Arc Browser is a Revelation on the Mac (Premium)

In 2001, Microsoft released Internet Explorer (IE) 6 alongside Windows XP, cementing its dominance of the web browser market, with usage peaking north of 90 percent worldwide less than two years later. And then Microsoft, inexplicably and without explanation, stopped updating IE.

Why it did so was a mystery at the time, and it remained so for years. Lacking other evidence, many attributed the sudden lack of focus to laziness: Having dominated web browsers as thoroughly as it had PC operating systems and office productivity suites, the theory went, Microsoft had simply stopped trying. There was no need to innovate in a market in which it was the only relevant player.

It’s a good story. But that’s not what happened.

Instead, Microsoft had directed the IE team to work on platform advancements for Longhorn, the planned successor to Windows XP, that would embrace and extend the web into more powerful native capabilities. This project was supposed to take a few years and culminate in even deeper integration between Windows and the browser. But Longhorn infamously sprawled out of control and was only released in diminished form several years later as Windows Vista.

Nevertheless, Microsoft’s neglect of IE was real. And by turning its Eye of Sauron elsewhere for so long, Microsoft gave a new generation of web browser competitors—most notably Mozilla Firefox at first, but then Google Chrome—the opening they needed. By 2010, IE’s share of the market had fallen below 50 percent, and then Chrome surpassed IE as the most-used web browser in 2012. It hasn’t looked back since, though the rise of Safari on Mac—no doubt inspired by Apple previously being forced to make IE the default on its own desktop platform—is also notable.

Like IE during the heady “browser wars” years of the late 1990s, Firefox, Chrome, and other modern web browsers achieved their respective levels of success via innovations that improved functionality and performance while taking advantage of the ever-growing sophistication of the underlying web platform. This makes sense at a high level: When products are new, innovation often comes quickly, and simple but limited products can evolve into more full-featured solutions that address more needs.

Flash forward to today, and browsers are more relevant and important than ever before. They are “super apps” that are used by more people, on more devices, than any other software. That’s been true on Windows for almost 20 years.

It is odd to me, then, that we don’t see more innovation in this space: Today’s mainstream web browsers—not just major players like Chrome, Safari, Firefox, and Microsoft Edge, but also less common solutions like Brave, DuckDuckGo, Opera, Vivaldi, and so on—are nearly identical from a functional standpoint, with the same basic user experience. The differences are minor, and moving between web browsers is seamless and drama-free.

On the surface, this situation doesn’t seem to mirror what happened to Internet Explorer in the early 2000s. After all, the makers of today’s browsers are actively updating these products, and on a much more aggressive schedule than ever before. But the lack of user experience innovation we’re now experiencing very much mirrors the IE of two decades ago. For all the updating, there’s very little of substance. These companies are not asking hard questions about how people use their products. They’re either maintaining their market positions or competing at a macro level, by emphasizing a certain strategic focus like AI (Microsoft and Edge) or privacy and security (Brave).

In personal technology, true innovation—disruption—requires an upstart, an outsider. A company with no market position to protect, no history to respect. The most famously successful example of this is Apple, which reimagined the smartphone so thoroughly with the iPhone that it established what is essentially a new market for a new product, and, in time, a sprawling new ecosystem. The iPhone was so disruptive that it killed off every smartphone platform that preceded it, and its only direct competition today is Android, which was (re)created in its wake as a direct copy.

Do web browsers need to be disrupted?

It’s a good question and a worthy debate. I’ve long felt that Windows should be left alone, so to speak, so that it can ride out its remaining years with dignity while the dust settles from the explosion triggered by the iPhone and the aftershocks of Apple’s next innovation, the iPad. And yet I’m arguing here that web browsers, which first put Microsoft and Windows on notice as a potential extinction-level event in the mid-1990s, still warrant attention.

Here’s my thinking, which I’ll condense down to a Windows-centric worldview for simplicity’s sake. We in the Windows community have spent the past 25 years fretting whether Microsoft would ever release a new native apps platform that could replace Win32 and bring Windows into the 21st century. And today, Microsoft has seemingly given up after multiple failures and is now engaging in the marketing version of making lemonade: Windows runs all kinds of apps, and Microsoft doesn’t care how you make them. There will never be a new native apps platform for Windows.

But that’s because there doesn’t need to be: The web platform that Microsoft almost embraced in the late 1990s before insular decision makers pulled the plug, forever changing history, has matured to the point where it’s ideal for desktop apps and this new era of heterogeneous computing. That Microsoft fought a futile, losing battle against this reality for over 20 years is now a matter for the history books. But the real proof point here, as it’s been for ages, is in what Microsoft does, not what it says. And Microsoft today is creating web apps.

