One alternative view on history that I think could be very useful, is a graph showing not just where I was, but also how I got there and where I went. It happens fairly often that I remember reading something through, say, HN, but I have no clue where the actual post was. Right now the best solution is to search for related words in my history and hope there isn't too much noise, or to search the site I remember coming from. That's incredibly suboptimal.
In fact, this is probably the reason I like Tree Style Tab[1] so much. Instead of a simple list, it shows my tabs in the context I opened them from.
I would like to do a custom search of the content of sites that I have visited.
Oftentimes, I'll remember I read something in the past few days, then it's a mad scramble to open various links to try to figure out if it's the right one.
This. Susan Dumais at Microsoft had a project called "Stuff I've Seen" [1] in the early days of desktop search. In a nutshell, it's indexing and information retrieval on the data that you use in your daily life -- web browsing, calendar events, etc.
I've often thought about trying to re-implement some of their ideas with a local caching proxy or a browser extension. If there's an open source attempt at this already, I'd love to hear about it.
Sounds a bit like a project we worked on for a while until various VC convinced us that getting people to change browsers was impossible, especially since you can't change the default on iOS:
Ok, this is one of those ideas that is clearly dumb but still just might work... you could hack a personal history report out of Google Analytics.
You create a browser extension that adds the GA pagetag to every site you visit. Have the Client ID isolated by tab, but persist from one domain to another, so every tab is reported as a different "User." Then you can use the Behavior Flow reports to see the next-page/previous-page/next-domain/previous-domain data for everything you viewed!
This of course has the enormous caveat that now all your browser history is setting in a Google Analytics account, instead of just only on the browser. But considered Congress just allowed your ISP to sell that data anyways, maybe that's not such a big drawback anymore?
I'm a bit hesitant to use a closed source browser. Browsers these days have such a large attack surface... I wouldn't trust something that is not developed and updated in the open
The most striking thing about Vivaldi's decision to remain closed-source is its provenance. Granted, von Tetzchner does have a history of working with closed-source software, but one would think after the loss of Presto (excepting that one brief leak[1]), that any ex-Opera heads would understand the potential value of open source and community contribution.
As someone who put a lot into the old pre-12 Opera community, I will never use a closed-source browser again. This is not a commentary on quality - Opera was hands down the best browser ever built, despite being closed-source - but rather on commitment. If Vivaldi fails, all of its users will be left out in the cold. Again.
Admittedly, it is not quite so bad a situation as Opera 12. Vivaldi is mostly/largely open source[2]. It still seems like a massive missed opportunity though.
Yes, this is a bit concerning as Vivaldi development was last I heard self funded by Jon von Tetzchner (http://www.wionews.com/science-tech/wionspecial-ceo-explains...). He claims it's otherwise meant to get funds from search engine and bookmark bundle deals. Sounds like a fragile base to build on in a niche browser aimed for power users, who tend to be unusually quick at removing/reverting exactly such features?
Isn't the way out of this via open source and community efforts essentially why Firefox has been successful despite not being OS bundled or backed by a huge search engine company like Google?
There is "Otter Browser"[1] that is FOSS. It aims to be an Opera 12 clone (it doesn't use presto of course). I don't use it but it seems to have active development.
As an "alternative browser" I prefer Brave. It seems to load tabs the fastest, even faster than Opera with its own native ad-blocking. I used to like Opera before, but I stopped using it when it was bought by that Chinese group, and after seeing what they did to Opera Max, and app I found essential for saving data roaming costs. But they pretty much killed its usefulness when they started pushing their ads through it all the time.
I wish http://midori-browser.org would run on macOS natively. This is one app that I miss from my few months with (unstable) Elementary OS. Clean, simple, light, very functional.
I second Brave! It's pretty fast, they have some smart security people on board to prevent them from making dumb mistakes (though that doesn't mean there can't be security flaws, obviously).
Given that I don't write code, I'm happier with the idea of a closed-source browser. The problem I have with using Vivaldi is that it's too much like any other Chrome hack without really bringing back the stuff that made Opera a wonderful browser in a fundamental and meaningful way. Then again, I feel that Presto (and the "limitations" of that engine) was one of the greatest strengths, along with a built-in mindset that the user should control all of what they load and/or see.
Honest question: do you personally inspect the code of the browser of your choice; trust others to do that; or are just are you feeling better about having that possibility?
