Safari is released to the world (2013)
donmelton.com
donmelton.com
Huh. I wish Apple still focused on these apps. Those apps used to be great at the time this guy was there. Nowadays it's all kind of lame.
But yeah these days all the development on desktop seem to go on JS/HTML - partly thanks to WebKit (which was a basis for Chrome later, which was basis of Electron) - which makes the app kind of suck, but honestly the development side is much easier. And people actually use phones in 2023 more than desktop. So, whatever.
Apple's reminders feel like a godsend by comparison. It does exactly what it needs to, and syncs flawlessly between my phone and Macs. I can set reminders through Siri, and it even has a really intuitive way of setting date/time for reminders, where you just type, e.g. "tomorrow 5pm" and it'll highlight the text and give you a suggestion to make the reminder happen then, you click it and it yanks the text out and populates the appropriate time. You can even set reminders to appear when you're messaging a certain contact, and it'll appear within the Messages app in that contact's conversation. And it got even better with the new lockscreen widgets in iOS16, where I can see my reminders for the day front-and-center on my lockscreen. I gotta give props to Apple for their reminders system, cause I didn't even have any sort of expectations in the first place, but it's continually impressed me and become a consistent part of my daily routines.
[1] https://www.androidpolice.com/2021/07/11/reminders-via-googl...
It's by far what I miss the most since I moved to Linux. Outlook for web wipes the floor with regular outlook IME (it's much, much faster) but it's still a pain to use compared to Apple Mail.
If anyone has any comparable solution for Linux, I'd be glad to know. That would be something streamlined, not having to cobble together random different programs for obtaining the mails from an Exchange Online server, some other app for not-quite-fully-working calendar support, etc.
I've found it kind of anemic and prefer the Linux mail clients (Kmail, Evolution, Thunderbird) and their plugins.
It was also very smooth, and I didn't have to wait around for it doing who knows what, but this hasn't been an issue with KMail when I used it regularly some 12 years ago (without Exchange, though). Only Outlook ever felt noticeably slow.
Some other creature comforts were great, like integration with quick look (tap space on an attachment, and you can see it in a popup).
I could also have a unified calendar with my personal (Gmail) and work (Exchange online) accounts.
I only had to tell it "this is my exchange account, and this is my gmail account, now do your thing" – and it did, complete with OAuth2 and all.
I haven't tried KMail recently, but last I've seen (a few years ago) it didn't support Exchange online.
Thunderbid had some Exchange support, but it was a bit wonky (around 2019). I don't think the calendar was actually working. Maybe things have improved, I seem to remember there was a major upgrade recently.
Evolution, IIRC, had some kind of support via EWS, but I never "clicked" with it so never tried it seriously. Installing a gazillion gnome packages didn't help its case, either (I don't use gnome otherwise). According to the Arch Wiki [0] it should support mail + calendar through MS's Exchange ActiveSync (which is proprietary, but I don't have an issue with that if it works).
[0] https://wiki.archlinux.org/title/GNOME/Evolution#Microsoft_E...
Safari is released to the world - https://news.ycombinator.com/item?id=5041354 - Jan 2013 (120 comments)
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
It looks like the Qt side of things doesn't get updated anymore. If you want to use webkit, there are still options, though.
However, in my experience the open source webkit engine and Safari are very different. I don't know why you'd use webkit for anything if Blink and Gecko are available.
This has not been my experience using Playwright which compiles webkit at every single new merge.
The Webkit clients I could get working on my machine all lacked features Safari reportedly support, leading to a lower score than Safari.
I don't know what's missing exactly, but there's clearly a distinction between the two, even if the core functionality is the same.
edit: ahhh, it's not KDE/Qt. Epiphany is GNOME. Okay
https://youtu.be/gPyrzvI7Mlc?t=3240
It's fascinating to see just how much new stuff was unveiled.
I still use Safari on macOS. It seems to retain great performance with great battery life.
Off topic:
What would be cool is if Apple released Safari for Windows (again) and Linux. Ok, probably not for Linux, since Apple doesn’t work well with the desired licensing schemes and since Apple isn’t capable of working with the diversity on Linux to release an end user application.
