likely to spring to mind, as opposed to the application's unique visual style," Faaborg said.