The more I think about this, the more convinced I am that I want a browser that functions like an IDE.
Multiple panes with tabs, that can be resized either to custom sizes or automatically (binary space partitioning).
It’s too much for a general purpose browser, which is how we’ve got the Chrome, Safari, Firefox we have. But I think the appeal of a more information-dense browser is broader than we think, anyone doing research would benefit from it.
Vivaldi can do tab tiling. Not nearly as customizable as an IDE as you describe, but you can vertically or horizontally tile multiple tabs on the same "page". Only browser I know that has that close of functionality
It’s highly limited compared to the tiling in Vivaldi. You can only tile 2 tabs at a time. Arc also doesn’t seem to have horizontal tabs so you lose 10-20% of your screen space to the vertical tabs bar which gets annoying. I do like Arc but there are some basic things missing that make it hard to use.
Maybe you tried it a while ago? You can tile unlimited tabs either horizontally or vertically (but not both). I've never found any reason to tile any thing vertically anyway, the web just isn't designed for small horizontal windows.
I use an ultrawide 32" screen and even for window tilling I use PaperWM (technically a Hammerspoon script because PaperWM isn't on MacOS). So I only need to navigate windows/tabs in one dimension (horizontally).
Arc does have the benefits of native UI, Vivaldi at least when I tried a couple of months ago was very janky, presumably because the browser chrome is written in React.
Somewhere in this thread people mentioned Stack browser which is actually closer to PaperWM in terms of how it tiles tabs. For both Arc and Vivaldi it's very annoying if you resize (going fullscreen and back) windows with tiles because the size of the tiles gets messed up. Stack just keeps the size and make them scrollable.
You likely know what I'm talking about, tab pinning is broken in all mainstream browsers, it isn't broken in Arc. In addition, you can pin groups of multiple tabs tilled side by side (not very useful without a wide screen, but still)
I'm not the person you asked but I love it. It's actually solved my tab hoarding problem and made my life so much simpler. Caveat: I'm an ADHD sufferer so my experience may not be typical.
I actually have ADHD(I) as well, and have been trying to battle tab/bookmark hoarding myself. It seems Vivaldi is working well, but I was wondering what Arc has to offer.
Early versions of Opera (maybe 20 years ago or so, I don't remember when exactly) had real MDI, where the tabs were subwindows rather than actual tabs. You could arrange them however you wanted, tile, cascade, etc.
For me, it was the pinnacle of browser UI. Everything that came after that (including of course later iterations of Opera, which removed the MDI) has been considerably worse.
I really, really liked a lot of Opera's ideas. I think someone on that team really understood good UI design principles, at least at the feature level. If memory serves it included an IRC client at least, and maybe even an FTP client? Can't remember about the latter, but definitely the former.
Perhaps a bit "bloated" for its time, but it was always pretty sleek to use. In fact I don't exactly remember why it never became mainstream. Maybe it had too many quirks or lagged behind others in terms of the new features, which... isn't entirely its own fault.
Looking at it now, it seems to have an identity crisis now. It looks like it has Whatsapp integration, among other things, which is the absolute last thing I'd ever want my browser to have an integration with.
> Looking at it now, it seems to have an identity crisis now.
Opera was sold to a Chinese company in the mid-10s and introduced suspicious stuff like a built-in free VPN. It is basically ignored nowadays.
The original CEO started Vivaldi, which is an excellent browser in terms of features - tiling tabs [0] were a favourite of mine. Unfortunately it is closed-source and chatty [1], but I still like on devices where privacy is not a concern.
Yuck, what a disgustingly misleading front page then. I installed it today after it was mentioned and played with it and really like it, but this is worse than Ungoogled Chromium. I guess I won't be switching to it. Thanks for the heads up.
Indeed, it did include IRC client, FTP client, email client and download manager with torrent support (not 100% sure if all at the same time, because it has been many years and versions blur in my mind).
The funny thing is that even though it included a lot of features I didn't use (for example, I never used the email client, not because it was bad, but I didn't use POP3/IMAP in general during that period), it was still much more efficient than all competitors in terms of speed, memory usage and even download size (at some point it boasted to be the only browser that fit in a floppy disk, although of course not for long).
I think the reason why it never became mainstream is that it was a paid product for a long time, while all other browsers were free. It eventually became free, but by then, competitors were too entrenched.
I also dislike Whatsapp but depending on country, it might be a necessary evil. In Spain for example, not having Whatsapp practically means forgoing any socialization with non-nerds. It's so ingrained in daily life that being without it is almost unthinkable for most people. Anyway, the current Opera doesn't have almost anything in common with the original one. Different engine, company, team, vision and even continent (owned by a Chinese company), only the brand and a thin sliver of identity remains.
(The sad thing is that I use it, because I still find it to be the least bad option for my needs...).
I remember reading about Opera back in the day, the reason it was not bloated was that all these features were packaged as separate dlls and only loaded when if you tried to actually use them
Also for a long time one of their main marketing points was that it had really small binary size so you could download it fast. They really spent a lot of engineering in making it performant
Opera got sold and the people who used to work there now continue their work at Vivaldi. Love their browser: strong focus on keyboard users, fast, privacy oriented.
it had irc client, rss feed reader, mail client all builtin. Its local caches were portable. Had shortcuts to hide images in current page, disable js (imagine early 2000s web ui with gifs everywhere) etc. Early opera truly was the best browser that ever.
And not an IDE like VSCode, but a fully customizable IDE.
It's so frustrating to yearn for basic features in VSCode only to be ignored for years. For example, on macOS, a good amount of space is wasted because we can't hide the menu/title bar. We can't customize tabs either. No vertical tabs like Edge. No tabs height customization. It's either their way or the highway.
A gripe with Brave: Why doesn't the mouse-over feature for the vertical tabs doesn't work if the cursor is at the very left of the screen? And why can't we move the vertical bar to the right side of screen?
VS Code has always had vertical tabs. It actually launched without horizontal tabs support and only added it due to community pressure. The reason was that VS Code/Monaco was originally designed to run in the browser, and the VS Code team wanted to avoid having a "tabs within tabs" UI.