Think about it. All the recent new Microsoft 365 apps—what we used to call Microsoft Office apps–are web apps. Microsoft Teams, Microsoft Loop, and the new Outlook are all built with web technologies. Microsoft Copilot, implemented first on the web in Bing, and now integrated into the Microsoft Edge web browser, has been implemented in Windows 11 as a web app. And even the legacy Office apps—Word, Excel, PowerPoint, and so on—long ago moved to a web-based extensibility model, which is what allowed Microsoft to bring Copilot to Microsoft 365 so quickly.

And it’s not just Office/Microsoft 365. Visual Studio Code and Microsoft Clipchamp are both cross-platform web apps. Even Microsoft’s decision to recreate its Edge web browser from scratch and use the Chromium code base as its foundation was an explicit confirmation about how important it was for the software giant to get it right. The web isn’t just the future, it’s the present. It’s everything, and in this case, not having a mobile platform helped Microsoft get there when it did. Apple has still not gotten this message, and it’s fighting the inevitable tooth and nail to protect its native mobile apps platform. Just as Microsoft did on the desktop, with Windows, for far too long. Again, innovation isn’t going to come from those with positions to protect.

Today, the three primary personal technology platform makers—Apple, Google, and Microsoft—have all embraced web technologies deeply. Each makes its own browser, most obviously and most notably. Google and Microsoft both allow developers to publish web apps in their app stores alongside native apps. And all three treat web apps as first-class citizens on the desktop. (Apple is an outlier on mobile because of its protectionist strategies.) Google has even created an operating system from its browser, ChromeOS, in effect realizing the early promise of Netscape in the 1990s. And while that platform started off a bit wobbly, it, like the web, has matured greatly. It’s not your father’s ChromeOS anymore.

With all that as a backdrop, I will reiterate my curiosity that the web browsers we use are all basically the same from a user experience perspective. The same as each other, and the same as the web browsers we used 25 years ago. Browser makers compete on what are ultimately subtle differences. And enthusiasts spin their wheels debating whether consolidating on rendering engines is a monoculture (it isn’t) while ignoring more profound questions.

Why aren’t we seeing more innovation in web browsers?

In February, I wrote about the one beacon of light in this market: A tiny startup called The Browser Company created a web browser called Arc that challenges the browser user experience assumptions and traditions that may—or may not—be holding us back. Arc sees the web browser as what it calls an Internet computer that will, in time, consume the desktop functionality of the platforms on which it runs—multitasking, app and window management, and so on—because the web is the platform. I also called on Mozilla, the idealistic but struggling makers of Firefox, to consider purchasing Arc and its makers before they’re snuffed out by Apple, Google, and/or Microsoft.

At the time I wrote that February article, I’d only used Arc on Windows. But this version of the product is incomplete and lacks many of the best features that are available in the more easily obtainable and mature Mac version of the browser. And so I was sure to install and use Arc on the MacBook Air when it arrived a few years ago.

And I am conflicted.

On the one hand, Arc for Mac is like a fully realized Renaissance masterpiece: I could immediately see that the team that built it thoroughly examined not just what a web browser does, but how it does those things. And that they were not afraid to evolve or reject familiar user experiences when they found a better way. This is a far more complete app than its Windows sibling, which resembles the Mac version only as much as a LEGO version of a vehicle might resemble its real-world inspiration.

On the other hand, Arc is still difficult to learn and master, and in a somewhat ironic fashion, its additional functionality on the Mac makes that even more jarring in some ways. This experience has shone a new light on the confusion I felt when I tried the Windows version back in January. It’s one thing to “know” that the Mac version of Arc is more full-featured than the Windows version, but it’s another thing entirely to experience it.

As before, I feel that certain users, certain types of users, will immediately grok Arc and see the value of its innovations. But I also see that many others, many mainstream users, will just be confused by this app. And that this confusion will drive them back to a more comfortable and familiar browser, like Chrome or Safari.

This is a problem. It’s also something I can relate to: In using the MacBook Air for the past few weeks, I have specifically worked to overcome the blockers that I’ve struggled with in the past when using non-Windows desktop platforms like macOS, ChromeOS, and Linux. I adapted to its confusing, inconsistent, and occasionally brilliant but different multitasking features, for example. (And did so without installing any third-party utilities that might have made this system more familiar.) But I still struggle from time to time. And it wasn’t until yesterday, in a bout of frustration tied to Affinity Photo 2 and my inability to make the Mac version of this app do something I needed, that I finally did what I always do and gave up. I opened up the closest Windows laptop—oh, I have many—fired up the same app, and immediately finished the task I couldn’t complete on the Mac.