Given the last few years of adventure in the OpenSSL land I make no assumption about superiority of the open sourced solutions security-wise. Hence the curiosity, what is the motivation of others.
No software is perfect, but I'm more confident that vulnerabilities will be discovered and dealt with on open source projects. Proprietary projects have different incentives.
I wouldn't have a clue what to do with browser code, and I don't inspect the code of the browser I use, but I am happier if people who can understand the code and who aren't dependent on the project for their living can check it.
Being hacked is only one of many security concerns. The authors of software don't always value the users wishes over their own, open source at least keeps them honest about what they're doing. A closed source browser (that I'm sure has some "telemetrics") who's main goal is to analyze browsing patterns is a privacy nightmare.
Agree, but beyond attack surface I just don't want to become accustomed to something that can be shut down, aqui-hire/closed down, or otherwise disappeared via the ongoing advance of technology.
This feature looks nice and I hope other browsers adopt and adapt and make more meaningful end-user advancements like this that are visible to the eye.
I wonder if for a company with less security resources than Google, it actually more secure to develop to develop close-sourced? Is source helpful when developing exploits? Intuitively it seems that it would be extremely helpful for the devious minded-individual.
Security researchers are very used to attacking binaries. The small bit of obscurity you get from hiding the source buys you very little security. And the most insecure internet-facing bits of code of the past decade - Flash and Acrobat PDF viewer - have always been closed source.
Their browser based on Chromium and when there is source available it's at least possible to see how fast they apply security fixes from upstream. For end-user of product there no way closed source browser can ever be more secure.
I always used and loved Firefox because it was open source, but there have been a few articles now showing exactly how insecure Firefox is compared to Chrome. For this reason I've started to use Chrome until Firefox ups it's game.
Apologises in advance for using sarcasm and an analogy:
Cars without the very very very best armour are completely unsafe. Don't use this nice, good, performant, practical and mostly bulletproof van, instead settle for the google armoured car because think if someone starts shooting at you with a 50 cal sniper rifle from a perfect angle!
PS: for extra goodness the hood of the google car is welded (closed source even if we know the engine from chromium), pulling trailers (real extensions) is not possible and by default it records your every turn of the wheel and sends it to Google because you never know...
PPS: Oh, at close range even the google car yields to a sniper rifle.
Me: happily using Firefox. Happily driving a reasonable family car. There are certain places I won't drive that car at night but I can live with that.
This is innovative and might be enough to get me to switch browsers. I'm perplexed at why Firefox, Chrome and IE only ever seem to innovate the engine and have basically given up on doing anything novel with UI.
OK performance is nice to have but frankly I think that's just managers shoving resources at that because they know how to measure it and a 10.3% increase in speed or memory efficiency or something is easy to sell in a meeting.
UI innovation is much harder to quantify, requires more work to sell, and the payoff takes much longer to manifest, not least because every change will generate a certain number of complaints from people who think their pet issue should have been addressed first, like UI designers are interchangeable with performance specialists.
Chrome shot to first place because it kicked Firefox's ass on performance and IE's ass on reliable rendering (an opinion which you may or may not agree with). But while extensions are wonderful they can't and don't replace core UI innovation. And tbh I'm really bored with Chrome at this point; from a user point of view nothing significantly new has happened for a really long time.
Look at the bookmark management, for instance. It's been the same since forever, and if I want to socialize that information or turn it into a feed or something I have to go looking for third-party solutions, which are thin on the ground because people are reluctant to develop too much for a platform over which they have little or no control.
You really need to both do longer studies and include (and distinguish in analysis) new vs. experienced browser users to separate out short-term familiarity effects from long-term usability.
Firefox Test Pilot [1] is a more recent approach to testing new UI in Firefox. It's entirely extension based. I think this is a smart approach because it lets Mozilla quickly test new UI features without building it into the core browser before it's been vetted.
However its full potential is not ready yet, since Firefox has been migrating to WebExtensions, and this has not fully stabilized yet. Long term I think the WebExtensions move also makes sense, since it gives extension authors a stable API to work with, instead of XUL which can break with each Firefox update. And the architecture makes it easier to optimize threading performance and some form of security sandboxing.
So the point is, I think more UI experimentation is on Firefox's horizon, but they have to first stabilize the technical architecture before they can go full throttle on that.
> There is a reason for that – as a rule, browsers don’t really want you to use history. They want you to search and find things multiple times because search royalties are part of their business model.