> For example, on macOS, a good amount of space is wasted because we can't hide the menu/title bar.
It was possible with extensions like titleless and monkey patch, but Microsoft did something recently that made these extensions not work on newer versions.
>It's so frustrating to yearn for basic features in VSCode only to be ignored for years
What is a "basic feature" to you is apparently not that to most people, certainly not to me. Just how important could it be to resize the tab height? And what would vertical tabs do that the open editors panel doesn't? I also don't follow regarding the menu bar -- that's the way all apps behave on MacOS? You can put it into fullscreen mode if you want.
Do you actually need tabs though? First thing I do after installing VSCode is usually getting rid of the tabs and just use the file tree and "opening files" pane (recently got rid of that too because you can just invoke its popup with a hotkey).
What annoys me more is that stupid minimap that somehow has become an enabled by default "feature" on modern editors.
VS Code has a vertical documentlist, and if you want different styling, you can build an extension for this. No reason to have a duplicate feature out-of-the-box.
That's such a wasted design opportunity - give the users freedom in using a proper text editing tool instead of whatever awfulness the OS designers got stuck in decade(s) ago
What are some examples? I finally got around to using a tiling window manager this week and have been loving it, but would love a browser without a status bar to go with it
I actually want a windowing system that provides the tree tab bar instead. I feel the paradigm of having windows around have been stuck for 40yrs and that the only major mainstream innovation is just manually tiling windows.
KDE has allowed grouping windows as tabs, but that never really took off. There's a vast amount of tiling WMs, but none of them are truly popular, and I also don't want to open a terminal, or worse, emacs, to deal with wifi/bluetooth/thumbdrives just to showoff how unnecessarily awkward some things get to be when you don't run some mainstream desktop.
I’ve been using pop os for the past few months and I’ve fallen in love with its snapping window manager.
I have a 4K screen plus the laptop (xps 13).
I can have two browsers open, then my ide. And then terminal on the laptop screen.
And I can easily slide the divider between the ide and browsers as necessary based on what requires the most space at the time.
Only issues I’ve run into is being able to move multiple stacked windows to a new location, all at once. I probably just haven’t found the right shortcut.
It’s not exactly what you’re talking about, as the browsers are independent, but I’d be keen to see new browsing options like that.
Sway (and i3?) actually has the notion of containers, which can contain multiple windows and can be selected and manipulated as a unit.
It works nicely. The only thing that I haven't automated yet is that I have a rather specific workspace layout on my 4k screen and converting it from/to a laptop-screen workspace when unplugged takes a few steps.
> The more I think about this, the more convinced I am that I want a browser that functions like an IDE.
Yep. That is exactly how I think about it as well.
The browser and the IDE are effectively the same metaphor of a "super-app", just different security assumptions about where the information resides, how much it can be trusted and what can be done with it, how active the user input etc. In systems such as Firefox OS or ChromeOs the boundaries dissolve further.
Thinking about open source desktops and their application ecosystems, instead of countless independent efforts that waste precious talent in incompatible replication there could be some sort of architecture where lower level apps (editors, file/tab/bookmark/tag managers, shells, webview or spreadsheet renderers etc) could be recombined in various ways in the browser/IDE (or run as standalone apps).
To save you time trying it out, their implementation of tiling is better than that of Arc. Unfortunately it suffers the same problem most mainstream browsers do: broken pinned tabs, and there is no pinned tab groups like Arc
Or, getting rid of tabs and going back to individual browser windows, but with some mechanism to visualise and navigate between them better on the desktop (e.g. window shrinks to 30% of full size on loss focus and stays to a side of the desk top until restored). I remember managing a fairly large number of Internet Explorer windows on Windows 98.
I thought about that too, the ability to group windows into tabs*, regardless of application is better managed by the desktop environment, rather than application.
UniPress Emacs for NeWS in 1988: Scriptable GUI, tabbed windows, pie menus, hypermedia authoring tool for HyperTIES browser.
Emacs served as an IDE with tabbed window and pie menus, for interactively editing, viewing, and navigating HyperTIES markup language documents, graphics, and interactive PostScript "applets".
HyperTIES browser and Gosling Emacs authoring tool with pie menus on the NeWS window system
>HyperTIES is an early hypermedia browser developed under the direction of Dr. Ben Shneiderman at the University of Maryland Human Computer Interaction Lab. This screen snapshot shows the HyperTIES authoring tool (built with UniPress's Gosling Emacs text editor, written in MockLisp) and browser (built with the NeWS window system, written in PostScript, C and Forth). The tabbed windows and pie menu reusable components were developed by Don Hopkins, who also developed the NeWS Emacs (NeMACS) and HyperTIES user interfaces. (Sorry about the quality -- this is a scan of an old screen dump printed by a laser printer.)
Emacs provides the pie menus you see popped up in the illustration (Articulate, Edit, New (Storyboard, Link, Picture, Target), Define) that control the HyperTIES browser from the custom text editing mode of HyperTIES storyboards (like web pages), which the HyperTIES browser (in the background, which emacs controls in a sub-process) formats and displays. HyperTIES also uses pie menus for navigation and in interactive "applets" programmed in PostScript.
How about taking another step back, and having a browser that functions as a scriptable user-customizable editable window manager like HyperCard?
Run the web browser directly on the hardware, and implement the desktop window manager with that, in a way that the whole system is a scriptable integrated development environment. Then it's easy to implement tabbed windows, pie menus, and user-editable HyperCard-like interfaces.
Here's a big step in the right direction of supporting pie menus on the desktop:
Simon Schneegans's "Kando" project aims to implement cross-platform pie menus on the desktop. Simon implemented the beautiful "Pie Fly" Gnome shell extension and WYSIWYG pie menu editor, so I am really looking forward to what he does with other desktop interfaces!
>Kando will be a pie menu for the desktop. It will be highly customizable and will allow you to create your own menus and actions. For instance, you can use it to control your music player, to open your favorite websites or to simulate shortcuts. It will be available for Windows, Linux and maybe macOS.