Most people, certainly most mainstream users, do not have this patience: I literally struggled with that task for over an hour, determined to figure it out. And why should they? Switching apps is hard enough as it is, but it’s next to impossible when a new app is both unfamiliar and difficult to learn, and there’s a perfectly good alternative you’ve been using for years, just sitting there on the PC (or Mac), mocking you and your silly efforts. It’s a wonder anyone tries at all.

The thing is, I love Arc. And with one major exception that could easily be fixed—or is perhaps already configurable, and I just haven’t found how to do so yet—I feel that this rethinking of the web browser is necessary, timely, and important. In short, the innovation here is real.

To make this case, all I can do is do what others have done before me and try to explain the high-level differences between Arc and whatever browser you use. You may or may not agree with some or all of these changes, may recoil at specific user experiences, or may feel that this is a simply a case of too much, too soon. I get it. As noted, I’m conflicted about this thing as well.

But here’s what’s happening.

Today’s web browsers are fundamentally multi-document apps that support both multiple windows and, in each window, multiple tabs (and, often, a Split View too). Each window and tab you open displays some webpage, typically on a remote server, and these apps all support standardized methods for switching between each of these webpages using the mouse, keyboard shortcuts, touch, or whatever else. Likewise, each window and tabs takes up memory and other resources, and because so many people keep so many tabs open at once, memory-saving features like tab sleeping have become common. (Just writing that, I can see where The Browser Company came up with the notion of the Internet computer: Web browsers, in many ways, are like mini-operating systems as it is.)

From an organizational perspective, you can pin tabs for a sense of persistence, though as many have noticed, if you close a browser window with pinned tabs while another browser window is open, you’ll lose the pinned tabs. (Usually: Safari handles this differently by putting the pinned tabs on new windows, too.) You can also create profiles to separate your work/school and personal browsing.

Like other browsers, Arc lets you display webpages in different views and switch between them using various methods, of course. (Including, yes, a Split View.) It’s just that it does almost everything differently. The Arc user experience is minimalist, with a single sidebar on the left and no traditional browser toolbars. This is non-negotiable, and will be off-putting to those, like me, who experimented with a vertical tabs layouts in browsers in the past and just didn’t like it.

But there’s a reason for this interface: Arc organizes the content you view in the browser into one or more named spaces, which are logical areas, each of which includes unpinned, ad-hoc tabs (like a normal browser), pinned tabs (which are nothing like those in a normal browser), an icon, and a theme. (Arc also supports multiple profiles, but let’s not confuse matters further.) You might make one space for work and one for home. One for each project you’re working on. Whatever.

The Arc sidebar displays one space at a time. But it also displays other items at the top—some basic browsing controls, a search/address bar, pinned extension icons, and favorites (which are very much like pinned tabs in normal browsers)—that persist in the sidebar, no matter which space you’re using. You switch between spaces using touchpad gestures, keyboard shortcuts, or their icons on the bottom of the sidebar.

Inside a space, you can have pinned tabs and unpinned tabs.

Pinned tabs persist, meaning they are permanent and will be there each time you restart the browser. But in addition to the expected functionality, pinned tabs can also be renamed and assigned custom icons. That’s fun.

Unpinned tabs are also treated differently than you’re used to. For one thing, they’re transient: They are auto-archived, by default, after 12 hours, so they disappear from the space never to be seen again. You can change the auto-archive timeframe—to 24 hours, 7 days, or 30 days—but you cannot disable it. But if you’re nervous about losing anything, don’t be (or at least try): Arc’s command bar—described below—can help you find any pages you opened previously. You can, of course, organize tabs you wish to keep around by pinning them in a current or new space. And there’s a handy “Archived Tabs” view in the Arc library, accessible from the bottom of the sidebar as well.

We all work differently, so all I can do here is explain how I typically configure a web browser and how I adapted that to Arc.

I explained the basics in How I Configure Brave: I pin four tabs (Gmail, Google Calendar, Twitter/X, and Mastodon) and keep several other tabs (Thurrott.com, OpenWeb, Thurrott.com threads, Thurrott.com users, Google News, and The Old Reader) in my first browser window and make sure it reloads this configuration when I need to restart the browser.