I think that that's a little paranoid. There's just little demand for a revamped history tab. I like this change a lot, but it's certainly a case of not knowing I wanted it until I saw it.
It might be paranoid from the aspect of history, but this is something that I definitely noticed with the URL bar when moving from Firefox to Chrome.
Chrome will often offer the first result as a search query, instead of a website I visited before. Feels very clear that it's a big part of their business model. Firefox is much more aggressive in serving history URLs. If I type "videos" in Firefox, it will give me "reddit.com/r/videos" as the first result, since I've been there plenty of times, but Chrome insists on offering the first result as a generic Google query for "videos".
Related sidenote, Firefox was also way superior in remembering which letters I type to go to which site in a more broad way. Chrome does it on a very basic level, prioritizing the sites that have the same string on the top level domain. Example, if I type "news" Chrome is always going to offer me sites that begin with "news" such as "news.ycombinator.com", even if I more often end up going to "randomsite.com/news" or "randomsite.com" whose page title is "News!" when typing that. Firefox remembers my preference, which means that after a few days of Firefox use, I'll have a bunch of 2 letter combos that take me to the exact site I want ("ne" to Hacker News, "vi" to reddit videos, "ap" to an apartment listing website which has "ap somewhere deep in their URL, or just in the title of the website, etc.), while Chrome will only give me top level URLs, and feels as useless on week 10 as it did on week 1. For sites where I don't remember the URL, I'll usually end up googling the page title in Chrome, while Firefox would serve me exactly what I need from history.
Part of all of that is just shit UX on Chrome's part, but it's a safe bet that shit UX drives significant revenue for Google.
I never made the connection that the poor UX leads to more search results traffic (makes sense), but I've definitely noticed that the URL bar in Chrome is much worse than the one in Firefox.
I remember the ridicule Mozilla got when they introduced the "awesome bar" (doesn't look like they're still calling it that), but it really did turn out pretty awesome. It replaces both bookmarks and history for me -- which is incredible if you think about it, those are both two very heavyweight interfaces.
It also makes using Chrome very frustrating, since I keep entering keywords that I know Firefox would have indexed (such as a ticket number in a bugtracker), while Chrome sometimes does and sometimes doesn't.
You've noticed more details than I have, but I keep using firefox over chrome because Firefox's awesomebar wipes the floor with what chrome is doing in my daily workflow. I use a lot of intranet sites/internal web applications continually throughout my workday, identifying a lot of unique resoures, and firefox works great at me typing a few characters of a resource ID or URL or page title and finding it out of my history. Chrome feels terrible at it.
About remembering which letters I type to go to which site, I assumed that's just the latest and greatest with browsers, and loved it without noticing that I do. But now that I know it's a Firefox feature, I'll knowingly love it now.
One thing that Chrome's address box has over Firefox is speed. Sometimes I (mindlessly) press f and enter expecting FF to autocomplete to facebook but I get taken to google search for f. (Which of course has facebook.com as the first result)
No, it's absolutely true for Chrome. One of the reasons I went back to Firefox, as Chrome makes it intentionally impossible to find things from your history.
It may be paranoid to believe it's fully and consciously intentional, but it also comes down to general tendency and mindset. Particularly in Chrome, developing a browser for a search company, with search in mind, and investing a lot of time and energy in making the browser "good at search" and integrated well with search, will naturally lead to a search-first mindset in developers, whether intentional or not.
If they want to monetize history via search they could still do that but as a side bar or some other mechanism. i.e., Provide extensive, rich history tracking, but also add some sort of optional search based on relevant context (just please do not make it automatic).
I've always wondered why browsers don't put the contents of pages I visit into something like clucene, so I can search pages in my history by their contents, not just their titles and metadata. I'm sure there are security and storage tradeoffs, but to me this seems like a much more complete solution than this revamped history UI.
There is a recoll extension for Firefox, which updates an index visible from the desktop program. Not quite as neat as integrated history searching in the browser but still useful.
Back around 2000, I wrote a web proxy server that did this. It was a proxy server mainly because browser extensions were not a thing back then. I never released it because I never solved some performance issues (it was using a very naive implementation in a lot of ways). I did get a lot of use out of it for about a year, though.
I'm sure there are security/privacy trade-offs. Storage is not such a big thing. I found that index size leveled off somewhat after a point. It probably would use enough storage to make it unreasonable for mobile browsers.
Man, if this was a thing I would definitely run it. All the plugin approaches fall down for me as I use FF and Chrome interchangeably for different tasks.