>Fly-Pie is an extension for GNOME Shell which lets you open marking menus via keyboard shortcuts. And — to the best of my knowledge — it is the first GNOME Shell extension with achievements!
>You can use it to launch applications, simulate hotkeys, open URLs and much more. It features a continuous learning curve which lets you gradually lift-off from a grumpie menu rookie to a snappie menu pielot. (You got it? Like pilot, but with a ). Once you opened a marking menu, you can seamlessly transition between three alternative selection modes:
>Point-and-Click: Select items by clicking on them or anywhere in the corresponding wedges.
>Marking-Mode: Select items by drawing gestures. To do this, click anywhere and drag your mouse. Pausing or making a turn selects the currently dragged item.
>Turbo-Mode: You can also "draw" gestures while holding Ctrl, Shift, or Alt without having to press your mouse button! This is especially useful when you opened the menu with a shortcut involving such a modifier.
Fly-Pie 7: GNOME Shell 40+ and a new WYSIWYG Menu Editor!:
>At Sun we experimented with implementing an X11 window manager in NeWS. We didn't have transparency at the time (1992), but we did support shaped windows!
>The NeWS window manager supported cool stuff (for both X11 and NeWS windows!) like rooms, virtual scrolling desktops, tabbed windows, pie menus, was easily extensible and deeply customisable in PostScript, and ran locally in the window server so it could respond instantly to input events, lock the input queue and provide feedback and manipulate windows immediately without causing any context switches or dealing with asynchronous locking, unlocking and event handling. You'd never lose a keystroke or click when switching between applications, for example.
[...]
And here's how I think you should design a programmable "window manager" these days -- but it would be much more than just a window manager! It would be great for integrating legacy desktop and mobile applications into VR, for example!
>Don asks Peter Korn: Hey I would love to bounce an idea off of you! I didn't realize how much work you've done in accessibility.
>There is a window manager for the Mac called Slate, that is extensible in JavaScript -- it makes a hidden WebView and uses its JS interpreter by extending it with some interfaces to the app to do window management, using the Mac Accessibility API.
>So I wanted to make pie menus for it, and thought of a good approach: make the hidden WebView not so hidden, but in the topmost layer of windows, covering all the screens, with a transparent background, that shows the desktop through anywhere you don't draw html.
>Then just make pie menus with JavaScript, which I've done. Works like a charm!
>THEN the next step I would like to do is this:
>aQuery -- like jQuery, but for selecting, querying and manipulating Mac app user interfaces via the Accessibility framework and protocols.
>So you can write jQuery-like selectors that search for and select Accessibility objects, and then it provides a convenient high level API for doing all kinds of stuff with them. So you can write higher level plugin widgets with aQuery that use HTML with jQuery, or even other types of user interfaces like voice recognition/synthesis, video tracking, augmented reality, web services, etc!
>For example, I want to click on a window and it will dynamically configure jQuery Pie Menus with the commands in the menu of a live Mac app. Or make a hypercard-like user interface builder that lets people drag buttons or commands out of Mac apps into their own stacks, and make special purpose simplified guis for controlling and integrating Mac apps.
>Does that sound crazy? I think it just might work! Implement the aQuery "selector engine" and heavy lifting in Objective C so that it runs really fast, and presents a nice high level useful interface to JavaScript.
>Soon after the invention of the movie camera, there was a "genera" of films that consisted of nothing but pointing a movie camera at a stage, and filming a play in one shot.
>That's the classic example of using a new technology to emulate an old technology, without taking advantage of the unique advantages of the new technology, before the grammar and language of film had been invented.
>Simply projecting desktop user interfaces designed for flat 2D screens and mice into VR is still in the "novelty show" age, like filming staged plays written for a theater, without any editing, shots, or film grammar.
>VR window managers are just a stop-gap backwards-compatibility bridge, while people work on inventing a grammar and language of interactive VR and AR user interfaces, and re-implement all the desktop and mobile applications from the ground up so they're not merely usable but actually enjoyable and aesthetically pleasing to use in VR.
>The current definition of "window manager," especially as it applies to X-Windows desktops, tightly constrains how we think and what we expect of user interface and application design. We need something much more flexible and extensible. Unfortunately X-Windows decades ago rejected the crucially important ideas behind NeWS and AJAX, that the window manager should be open-ended and dynamically extensible with downloadable code, which is the key to making efficient, deeply integrated user interfaces.
>For example, the "Dragon Naturally Speaking" speech synthesis and recognition system has "dragonfly", a Python-based "speech manager" that is capable of hooking into existing unmodified desktop applications, and scripting custom speech based user interfaces.
>Another more ambitious example is Morgan Dixon's work on Prefab, that screen-scrapes the pixels of desktop apps, and uses pattern recognition and composition to remix and modify them. This is like cinematographers finally discovering they can edit films, cut and splice shots together, overlay text and graphics and pictures-in-pictures and adjacent frames. But Prefab isn't built around a scripting language like dragonfly, NeWS or AJAX.
>Here's some stuff I've written about the direction that user interfaces should take to move beyond the antique notion of "window managers", and enables much deeper integration and accessibility and alternative input and output methods.
>Glad to see people are still making better window managers! [...] I think extensibility and accessibility are extremely important for window managers. [...] I'd like to take that idea a lot further, so I wrote up some ideas about programming window management, accessibility, screen scraping, pattern recognition and automation in JavaScript. [...] Check out Morgan Dixon's and James Fogarty's amazing work on user interface customization with Prefab, about which they've published several excellent CHI papers: [...]
>Imagine if every interface was open source. Any of us could modify the software we use every day. Unfortunately, we don't have the source.
>Prefab realizes this vision using only the pixels of everyday interfaces. This video shows the use of Prefab to add new functionality to Adobe Photoshop, Apple iTunes, and Microsoft Windows Media Player. Prefab represents a new approach to deploying HCI research in everyday software, and is also the first step toward a future where anybody can modify any interface.