Recreating this in Arc, I configured the default space (which I imaginatively named Work) with Gmail, Google Calendar, Twitter/X, and Mastodon as favorites, and then I pinned those other pages to the same space. So that’s nice and straightforward, and I took a few seconds to customize the theme color and space name, and enable Arc Sync so that this configuration appears on other computers (including PCs). I also pinned the same extensions I pin in other web browsers—Dashlane, Dark Reader, and Save to Pocket—so they’re visible at the top of the sidebar.

The sidebar can be resized or hidden, and if you choose the latter, you can toggle it back on with a keyboard shortcut or by mousing over to the app’s left edge. This is a particularly nice setting if you use the browser in Full Screen mode on the Mac. But even on Windows, it creates a minimalist look that really appeals to me, and navigating with the keyboard is both seamless and, to me, ideal.

There is a lot more to Arc, but I would like to focus on two key points. Both are tied to the uniqueness of this app. But one is straight-out wonderful while the other is, at least to me, curiously limiting. (It is this second feature where I may be missing something.)

When I want to open a new tab and search or navigate to a specific page, I type Ctrl + T (in Windows, or Cmd + T on the Mac) and start typing. I will also sometimes just want to replace the page I’m currently viewing, in which case I type Alt + D (Windows only) or Ctrl + L (Cmd + L on Mac) to select the address bar, and then I start typing.

In Arc, these keyboard shortcuts both bring up the Arc command bar, a floating toolbar of sorts that resembles (and works like) the command palette in Visual Studio Code or Windows Terminal. The command bar works a lot like the address bar (or search) in your browser, in that it supports auto-complete and understands your browsing history. (Ctrl + T will open a new tab, while Ctrl + L will work with the current tab.)

But the reason I compared it to the command palette in those other apps is that it does so much more, too. Among other things, it integrates deeply with Arc-specific features, so you can do things like move the current tab to pins, favorites, or a new space. It can be used to search the current site. Or to access an extension. And if you enable Arc Max, you can even use it to prompt ChatGPT.

The command bar makes sense because it uses keyboard shortcuts everyone knows, and it provides the same basic functionality. It feels familiar and works as expected, basically. But it’s also nice for power users because it can do so much more too.

The bit I struggle with is tied to multitasking.

With mainstream web browsers, as noted, there are windows and tabs. On Windows, you switch between these windows with Alt + Tab and you switch between tabs in a window using Ctrl + Tab. (The Mac is similar, but inconsistent between apps when it comes to choosing a window. Sometimes Cmd + ` works, sometimes not.) This is straightforward because it’s well-understood.

Aside from its unique tab auto-archive functionality, Arc lets you switch between windows and tabs, of course. Except that it doesn’t, not strictly speaking. That is, if you open Arc from a dead stop, none of the tabs you’ve pinned or favorited are open and so you can’t “tab” to any of them (using Ctrl + Tab). In a mainstream browser, these tabs–pinned tabs and normal tabs—all load with the browser app.

OK, that’s no big deal: I can open each of those pages—the four favorites and the 6 pinned tabs—easily enough on first boot. Except that doing so doesn’t solve the problem: Arc only displays the five most recent tabs when you type Ctrl + Tab. All other browsers display all the tabs.

Hm.

Naturally, I assumed I was missing something. I’ve spent more time than I care to admit scrolling through the app’s settings interface, assuming there’s an option I’m missing. And I’ve Googled it, of course. From what I can tell, this was a design decision, and The Browser Company expects you to use that command bar I just touted to access those other tabs. Or maybe just leave the sidebar open and use the mouse. And … Yeah. I am not OK with that. This should be my choice.

(Like other Chromium web browsers, Arc supports a Task Manager feature, yet another way that browsers resemble operating systems. You can also access this feature from the command bar and then use that to access individual tabs. You know, if you hate yourself.)

One might argue that this feature–it’s called “recent tabs” not “all open tabs”—falls into the same category as the auto-archive tab feature. It’s what a designer might call an opinionated feature. I get that. I just don’t like it. And in the end, I am all about choice. This gets in the way of my needs, which I think usurps the entire point.

Is this a dealbreaker? No. It’s just a curious limitation that may or may not be consistent with some of the app maker’s other design decisions. Just as Arc itself is an innovative new take on web browsers that may or may not be the future. To be fair, I am getting worrying Windows Phone vibes here.

Before moving on, I want to at least mention that there is a lot more to Arc than the basics I’ve described, and I’ve tried to limit the conversation a bit because there’s a real usability hump to get over first. But the more you use this browser, the more you’ll discover; it really does reward you that way. And maybe that’s a good topic for another day.