This would be a good additional feature for Pinboard perhaps (which I've been intending to subscribe to for ages but never get around to).
This seems very cool and all, but surely 99.99% of regular users have never really thought about seeing their browsing history in a calendar view, or wanted or cared for statistics about it. I am fairly sure a significant proportion of users aren't even totally sure how to access their history - apart from during brief moments of paranoia when they want to keep nefarious things from a partner - and, if they really were looking for something they know they'd seen before, surely they'd just think, "I found this before, I'll just google for it again." (Similarly bookmarks, frankly.)
I guess my point is that this probably appeals to a very small subset of users, but it seems like a massively over-engineered solution for the majority of people, and feels a little bit like the result of trying to answer the question "What can we do that's different?" with a "Wouldn't it be cool if...", rather than solving problems that people genuinely face. Am I being unfair?
My read on the philosophy of Vivaldi is that it's deliberately developed for the minority of users. It's a reaction to the trend of browser features tending towards a generic middle.
They make take it too far (developing features that 1 or 2 users want), but it's still somewhat refreshing to see anyone set out with that kind of mindset.
Also worth noting: users often don't know what they want until they see it. Tabs in browsers were a 'power feature' NetCaptor introduced (and Opera copied), then they got really popular.
Absolutely not - and I didn't mean my post to come across as negative. If this is purely a feature intended for power users - or if the whole piece of software is intended for them - then that's totally fine. Aside from the slightly meaningless strapline, a lot of the copy on the Vivaldi homepage is definitely promoting the sort of customisation only power users care about - so perhaps you're bang on the money there.
From my own perspective, I can't quite decide whether I'm not a power user because this feature doesn't much interest me (which I'm fine with), or whether it would interest me more if I actually tried it; I guess I am interested in the product journey that led to this being built, because it is clearly a large amount of effort for a - to me - quite unusual tent-pole. Likewise, for me, it's hard to see a doughnut chart of "Link transition types" as anything more than a novelty that you would look at once and then hide.
> Does all software have to be lowest common denominator?
I'd like to believe that it's not, but my humble observation is that it ends up being so. A friend just told me that it has to do with economies of scale. I never liked economics anyway.
On a technical level, it has to do with complexity. By reducing it, you lose X% of your customers who depend on the flexibility and features but you serve way better to the remaining (100-X)% and end up with an increased market share as long as the increase in quality is meaningful and X < 20 or something.
In a hard market to crack it makes sense to try and do something different. If it fails then they're no worse off then before. If it works then it's a massive boom for them.
For what it's worth, this is a feature I've long wanted but hadn't realised it until now (I've always hated the way how browsers normally organise the history).
This might be true but in that case I'm in the 0.01% who want to know how much time I spend on HN and other sites instead of working. I ended up installing a time tracker(ManicTime) but I just realized that history statistics in the browser would be good enough for me.
Yes, you're being unfair. 'Solving problems people genuinely face' is never going to result in innovation, because it's maintenance for the existing paradigm. You are not going to find brilliant new ideas by pulling support tickets.
Take the most innovative thing you can't think of and then apply the same arguments to it. It's easy to see what problems things solve after they come into existence, less so in advance. That's why innovation involves the risk of failure.
Thank you. I've always enjoyed this quote from Francis M. Cornford, which perhaps takes an opposing view:
Every public action which is not customary, either is wrong or, if it is right, is a dangerous precedent. It follows that nothing should ever be done for the first time.
This seems like one of the first features that will make me want to use Vivaldi. It's a project I've been rooting for a while, and this might be the turning point for me.
I personally like the built-in notes app with folders, it opens in a sidebar.
It's a nice looking browser too. Been using it on Win & Linux for the last few months and it's grown on me a lot.
Readit News
In fact, this is probably the reason I like Tree Style Tab[1] so much. Instead of a simple list, it shows my tabs in the context I opened them from.
[1]: https://addons.mozilla.org/en-US/firefox/addon/tree-style-ta...
The tails project experiments with this in Servo and browser.html:
https://medium.freecodecamp.com/lossless-web-navigation-with...
Discussion: https://news.ycombinator.com/item?id=13518919
I would like to do a custom search of the content of sites that I have visited.
Oftentimes, I'll remember I read something in the past few days, then it's a mad scramble to open various links to try to figure out if it's the right one.