>Here are some other interesting things related to scriptable window management and accessibility to check out: aQuery -- Like jQuery for Accessibility
>It would also be great to flesh out the accessibility and speech recognition APIs, and make it possible to write all kinds of intelligent application automation and integration scripts, bots, with nice HTML user interfaces in JavaScript. Take a look at what Dragon Naturally Speaking has done with Python:
>I would like to discuss how we could integrate Prefab with a Javascriptable, extensible API like aQuery, so you could write "selectors" that used prefab's pattern recognition techniques, bind those to JavaScript event handlers, and write high level widgets on top of that in JavaScript, and implement the graphical overlays and gui enhancements in HTML/Canvas/etc like I've done with Slate and the WebView overlay.
Vertical tabs without nesting (trees) is such a wasted opportunity. Nesting lets you close/minimise/etc a group of a related tabs, eg. all the search results you opened from a single query. I hope Brave's implementation at least lets you switch tabs using the mouse wheel, which is probably my favorite feature of the Firefox vertical tabs addons.
BTW, I was a TST user for many years, and recently switched to Sidebery, which seems to do the same and more, but is a bit more polished.
The only thing those Firefox extension lack is the ability to hide native tabs (I understand that this is a limitation of Firefox extension APIs). You have to manually modify your userChrome.css to do so.
I swear they said they were going to add hiding the top tab bar back in as a native option back in 2017, but I'm still using the userChrome.css hack today.
Yes, that's what they're talking about. It ends up wasting space to have your vertical tab sidebar without the ability to hide the default tab bar at the top of the window.
I used to think like this but now that I'm no longer a student with too much time on my hand, I find features that allow up to a hundred of tabs detrimental to my productivity. Simply vertical tabs strike the right balance.
Just installed and configured Sidebery and after a few minutes I already feel like it completely outclasses TST for Firefox. Thanks for the recommendation!
Sideberry is fantastic. Not only is it a great replacement for Tree Tabs, I've also found it to be a better alternative of Simple Tab Groups. With `userChrome` entries to auto-hide Sideberry and the native FF tabs, it's improved the web experience on my smallish laptop screen tremendously.
Agreed! Here's my current `userChrome.css` to hide the tab bar when Sidebery is open, IIRC adapted from their repo because theirs didn't work fully for me. Admittedly this is hacked together, it was mostly just trial and error, but it works!
To make it work only when Sidebery is open, make sure you add a preface value in the Sidebery settings and use the same value in the css above; in my case I just use an invisible character. Couple it with Ctrl-E keybind to toggle the Sidebery panel, and it's been a big quality of life improvement for me.
I think the preferences are a bit easier to understand, and I didn't have to install an add-on for an add-on to have mouse wheel support. I'm sure there's more, but it's been a few months and I don't quite remember.
TST remains great, though, very versatile and extensible.
Brave currently offers groups, which can be minimized to achieve some of the effect you mention. But like you, I'm hoping they bring true nesting in the future!
It's strange to me how a key selling point of these vertical tabs is "save on vertical space", but they almost always still include a header bar. You can see in the video they posted that it saves almost no vertical space, while now taking up a massive amount of horizontal space.
> It's strange to me how a key selling point of these vertical tabs is "save on vertical space", but they almost always still include a header bar.
Honestly, that's not the main selling point of vertical tabs for me. Horizontal tabs don't scale for more than a dozen or so, but vertical tabs do. The "wasted" horizontal space is useless for reading most page, but displays enough data for practical navigation, and the vertical organization allows more-developed vertical scrolling technologies to be used with the tab list (e.g. scroll wheels).
> the vertical organization allows more-developed vertical scrolling technologies to be used with the tab list (e.g. scroll wheels).
Firefox also lets you scroll the tabs horizontally. Which I prefer to Brave's approach of simply not displaying the last tabs in the bar if it's too many, which means newly opened tabs simply never show up while the others are shrunk so far they become difficult to hit with the cursor.
You can also remove that unnecessarily large sidebar header that Firefox insists on not letting users disable via normal means with userchrome mods. Some such setups can be seen at https://firefoxcss-store.github.io/.
"tab bar in titlebar" has been the Firefox default for a number of years at this point. If Linux, your distro/packager of choice may have overridden that. The only true exception to this is tile based WMs (which don't have title bars) but in that case it's still above by default.
It's not ridiculous at all if you have to compare two documents side-by-side which is something I do 10 times a day while coding whether its documentation+code or doing a git diff
Meanwhile here I am, using an ultrawide 21:9 monitor and loving it as I can easily split it in two or three columns whenever I need. I don't understand why people like squarish monitors so much, the human FOV is not square.
How is the human FOV relevant? I just want to be able to fit as much vertical content into the visible area as possible. Because almost all documents we use are vertically organized.
FWIW, I'm using a single ultra wide screen and never have any full screen windows, always two side by side. Most Linux DEs and even Windows support this out of the box, for macOS there's spectacle.
That said, working on a web project right now, we _are_ ensuring it looks good when maximised on ultra wide monitors, because we figure people might actually do that. But I'm with you there, for most web content it's ridiculous.
I found that 21:9 works very well for me for work (gaming too, mind you). Sheets can show many rows, I can compare things side by side, I can maximize a console with ridiculously long lines, and I can dock the dev tools in the browser to the side. It's a like two 5:4s, side by side, without the bezels.
I have a 16:9 screen in portrait orientation, excellent for reading/writing etc. One small annoyance of 1080 as horizontal resolution is that it is just slightly too thin for some web layouts.
Sat next to it is a 32" 2560x1440 unit (they have near identical pixel pitch, so I don't have to worry about oddities like differential scaling between screens) that most of the time is split into two halves or ⅔+⅓ (under windows, the FancyZones util in PowerToys is handy for this).
I liked the arrangement so much at home that I've replicated it at my own expense in the office.
Yes, they don't follow their own principles to the logical conclusion
In Vivaldi it's much better - you can get rid of the header bar completely and only have the search/tool bar, styled to be narrower, saves real space
And since horizontal space is cheap in most websites, the latter part is less of an issue (though you still don't need it to be massive)
Arc is way ahead on browser UX. Opening links in Pinned Tabs into a modal over the Pinned Tab content is exactly how I want to use link aggregator sites
The trick is to hide all extra stuff like address bar, tabs, settings, OS related bars, etc. and alonly show the browser stuff upon Ctrl+L.