Safari on Windows and Linux may be probable (not even possible) only in an alternate universe, since Apple focuses on OS support for its applications in (roughly) this order: iOS, iPadOS, macOS, tvOS/homeOS and miscellaneous (this includes Android and Windows). Anything outside iOS seems to get really smaller budgets, smaller teams and very little attention.
Additionally, there is no money in Safari on other platforms, unless Apple does come up with the oft rumored Apple search engine and expands its advertising ambitions even more (considering how poorly Siri has been improving over time, any such search engine would be better not being developed or released at all).
But aside from that use for it, Safari for Windows would have 100% been a play to try and keep Google from taking over the web, and I guess that didn't matter to them.
I also installed it immediately on Windows. It was great for testing issues and seeing how things would render, but it was not performant.
It seemed to do that same thing as iTunes and QT where it felt like Apple just ported their whole GUI stack on top of Windows. It looked and worked exactly like in a Mac. But that came with a big performance cost.
Even as a big fan I couldn’t stand to use it as my main browser on my windows machine at work. I only used it for testing things. Otherwise I stuck to FireFox.
I’ve always felt like Safari for windows was 100% about the HTML based “apps” before the App Store (so Windows users could develop them) and so companies could test their sites for iPhones without having to buy them when they were still niche/expensive.
As iPhones rocketed to popularity, they didn’t seem to see it as necessary. The popularity of the iPhone alone was enough to force testing. And the iPod touch was there too as a cheaper testing device.
I know it came out 4 years before the iPhone. Maybe it was laying the foundation? I just can’t see Apple truly wanting to be in the Windows browser market.
* AppleWin, Apple's port of WebKit for use in iTunes for Windows. I'm pretty sure it depends on DLLs that reimplements Cocoa APIs on Windows, those DLLs are shipped with iTunes. (edit: one can download the support libraries here [2])
* WinCairo, maintained by Sony and uses Cairo for rendering. Maybe someone can fill me in why is this a thing, since Sony only uses WebKit on the PlayStations.
[1]: https://fujii.github.io/2019/07/05/webkit-on-windows/
[2]: https://developer.apple.com/opensource/internet/webkit_sptli...
As for why Sony maintains WinCairo, I believe they use Cairo on their platforms, and they mostly use Windows for their dev systems (as does most of the games industry).
edit: oh it's "GNOME Web". OK
I did watch this, and at the time was only a year into using Mac OS X coming from FreeBSD and Linux before that, where I used KDE. So when "KHTML" went up on the screen, I was right there with it.
I don’t like the analogous story on sports though.
Yahoo and Google look pretty similar these days (clearly innovation moved to mobile apps):
https://web.archive.org/web/20130701000055/http://www.yahoo....
http://web.archive.org/web/20130308002617/http://www.google....
Demo focused so much on bookmarks, and Jobs was showing off Flash during it!
Firefox is a huge PITA to set up right, but once you manage to get it set up with Sidebery, MUC, etc, it works well.
I love how at the time he thought that lacking tabbed browsing was no big deal, since its a broken paradigm anyway. Surely Apple has something better than tabs up their sleeve!
But KHTML - basis for Chrome, Safari and Edge - won the web browser for open source.
In the time since Safari's launch, Safari's WebKit has completely proliferated. Chrome and Blink came to the forefront and took even more marketshare. We live in a browser monoculture, dominated by the giants. Firefox is just a blip on the radar.
Isn't this what Microsoft got sued by the Department of Justice over? Times are so different now.
Mozilla barely ekes out an existence, and it's only thanks to the good graces of Google. Google's fear of antitrust action, that is.
Mozilla nurtured Rust, and we should all be thankful for that. But Firefox doesn't look healthy, and it's such a shame. I'd welcome a world where Firefox was back at 30+%.
Outside of Safari, WebKit's popularity is largely owed to how easily it embeds — regardless of UI toolkit, platform, or language you're working with you can probably embed it, and so WebKit has shown up in countless embedded applications for over 15 years now.