I've often thought about trying to re-implement some of their ideas with a local caching proxy or a browser extension. If there's an open source attempt at this already, I'd love to hear about it.
[1] http://susandumais.com/SISCore-SIGIR2003-Final.pdf
http://www.azinman.com/#/stateless/
You create a browser extension that adds the GA pagetag to every site you visit. Have the Client ID isolated by tab, but persist from one domain to another, so every tab is reported as a different "User." Then you can use the Behavior Flow reports to see the next-page/previous-page/next-domain/previous-domain data for everything you viewed!
This of course has the enormous caveat that now all your browser history is setting in a Google Analytics account, instead of just only on the browser. But considered Congress just allowed your ISP to sell that data anyways, maybe that's not such a big drawback anymore?
It's interesting that many businesses (and probably governments) have better access to a user's browsing history than the user.
http://www.ghacks.net/2014/11/14/firefoxs-interest-dashboard...
As someone who put a lot into the old pre-12 Opera community, I will never use a closed-source browser again. This is not a commentary on quality - Opera was hands down the best browser ever built, despite being closed-source - but rather on commitment. If Vivaldi fails, all of its users will be left out in the cold. Again.
Admittedly, it is not quite so bad a situation as Opera 12. Vivaldi is mostly/largely open source[2]. It still seems like a massive missed opportunity though.
[1] https://github.com/github/dmca/blob/master/2017/2017-01-12-P...
[2] https://vivaldi.net/userblogs/entry/a-few-words-about-open-s...
Isn't the way out of this via open source and community efforts essentially why Firefox has been successful despite not being OS bundled or backed by a huge search engine company like Google?
I've tried it out and it seems nice and powerful.
Yes I know they publish some source and no I don't use Chrome - for the same reason.
[1] https://otter-browser.org/
https://brave.com/
Best of all, it's open source: https://github.com/brave/browser-laptop/blob/master/LICENSE....
How come? Is there an upside to this I'm not seeing?
Given the last few years of adventure in the OpenSSL land I make no assumption about superiority of the open sourced solutions security-wise. Hence the curiosity, what is the motivation of others.
I wouldn't have a clue what to do with browser code, and I don't inspect the code of the browser I use, but I am happier if people who can understand the code and who aren't dependent on the project for their living can check it.
This feature looks nice and I hope other browsers adopt and adapt and make more meaningful end-user advancements like this that are visible to the eye.
Cars without the very very very best armour are completely unsafe. Don't use this nice, good, performant, practical and mostly bulletproof van, instead settle for the google armoured car because think if someone starts shooting at you with a 50 cal sniper rifle from a perfect angle!
PS: for extra goodness the hood of the google car is welded (closed source even if we know the engine from chromium), pulling trailers (real extensions) is not possible and by default it records your every turn of the wheel and sends it to Google because you never know...
PPS: Oh, at close range even the google car yields to a sniper rifle.
Me: happily using Firefox. Happily driving a reasonable family car. There are certain places I won't drive that car at night but I can live with that.
(Feel free to bookmark and reuse. : )
OK performance is nice to have but frankly I think that's just managers shoving resources at that because they know how to measure it and a 10.3% increase in speed or memory efficiency or something is easy to sell in a meeting.
UI innovation is much harder to quantify, requires more work to sell, and the payoff takes much longer to manifest, not least because every change will generate a certain number of complaints from people who think their pet issue should have been addressed first, like UI designers are interchangeable with performance specialists.
Chrome shot to first place because it kicked Firefox's ass on performance and IE's ass on reliable rendering (an opinion which you may or may not agree with). But while extensions are wonderful they can't and don't replace core UI innovation. And tbh I'm really bored with Chrome at this point; from a user point of view nothing significantly new has happened for a really long time.
Look at the bookmark management, for instance. It's been the same since forever, and if I want to socialize that information or turn it into a feed or something I have to go looking for third-party solutions, which are thin on the ground because people are reluctant to develop too much for a platform over which they have little or no control.
"Boy howdy do normal users have hard time with non-conventional browser interfaces!" https://medium.com/project-tofino/re-defining-the-tofino-pro...
However its full potential is not ready yet, since Firefox has been migrating to WebExtensions, and this has not fully stabilized yet. Long term I think the WebExtensions move also makes sense, since it gives extension authors a stable API to work with, instead of XUL which can break with each Firefox update. And the architecture makes it easier to optimize threading performance and some form of security sandboxing.