It's so much nicer, especially with an wm that's just gets out of the way - awesomewm
I struggle finding good page <titles> with the amount of space browsers give you. You should at least be able to see which tab is what if you have 3 for a website.
OmniWeb (from Omnigroup, developers of OmniGraffle, OmniPlan, OmniOutliner et al) was their web browser offering in a particularly fecund period of browser development in the early 2000s, which had vertical tabs. The tabs themselves were little thumbnails of the contents of the web page in question.
I used OmniWeb as my main browser for many years, right up until Firebug and Firefox became the preferred development environment. Great browser, a lot of great UI ideas out of that shop.
OmniWeb was great! I think Opera still did it first, though.
I moved on to Safari after it died, but I missed vertical tabs dearly and still do. Hopefully this idea catches on well enough for the Safari team to finally steal it.
Except that you could actually use it instead of being a legendary project we only heard about but very few people actually used. That makes it all the more painful that it was taken away.
Yes, I remember Opera having it around that same time, and Firefox having a plugin to copy it. I think the Firefox plugin had full "tab tree" support though. I believe it died after one of the many updates that changed how plugins worked, although I haven't checked in a couple of decades to see if something similar has appeared.
I've been using Firefox with Tab Center Reborn, and a config which collapses the tabs so only the favicons are visible. Using Shift-Tab, Ctrl-Shift-Tab and Ctrl-T, Ctrl-W to go around and open/close is so logical for me. First time I heard about Tree style tabs I was so against them because they occupied significant amount of space and I didn't know you could collapse them. I don't see myself ever going back horizontal tabs. Vertical is so much better, especially on smaller screen (like 1366x768), where horizontal space isn't needed that much as vertical is.
Tree Style Tabs work well, because screen estate is not wasted on 16:9 and 16:10 style displays. Web browsing is mostly reading task and because of usability, line length cannot exceed a certain width. Thus, there is more screen estate horizontally around the page than vertically on top and bottom of the page you are reading.
I liked vertical tabs on my main monitor where there is plenty of space. But on the laptop, especially with using split screen i got too crowded. I have since made the tabs completely invisible and just navigate trough them with the Vimium extension.
This has some quirks because of the limitations a browser extension has, but I now much prefer to just use keyboard shortcuts and search to jump to tabs.
There is the issue of pages which are designed only for very wide screens and for mobile. In between screen sizes will cause the main content to be annoyingly skinny, if it isn't totally unreadable. These sites are relatively rare, thankfully.
The more enhancements I see browser vendors making to tabs, the more I think they’re reinventing bookmarks/favorites.
Safari introduced “tab groups” a while back, which to me is an insane mess of complexity. When I think about it though they’re just folders and each tab is a bookmark/favorite.
The vertical tabs in Brave remind me of bookmarks/favorites too.
For me, tab groups are nothing like bookmarks. Tab groups are a way of managing already open tabs in related group; showing and hiding groups as you need them. I use a similar tool in Firefox and I always have tabs related to about 4-5 projects open at once. I use the tab group tool to switch between the tabs by project so I only need to deal with 10-15 tabs at once.
These are tabs that I don’t want to close as they have context and sometimes ongoing edits. It’s a bit like having multiple windows each with tabs but this keeps it together in one window and makes it easier to manager.
Without this, I would be facing 50-75 tabs and trying to figure out which ones were related to the project I needed to work on at the current time.
I don’t follow. Tabs are nothing like favorites. You go to a favorite and open links within it in new tabs. Eventually the tabs will be closed and you’ll often never visit those links ever again, as opposed to a favorite.
As for tab groups, I could never get used to them, but I do use vertical tabs and find them quite useful.
I have so many bookmarks which I have never revisited ever.
What counts isn't some abstract storage concept but the UI and UX to manage them. Bookmarks are this drawer in your basement where you have to actively spend energy to find them. Tree tabs are self organizing. How they are are stored is irrelevant to me.
Arc has "pinned tabs", but not "bookmarks". While they can be used similarly, I find it quite annoying to use the search to pull something out of my "bookmarks", only to have it open a directory hierarchy eight levels deep when I switch to it.
Likewise, if I navigate away from the page on the pinned tab, I lose my "bookmark".
The reason I like it is that he’s unified tabs and bookmarks almost, it’s a great UX that looks like some things you might have seen but still feels different.
He’s also got wicked design sense — but I’m biased so. I also feel like I’m doing him a disservice — he’s got one of those silly lifetime deals going right now.
I'm curious but somewhat discouraged by the lack of a single screenshot. I don't want to install a minimalist browser that won't have extension capability and a lot of other features just to see if I like one thing that it does.
This does look spiffy despite the other limitations. I fully agree that treed history makes far more sense than a flat list and wonder why this has been such a blind spot for other developers. I'll give it a whirl.
Gonna be honest, that website is doing a big disservice for the product. No screenshots and the navigation disappears when I click a link so I have to hit the browser back button to go to the next page.
No screenshots is huge though. That should be front an center for something that is all about the UI experience. Why would I commit $100 dollars to something I can't see.
I can't tell if this is a joke or not. Your comment is completely earnest, but the website seems to be a send up of websites that don't tell you anything about the product.
Would you mind providing just a little more information? This is clearly an issue but I'm not sure what it could be -- I'm not sure sentry is installed on the site either.
What page were you on where you saw that? Were you not able to load the page itself?
If you wouldn't mind opening up the browser console, I'd love to know what went wrong.
I have the same on Mull on Android, just the homepage from clicking the link above. Annoyingly there doesn't seem to be a dev console on mobile Firefox so I can't see what the error is.
