> Most people switch browsers for one reason: speed.
Is that true? Maybe it is and I'm out of the loop but I can't remember the last time someone complained about browser speed. The bottleneck seems to be website bloat more than anything else. Would love to see this argument quantified.
Nowadays users switch browsers to escape from AI nonsense. But in all seriousness, just enabling an ad-blocker significantly increases the speed of the browser, because, as you correctly noticed, website bloat is the largest bottleneck. And usually "raw" website content is only small fraction of all other stuff that gets loaded from various remote sources to show you ads and track you better.
And to take speed point even further - disabling JavaScript does wonders to website speeds, you won't believe how quickly some websites are loading. Logging in to banking website might not work at all, though.
I've been playing Dragon Age Origins recently, and I've been popping into the Steam overlay browser to look up some stuff, which frequently leads me to the wiki. And oh my god, I can't believe how bad the internet is without adblock these days. Every page visit, it pops up ginormous video ads that cover 90% of the web page, and it needs to chug along to get the initial render done before I can collapse it.
> You will probably already feel this difference every time you launch it.
How many times a day / week / month do you launch your browser from scratch ?
It is also a moot point with modern processors and modern OSs.
Even more so in Orion's target macOS market where you can leave an app open without any windows open (not minimized, I mean not open at all) , so its ready to go at a click.
Unless you are regularly using all your RAM, there is a very good chance the browser's files are going to stay in disk cache and be effectively instant to load. 100MB is an imperceptible load time difference on a modern system where memory speed is 50-100GB/s or more.
When all my windows are in one big stack (Windows, macOS), then multiple windows just get in the way and I’d agree.
But over the last couple months using better window managers like sway or niri, I tend to open new app/browser windows next to the windows they are related to.
It was true, and it was what made Google Chrome popular in the first place. Internet Explorer and Firefox were dead slow to start at the time while Chrome started instantly.
We just don’t know how bad slow browsers can be because all others have caught up.
That was a funny period of time because you could very transparently see the clear application of a corporate team that was tasked with improving the “startup speed KPI”.
During that time IE startup time went from a dozen or so seconds to also instantaneous. It was even faster than chrome sometimes. But that was just the startup. The application wasn’t ready to accept any user input or load anything for another 10 or 15 seconds still. Sometimes it would even accept input for a second then block the input fields again.
It’s the same mentality all those insanely slow webapps do when they think some core react feature for a “initial render” or splash screen etc will save them from their horrific engineering practices.
Google did a great job communicating Chrome's improvements over speed (both with startup and prefetch) and reliability (isolated and sandboxed tabs) during its launch. When you saw it, you knew that it was basically game over for any browser that had chosen to stagnate until then. They destroyed the competition.
I think Google gained more users with its aggressive advertising campaign than with its speed (except for power users). If someone used a Google product like search, email or youtube in a non-google browser, Google would always show an ad encouraging them to switch to Chrome.
At the time, the argument for Chrome was that Firefox and IE were bloated and their memory requirements were too high.
A system with less than 64 Megabytes of RAM (most computers of the time) would have to lean heavily on spinning rust virtual memory, making everything slow.
However, since then Chrome has become one of the biggest memory hogs that people commonly run.
I don't think lean memory use was the biggest claim Chrome had made. That was the game between IE and Firefox. Google had specifically promoted faster startup times, faster web browsing experience, and tab isolation / sandboxing so a crashing tab wouldn't bring down other tabs with it.
Netscape 6, which was released in 2000 and based on the Mozilla Suite (now SeaMonkey) recommended 64MB of RAM. The Mozilla Suite was the basis of the Phoenix project (later renamed to Firefox) and they shared the same technological underpinnings: Gecko engine, SpiderMonkey JS engine, XUL interface, XPCOM, etc. Phoenix/Firefox was about using the Mozilla technology to deliver just a browser, independent of the suite, with aim of being lighter weight. So while Firefox didn't exist yet its heavier predecessor did.
Unless it's ungodly slow, to the point where it's beyond being noticeable, speed is the last thing I care about when it comes to browser. Most of the options available are reasonably fast and differences are not huge enough.
Gentle reminder that if you're commenting on hacker news articles you are likely the outlier in the "why people switch browsers" reasoning. Friends and family constantly surprise me with their tech choices and how they interface with the digital world whenever I'm home on holidays.
It was the primary motivating factor behind the previous major browser shift, though there were also other large factors.
Remember that users often don’t correctly figure out which part of the stack is causing something. I’m guessing people generally don’t ID the browser as the performance bottleneck unless they’re familiar with browsers of significantly differing speed, and when not it comes out as asking for faster internet, faster websites, or a faster computer, all of which we hear constantly.