So the point is, I think more UI experimentation is on Firefox's horizon, but they have to first stabilize the technical architecture before they can go full throttle on that.
[1]: https://testpilot.firefox.com/
I think that that's a little paranoid. There's just little demand for a revamped history tab. I like this change a lot, but it's certainly a case of not knowing I wanted it until I saw it.
Chrome will often offer the first result as a search query, instead of a website I visited before. Feels very clear that it's a big part of their business model. Firefox is much more aggressive in serving history URLs. If I type "videos" in Firefox, it will give me "reddit.com/r/videos" as the first result, since I've been there plenty of times, but Chrome insists on offering the first result as a generic Google query for "videos".
Related sidenote, Firefox was also way superior in remembering which letters I type to go to which site in a more broad way. Chrome does it on a very basic level, prioritizing the sites that have the same string on the top level domain. Example, if I type "news" Chrome is always going to offer me sites that begin with "news" such as "news.ycombinator.com", even if I more often end up going to "randomsite.com/news" or "randomsite.com" whose page title is "News!" when typing that. Firefox remembers my preference, which means that after a few days of Firefox use, I'll have a bunch of 2 letter combos that take me to the exact site I want ("ne" to Hacker News, "vi" to reddit videos, "ap" to an apartment listing website which has "ap somewhere deep in their URL, or just in the title of the website, etc.), while Chrome will only give me top level URLs, and feels as useless on week 10 as it did on week 1. For sites where I don't remember the URL, I'll usually end up googling the page title in Chrome, while Firefox would serve me exactly what I need from history.
Part of all of that is just shit UX on Chrome's part, but it's a safe bet that shit UX drives significant revenue for Google.
I remember the ridicule Mozilla got when they introduced the "awesome bar" (doesn't look like they're still calling it that), but it really did turn out pretty awesome. It replaces both bookmarks and history for me -- which is incredible if you think about it, those are both two very heavyweight interfaces.
It also makes using Chrome very frustrating, since I keep entering keywords that I know Firefox would have indexed (such as a ticket number in a bugtracker), while Chrome sometimes does and sometimes doesn't.
As Henry Ford apocryphally said, 'If I had asked people what they wanted they would have told me to bring them a faster horse.'
http://help.opera.com/Mac/10.50/en/history.html
https://addons.mozilla.org/en-US/firefox/addon/recoll-indexe...
The one complication I found was ensuring that I blacklisted sensitive websites such as banks.
I'm sure there are security/privacy trade-offs. Storage is not such a big thing. I found that index size leveled off somewhat after a point. It probably would use enough storage to make it unreasonable for mobile browsers.
This would be a good additional feature for Pinboard perhaps (which I've been intending to subscribe to for ages but never get around to).
I guess my point is that this probably appeals to a very small subset of users, but it seems like a massively over-engineered solution for the majority of people, and feels a little bit like the result of trying to answer the question "What can we do that's different?" with a "Wouldn't it be cool if...", rather than solving problems that people genuinely face. Am I being unfair?
My read on the philosophy of Vivaldi is that it's deliberately developed for the minority of users. It's a reaction to the trend of browser features tending towards a generic middle.
They make take it too far (developing features that 1 or 2 users want), but it's still somewhat refreshing to see anyone set out with that kind of mindset.
From my own perspective, I can't quite decide whether I'm not a power user because this feature doesn't much interest me (which I'm fine with), or whether it would interest me more if I actually tried it; I guess I am interested in the product journey that led to this being built, because it is clearly a large amount of effort for a - to me - quite unusual tent-pole. Likewise, for me, it's hard to see a doughnut chart of "Link transition types" as anything more than a novelty that you would look at once and then hide.
I'd like to believe that it's not, but my humble observation is that it ends up being so. A friend just told me that it has to do with economies of scale. I never liked economics anyway.
On a technical level, it has to do with complexity. By reducing it, you lose X% of your customers who depend on the flexibility and features but you serve way better to the remaining (100-X)% and end up with an increased market share as long as the increase in quality is meaningful and X < 20 or something.
For what it's worth, this is a feature I've long wanted but hadn't realised it until now (I've always hated the way how browsers normally organise the history).
Deleted Comment
Take the most innovative thing you can't think of and then apply the same arguments to it. It's easy to see what problems things solve after they come into existence, less so in advance. That's why innovation involves the risk of failure.
Every public action which is not customary, either is wrong or, if it is right, is a dangerous precedent. It follows that nothing should ever be done for the first time.