It probably happens due to the disabled hardware acceleration (or just WebGL). At least it's happened to me. Not surprising as the page draws a literal 3D horse.
thanks for this debug -- I think you might be right. I've been talking to the dev and he has to fix his sentry setup! This... might be a while. I'll pass this on though
I thought this was dumb until I opened his Twitter and read through it for a bit and saw what his vision for it was and how he's creating it. Now I'm interested.
Needs extension support before it's really viable though.
Why do people need dozens or hundreds of tabs to the point that they need groups for them? We've had bookmarks (and folders for them) for decades... It made way more sense when you could expect tabs to always be loaded, and not behave like another set of bookmarks.
It's more convenient to keep tabs open. With Tree Style Tabs on Firefox and Orion's vertical tabs I can retain the relationship. The browser generally keeps it cached so even if the site goes offline I can read what I need to. If I have a form partially entered, I can resume it. If the page has infinite scroll but doesn't update the URL, I can stay where I was. Some sites tie resources to a particular session so resuming from a bookmark won't get you to the resource anyway.
But, I can also find things faster if they're on a pile on my desk than I can if they're put away in a drawer so I'm sure it's different for everyone.
As for tooling, I think browsers just aren't very good at bookmark management. I find the sidebar for bookmarks borderline useless, but a tree of tabs in the same space is quite efficient for me. The bookmark manager isn't much better. We can use folders, but not really have the rich links the web was designed for. I often have both the HN discussion and article open with either a parent-child or sibling relationship in my vertical tabs. That's not easily replicated with bookmarks.
The other tooling issue is tab management isn't great. Chrome's shrinking tabs is terrible HCI; you can't read them and they become harder to click. Finding a tab can be frustrating. It's often easier to open a new tab to complete a task than go hunting for the one I want. Pinning tabs doesn't work great with multiple windows and seeing just the favicon isn't super helpful if you need to pin two pages from the same domain.
Between the tabs I keep open because that's how my brain works best and between tabs I get stuck with because browsers haven't provided better tooling, it's easy to grow into the hundreds and then periodically cull them.
Bookmarks have management overhead and are too easily forgotten about. For a lot of people tabs serve as a sort of to-do list or with groups, repositories for sites that will briefly be useful for a certain task in the near future. They naturally get closed as they’re done with, eschewing any kind of cleanup step as is necessary with bookmarks.
That sounds like how I use tabs but doesn't explain how people end up with enough tabs open to make them impossible to navigate in a horizontal layout and without grouping them. When I'm done with a tab, I close it. If I'm going back to it later, I leave it and... go back to it later. If I don't, I close it... or I bookmark it. I don't see how you go from using tabs as a short-term "read later" feature to having several dozens of them open at any given time. And while deleting bookmarks takes a second to do, so does going through your tabs to get rid of the ones that haven't been useful in 2 months. Personally, if I run out of horizontal space I consider it a good reminder to clean up, but then I see people with so many tabs that all they see is a ton of icons, and I wonder if they even know that bookmarks exist. That image came to mind as I read some of the other comments under this post, and I feel like we're just reinventing bookmarks for people who don't realize that bookmarks exist (bookmarks have been deemphasized for a while through trends like hiding the bookmarks toolbar by default).
I used that on Opera way back in the day (before the Blink rewrite). Back then there weren't any Firefox extensions that I liked (either too complex or too 'janky', IIRC) so I just gave up on vertical tabs and grouping, but maybe I should look at it again as that was years ago. Thanks!
Because bookmarks are a bad separate UI, the tabs and sessions are right there and retain your currently working scheme, and also don't include something you worked on 10 years ago and saved in the same folder
And the content of tabs is also cached/loaded unlike with bookmarks
I've had vertical tabs in Vivaldi for a little while now. I hope you guys enjoy it over on Brave as well, they change how you use tabs. Strongly recommended.
Vivaldi is great, but I've hidden the tab bar, and use the window panel instead. Then you get a nice tree view of tabs (nested, grouped by domain etc) and windows. Kind of treating my tabs as bookmarks, which is probably why it consumes a lot of memory :-)
I guess the last part is fixed in later chromium by hibernating tabs now?
Never liked vertical tabs. Horses for courses, personal preference and all that. My muscle memory for normal/horizontal tab layouts is so ingrained in me, that using anything different would take on a huge unlearning.
This basically adresses my biggest gripe with tabs in an elegant way:
The primary input mechanism for dealing with a large amount of "list entries" in any GUI software, nowadays is scrolling. While scrolling can be implemented in horizontal tabs, browsers prevented it from being used efficiently. Tabs all have to fit on that bar, and if they don't, for some reason you couldn't scroll but had to click little buttons, or if you could scoll, it wouldve been a bad UX, since tabs horizontal scrolling goes too fast for the user, or takes too long.
Vertical tabs. Nice. From now on, any browser not offering this feature, shouldn't even bother applying. Nicely done Brave!
Vertical tabs solve a core issue in horizontal tabs. Tabs are inherently 2D: the number of tabs vs the length of their titles. Horizontal tabs can scale only to one dimension. You can choose to either see many of your tabs at once, but not be able to read any of the titles, or be able to read the titles but have to scroll absurd distances just to view what tabs are open.
Vertical tabs give you a 2-dimensional view of your tabs. You can see many of your tabs and be able to read as much of the titles you want by scaling the horizontal size of the area independently.
I think it's brilliant. I also don't see how browsers could "fix" horizontal tabs without resorting to taking away valuable vertical space from the content that is the primary purpose of browsers.
Window frames should allow any tab on any edge, because it's a useful method of organizing the tabs!
You should be able to easily drag any tab to the top, bottom, left, or right edge of the window, and pop up pie menus on the tab to perform window management (and custom) commands on the tab or all tabs on the same edge, including covered-up windows.
For example: Lots of "todo" tabs on the left, "done" tabs on the right, a few "in progress" tabs along the top, "problem"
tabs along the bottom, so you can drag tabs between "todo" to "in progress", "problem", or "done", or select "close all left tabs" from a pie menu to clear out the "done" tasks, to easily manage your tasks with tabs.
You could even partition edges into different groups of tabs, if you wanted. And manage multiple tiled windows inside the frame. A pie menu on the tab will let you open it up to full frame, split to bottom, split to top (with the obvious directions).