And while Blink isn't designed to embed as cleanly as WebKit does, Chromium can be pared down enough that it in itself can be used as a platform, which we've seen in Electron. While XULRunner was similar in spirit, it was cumbersome and clunky, which is why at one point one of the more popular Gecko browsers (Camino) was instead written with AppKit. Electron isn't exactly an elegant swan itself, but it's considerably more responsive and is arguably easier to write a passable application with.
Embedding being killed also meant that there can only be one Gecko-based browser, because lack of embedding means being chained to XULRunner and prevents other Gecko browsers from being significantly different. So now if Firefox fails, Gecko is dead. If browsers like Camino still existed they could carry on the torch, but now those are gone.
I know that embedding had serious technical issues but I think they were worth solving. By dropping embedding support they threw out the baby with the bathwater.
> The KHTML library (as wrapped by KFM) was surprisingly usable back in the day on Linux. But who'da thunk it'd turn into one of the most important pieces of code in the world? Congrats to all involved.
It’s because of those pushes the web got faster and FF made such huge strides in performance (speed + memory usage).
Together MS started to realize they had to release IE 7 after sitting on IE 6 for 5 years in 2006.
Then Chrome came along in ‘08 to push everyone further.
I wonder if we would have gotten Rust if FF had pushed on as the best browser.
It’s unfortunate FF isn’t bigger. Chrome’s dominance (made worse by Edge becoming a reskin) really worries me.
> Mozilla nurtured Rust, and we should all be thankful for that. But Firefox doesn't look healthy, and it's such a shame.
Mozilla had a second chance with Rust and had years to make tons of money out of it or at least lessen their dependence on Google. Instead they literally threw that opportunity away and started chasing expensive distractions.
> I'd welcome a world where Firefox was back at 30+%.
I'm afraid that ship has sailed and Firefox is completely doomed and almost fully dependent on Google's money, especially with their shrinking browser market share. Chrome's dominance will just be more entrenched to the point it needs to be an independent non-profit, governing the browser engine as a standard and not controlled by Google.
As the only competitor for the status quo their efforts over the past decade have been embarrassing, gets even worse when we look even further back and see how they let IE dominance happen but never forget they existed through that, managed to take back some control then squandered it and let exactly the same history play out a second time while burning money on overpaid no-talent no-vision know-nothing execs.
I hope the EU/US Government makes this happen _before_ they force alternate browser engines on iOS -- that ~50% of mobile market share is the only thing preserving any resemblance of the open internet right now, and I fear that the lag between those two regulatory actions will effectively gift Google a monopoly on browser engines.
As would I, but using Firefox and following its development since 1.5, the outcome is entirely understandable. They chose a dick-measuring contest with Google over sticking to what built its usershare in the first place: user control. Until Google finally kills off competent adblocking, Firefox offers me nothing Chrome doesn't. And it's damned sad. They threw the ten percent under the bus so they could chase that 60% fantasy and this is the inevitable result.
> Isn't this what Microsoft got sued by the Department of Justice over?
They were integrating the browser into the OS, wouldn't allow alternative browsers to be installed, and took actions to get websites that would work only on Windows (notably with ActiveX). We don't really see that today.
[0] https://developer.mozilla.org/en-US/docs/Web/API/MediaSource
But it is cruel reality.
Most expensive phone but Safari crashes more often than 100 eur 5 year old Android.
https://www.google.com/search?q=A+problem+repeatedly+occurre...
https://bugs.webkit.org/ is full of bugs which are never fixed or will take years.
You will never know where new iOS release breaks rendering engine again and new bugs appear out of nowhere.
And it is almost impossible to debug them, because macOS safari is different than safari on iphone or ipad...
And what about bug in webview in some app or firefox/chrome/edge on iOS...
I'm sure Chrome and Firefox issue trackers also have plenty of bugs which will never be fixed - that's just generally what happens with long-running software projects.
Safari needs big iOS update to fix bugs...
It would be great if they atleast allow other browsers. So there would be alternative. Webkit bugs kill all browsers.