I wouldn't say it's only speed.
I've been Firefox for years, but eventually ended up surrendering Apple eco-system. with Apple silicon, Firefox at least then wasn't sleeping that well, and the tab sync of FF between my devices was also less than I've desired.
So performance is general is more like it...
that includes not hurting my battery life.
I've used all 3 browsers (chrome/safari/ff) daily doing web dev for years now and I'm convinced Safari just feels faster as a cohesive Mac app, with the animations and what not, but isn't in general when using the internet day-to-day. FF is little different than Chrome/Safari.
Also as a dev Safari is becoming the new IE. I've had a whole suite of Safari-only bugs in the past 2yrs and lots of browser crash reports from users.
I have definitely switched in iOS to orion for the support of firefox and chrome extensions. Have not the slightest idea how different browsers in mobile compare in speed. But if it was abysmally slow I would have had seconds thoughts about it probably.
From my perspective, all browsers are fast enough and within a couple of percent the same performance. I value features, privacy, etc. More than raw speed.
I switch(ed) for simplicity and privacy. Haven't found any yet. Camino and Firefox used to be that; and the browser on ElementaryOS (which IIRC was just a cleaned Firefox but not sure). Not anymore. Stopped using ElementaryOS, and every other browser collectively decided to aspire for FUBAR.
Now I think I'll just keep switching until there's one decent browser left which hasn't been AIed.
Hmm, I did switch to Safari from Firefox because I couldn't put up with how slow everything felt. Ironically I now find Safari quite laggy, whereas Orion or Brave with uBlock make for a better experience. I do agree most people either don't switch browsers or switch to something they heard is good. Maybe Kagi have better intel than us.
A lot of people switched away from Firefox / IE to Chrome when it launched.
Orion is faster than Safari on the same Mac. And it isn't rendering speed, but basic UI interface, multi-tabs usage. It is annoying because you see what Webkit is capable of and somehow Apple is not doing such as great job for Mac Safari. The difference is especially true on x86 Mac.
I use Kagi as my daily driver on mobile, and have it constantly as my second browser (next to FF Dev) on desktop for the same reason I use Kagi Search, support of the concept. It doesn't hurt that the browser is pretty good performance and experience-wise.
Applications that use browser engines for rendering tend to be a bit sluggy compared to native applications, yes. But I don't think a common complaint is that a web browser as a standalone application is particularly slow either running or starting up. People tend to say stuff gets slow once they have a ton of tabs open, which makes sense.
Don't know how common it is, after all, people are used to all the slow stuff out there, maybe they don't even complaint when it's less frequent actions like opening a browser. Though at least for a ton of tabs, there are hibernating solutions, so very annoyed people can at least find a workaround, unlike with the unfixable startup delay
tbh, it's been 10 years I don't use Chrome... When I used Windows, my main browser was Edge, and was quite fast to open as far I remember, but I believe on Windows Edge stays open in background all the time anyway...
Also, what's "instantly", how much time exactly on a cold/hot start does it take after you press a key and a fully drawn/functional window is shown?
Checked on an old Windows desktop (hot start), and while the empty window is shown in 0.8 sec (not too slow, but still not instant, those would be some old native apps taking 4 times less time), the fully drawn/functional window takes 2-3 sec, so definitely nothing instant
Recently as in the last 8 years when they overhauled it. It really was slow as heck back in 2016, but the e10s effort really, really paid off in terms of performance.
It runs noticeably faster than chrome on my 12 year old laptop. Plus, it isn't riddled with invasive tracking garbage.
Even before then, 99% of the difference came down to whether chrome and firefox were properly using gpu acceleration. (Both could be easily misconfigured.)
I never saw a situation where the actual engine performance mattered in real world scenarios.
These days, all the engines are comparable, except that Google sabotages safari and firefox on its own sites.
Big fan of sublime merge. I recommend it a lot to people who need to dip their toes in source control and want some layer of abstraction, but also want to feel like they’re connected to the underlying tool (git). Merge balances this very well.
The vast majority of developers working with git daily don’t know what a bare repo is, or that it exists at all. It’s not obscure knowledge as such, it’s just never come up for them as something they need.
The vast majority [0] of developers working with git daily have no mental model of git repos and just do a mental equivalent of copy'n'pasting commands and it's enough to let them do their work (until something breaks at least), so it doesn't seem like a particularly good indicator of whether something is obscure or not. There are many obscure things hiding in git, bare repos aren't one of those :)
These types of issues have to be the largest footguns in the react architecture. Relying so heavily on the linter to avoid the initial dependency mismatch issues and then on memoising to avoid the re-triggering can really feel like dancing on eggshells.