Any sub-dialogs a window opens up could have their tabs laid out in a stack along a certain edge from a certain corner.
You could designate edges (or sections of adjacent edges corresponding to tiled windows) to show the tabs of sub-windows of the selected window (to implement tree browsers).
The app or document in the frame should be able to export custom commands that appear on the tab pie menu, so you can do stuff with the window without bringing it to the top or disturbing the currently displayed windows.
The point is that the user should be able to decide what tabs to put where for what reason, not that one edge is the right edge and all other edges or wrong, or using more than one edge at a time is wrong.
What I've learned trying to analyse my preference for vertical tabs is that it's not so much about saving horizontal space vs vertical space. It's more just about showing more information on screen before needing to scroll.
With regular horizontal tabs you either squeeze all tabs to fit the window width, in which case you stop being able to read page titles at some point. Or you have a horizontal scroll on the tab bar which also means you can only read so many titles on screen (probably fewer) before needing to scroll.
The vertical tabs switch this up by being able to show a larger number of tab titles before you have to scroll down. Of course this depends on the width of your vertical tab sidebar but having word wrap or configurable height on horizontal tab bars isn't ever presented as an option so can't compare.
Another clue to understanding my preference was when I realised I actually really like being able to have multiple rows of horizontal tabs vs a horizontally scrolling tab bar. So I think it just came down to the density of information visually available before having to do any scrolling.
I do think it all just comes down to personal preferences and muscle memory and individual use cases (for example, some people like having 50+ tabs open, others don't). So I am very hesitant to call one approach better. I just don't understand why we can't have both options more available everywhere.
The crypto wallet video ad in the middle of the article is intrusive and frankly infuriating, especially when there is no video illustrating the lifecycle of the new vertical tabs.
That's Brave for you. The founder is a crypto nut, and the browser is adware (shows ads itself instead of merely rendering ads on third party websites).
Because you opted out. The browser is ad-supported though, their whole business model is supposedly built on the users opting in to view strictly controlled ads for cryptocurrency rewards, at the same time stripping the existing ads from the sites, forcing the websites to use Brave Software as the middleman if they want a share of their ad revenue.
GP's point was that the browser is adware. It really is - it exists on the advertisement revenue, even if you could use it without viewing ads. Rewards are neither opt-in nor opt-out: during the installation you're offered to try them, have to make your choice, and disable the UI elements afterwards with a relaunch.
It had a major downside of putting browser chrome over the page area and hiding the URL. Over time the latter has become the default anyways but the former is still iffy. Safari approached this concept more recently, as the only browser I'm aware with this as an officially supported option, but put the URL bar directly in the tab, which is a bit more awkward to use. On Vivaldi there is the option to hide it but then activate it via a keyboard shortcut/gesture, which functions similarly but can cause extra switching between keyboard based navigation and mouse based navigation (the latter of which can be very hard to avoid). Because of the way Vivaldi is implemented it's always performed a lot worse than other Chromium based browsers for me as well. In Firefox there were several CSS hacks for this but it was constantly falling apart on updates so I gave up on it.
I guess the long story short is: as a very light tab user (1-8 throughout the day) I miss this old option in Chrom* greatly.
I really liked brave and this is an interesting feature, but I had to stop using it when I tried syncing my bookmarks - for some reason, a bunch ended up duplicating, and I have a lot of bookmarks that are well organized.
Otherwise, I would have kept using it - it was fast and energy efficient on my intel macbook, way better than chrome.
I also have to say - I hate when I have so many tabs open. It's basically a manifestation of my ADHD. I need to focus one thing at a time.
Im conflicted, I like the general idea behind brave, their vertical tabs look good, but the browser deeply integrates this crypto grift and robs sites I visit and have their ads allowed of their revenue then wants them to "opt-in" to their grift to get the money they stole.
They're disabled by default. It's better to try it out than to read opinions on the internet about the browser and form a viewpoint, I say this because I too have been guilty of the same.
IIRC, the background processes for these ad and crypto services still run in the background even when you disable them. I haven't checked the browser in a long while. But it's easily verifiable by opening the brave task manager.
I checked to. On my computer, with a fresh install, i have to kill Utility: Bat Ledger Service from the brave task manager every time the browser is started. Thankfully it stays closed.
The ads and crypto stuff are also not exactly as simple to opt out as "literally disable it in the welcome screen when you first install" as one commenter enthusiastically mentioned.
Crypto stuff is definitely disabled, even if you skip the welcome screen. Actually, I don't think they even ask you to enable it on the welcome screen either.
All of it is definitely opt-in. Even Brave Rewards are disabled by default.
Now if only you could hide the tab-favicon sidebar...
Arc does this, but they don't give you the URL bar. It may sound nitpicky (and it is) but I really want the hidden tab sidebar AND a visible URL bar. Here's hoping either Arc or Brave will make this an option.
I'm on Arc. If you're ok with the url bar being visible where you want it to be visible — you can trigger it per-domain (tab?) then this is how it looks like when you enable Developer Mode for a "site":
Thanks for this. I had seen that once on localhost and wondered how to get that URL bar for all URLs. I didn't know you could use it per-domain, which may be good enough. Cheers.
Reason I dropped chrome around 2011/2012, they removed the options to have tabs laid vertically as not enough users used it (it was an hidden option and badly implemented).
Tree style tabs is the main reason I've used firefox since.
Tabs are really a bad design, maybe made some sense with 4:3 monitors but text was horizontal back then too.
More and more I've wanted a browser based on WebKit/Blink from a company with the original Mozilla's values. I know this will get a lot of flack, but Gecko feels way too antiquated for what I want out of a modern browser. Not to mention, browsers are significantly complicated that splitting developer efforts just doesn't seem like a good idea; Firefox keeps falling further behind in term of modern features.
I don't really think anyone is copying anyone here. This is a popular feature for browsers that has been around for over a decade.
Orion[1] is another recent MacOS browser that released before Arc in 2021 and has vertical tabs. Edge had vertical tabs as an experiment back in 2020[2] and is full feature now. Vivaldi has had vertical tabs since 2015[3]. It might not count since it's not built in, but Firefox has had vertical tab extensions since 2007[4].
Yep, OmniWeb was the first place I saw the of the vertical tabs back in the mid-00s. Blew my mind.
The concept of “drawers” that OS X had at the time was well suited for this kind of thing, with some “free” (in terms of dev effort) flexibility of allowing the user to choose which edge they unfurled from. It’s too bad they fell out of style in favor of much more static sidebars.
I used Opera over other browsers for a long time for performance reasons, it had M2, my favorite email client ever to this day, which got rid of traditional "folders", there's an alternate timeline somewhere where Opera won the browser wars and they already have flying cars and a federated web.
The crucial difference is keeping the address bar- this is where I think Arc is falling short. Maybe it's me, but I have an extremely difficult time reprogramming my expectations of where the address bar should be.
I've been using vertical tabs for well over a decade starting with Opera, then tree style tabs for Firefox, and now Vivaldi. I refuse to use a browser without the ability. Arc hasn't innovated anything there.
Vertical everything. It makes everything so easy to read and scan. I encourage vertical definitions for jr devs, especially ones tempted to put a 20 conditional if statement horizontally (no I don't want to talk about it).
The Orion browser has had this for a while and has the best native implementation by far I just hope they don’t remain pigeonholed into apple OS as they currently are
Lol! I have the decency to open multiple browser windows (large monitor). But, ja - unfortunately I tend to send stuff to the bookmark-graveyard quite quickly.
The latter is more valuable to the ad tech industry. In fact, they benefit from browsers with crippled bookmarking systems. Every time a user types in "facebook" instead of facebook.com into the omnibar, that user is directed to a Google/Bing/Brave Search page with ads.
I can't live with Braves terrible bookmark manager, and inability to save history for more than a few months. But I can't live without the flawless ad blocking.
Data that is shared with Google/Bing/etc. It worries me that the omnibar is enabled by default and records what people search for, let alone employees of major companies, governments, and institutions.
Brave gives a very bad vibe (mostly due to the affiliate links scandal), I prefer to just use my favorite browsers (ff/edge) + uBlock Origin and Privacy Badger which give good privacy/adblock experience, are open source and customizable to my liking.
A minor autocomplete mistake from 2 years ago (quickly fixed and apologised for by the CEO [1]) is framed as a 'scandal'.
Meanwhile Mozilla is showing signs of serious, systemic internal org issues, and no problem? [2]
It's pretty clear a user / privacy-first model is central to Brave's mission, so these drive-by comments griping loudly about minor, historical issues never feel intellectually compelling.
Brave's encroachments can be turned off. On a new install, I disable the search bar Rewards icon, extensions area Wallet, and new tab page Sponsored Images, News, and Cards. (And maybe others still). Each of these requires active engagement from a user to produce and signal for a 3rd party.
Meanwhile, they also fund novel privacy and security research [1], including preventing advanced fingerprinting. In essence, the maintain a patch set on top of Chromium (the open source base of Chrome, that Edge also builds on) which more convincingly respects user privacy, security, and choice[2][3][4]. See how much effort they put into keeping Brave Ads convincingly private [5].
Each of these is less intrusive than changes (to stay recent and relevant) include direct partnership for promotions [6].
Or to base beliefs on a series of news reports that sound bad, see a longer list [7], though it's quite noisy list.
Now all it needs is pie menus (like in Blender, The Sims, etc) that you can pop up by clicking on the tabs, and it will be as fun, efficient, and easy to use as UniPress Emacs was in 1988!
HCIL Demo - HyperTIES Authoring with UniPress Emacs on NeWS:
I later implemented pie menus and tabbed windows for the NeWS Toolkit (TNT) "Open Look" window manager in 1990 for Sun OpenWindows (X11/NeWS), that let you drag the tabs around to any edge of the window, and wrap all your windows in tabbed frames, even including X11 windows!
NeWS Tab Window Demo (This video may not play in some regions due to copyrighted music):
>Don Hopkins developed and released several versions of tabbed window frames for the NeWS window system as free software, which the window manager applied to all NeWS applications, and enabled users to drag the tabs around to any edge of the window.[5]
>The NeWS version of UniPress's Gosling Emacs text editor was another early product with multiple tabbed windows in 1988.[6] It was used to develop an authoring tool for Ben Shneiderman's hypermedia browser HyperTIES (the NeWS workstation version of The Interactive Encyclopedia System), in 1988 at the University of Maryland Human-Computer Interaction Lab.[7][8] HyperTIES also supported pie menus for managing windows and browsing hypermedia documents with PostScript applets.
HyperTIES browser and Gosling Emacs authoring tool with pie menus on the NeWS window system:
>HyperTIES is an early hypermedia browser developed under the direction of Dr. Ben Shneiderman at the University of Maryland Human Computer Interaction Lab. This screen snapshot shows the HyperTIES authoring tool (built with UniPress's Gosling Emacs text editor, written in MockLisp) and browser (built with the NeWS window system, written in PostScript, C and Forth). The tabbed windows and pie menu reusable components were developed by Don Hopkins, who also developed the NeWS Emacs (NeMACS) and HyperTIES user interfaces. (Sorry about the quality -- this is a scan of an old screen dump printed by a laser printer.)
I see the name “Brave Browser” and my brain just refuses to engage. Their marketing team ruined the product for me before I ever got the chance to try it.
There are no problems with the browser. It's the PR surrounding it. Clickbaity tech news websites have been trying to ruin Brave's reputation while Brave users continue to enjoy a fully FOSS browser with the best privacy period.
#/media/File:HyperTIESAuthoring.jpg){kind=link}
Multiple panes with tabs, that can be resized either to custom sizes or automatically (binary space partitioning).
It’s too much for a general purpose browser, which is how we’ve got the Chrome, Safari, Firefox we have. But I think the appeal of a more information-dense browser is broader than we think, anyone doing research would benefit from it.