Not really. Even if someone TRIED to make a Safari only page, it’d still render fine in any other browser. Trying to make a “safari only” page would end up with a page that’s rendered just as well on safari as any other browser. Making a “chrome only” page means a page that’s only rendered properly in chrome.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Safari Team Asks for Feedback Amid Accusations That 'Safari Is the Worst, It's the New IE'
- Thread starter MacRumors
- Start date
- Sort by reaction score
Sure, but to make a "safari only" page, you NEED a Mac. And here comes the problem that just testing a page for safari is a pain. Every developer needs an Browserstack account or must own a Mac.
Btw Firefox and Chrome do also not share the same engine and in most cases it just works.
What I’m saying, though, is even if you were to get a Mac and TRY really hard to make a “safari only” page, it would still NOT be Safari only, it’d render fine in ANY browser. Relatively speaking, it’s effortless to create a “chrome only” browser, and as a result, you see those often.
We actually had an internal site that was sort of Safari-only, although it did it by sniffing the user agent. It was a case of "if Safari, make it look iOS-like, but if not then make it look Android-like". Then work stopped supplying Androids and the Android 'version' of the site disappeared.
Yeah, sort of safari only.
But still, the page, as it is, works as expected on all browsers, including the redirect. It’s actually possible, without user agent shenanigans, to have a properly written chrome site to NOT render in safari.I get the point with "only", however the keyword is effortlessly. It also makes no sense to have the most "following the norm" browser just running on Mac. And sadly developers are also lazy
SOME rendering engine has to be the “following the norm” one.I get the point with "only", however the keyword is effortlessly. It also makes no sense to have the most "following the norm" browser just running on Mac. And sadly developers are also lazy
In the past, IE was certainly NOT the “following the norm” one (just like chrome today). And, were it not for the years of WebKit’s adherence to the norms, chrome in it’s current state wouldn’t even exist.Devs want to have a quick “out” for anyone that asks them why their site doesn’t work in Safari. And “because I designed it just so it would work in chrome” raises questions (like “Why?”). “Because Safari is the new IE” doesn’t!
There are plenty of lesser and older browsers that would not be able to render sites that specifically targeted Safari's level of development. So I'm curious why Safari sites should limit themselves to Safari's capabilities and not some other browser. Internet Explorer only supports 312 HTML5 standards (compared to 528 for Chrome and 471 for Safari). Netscape only supports 37. Why should the internet be limited to Safari's development, and not IE's or Netscape's?
Rather than complain that the web doesn't work well on your chosen browser, why not choose a browser that can handle the modern web?
There are no “Safari sites”. Sites that render well in safari render well in any other browser. The internet is NOT limited to Safari’s development as is witnessed by the number of chrome only sites.
Web developers are free to do what they want. “What they want” used to mean “code specifically for IE”. Now, it means “code specifically for chrome”. When someone asks why their site doesn’t render in Safari, there’s no shame in saying, “I wrote it specifically for Chrome and have taken zero effort to ensure it works in Safari, so use Chrome.” If a customer wants, they can download and use Chrome. If not, there’s plenty of internet out there that DOES render in Safari that should keep them busy Again, you are missing or ignoring the point. For example, you claim that sites that render well in Safari will render will in any browser. That is obviously false. Safari supports many modern web standards that older browsers do not. Are you seriously suggesting that Netscape Navigator 2.0 could properly handle all the HTML5 standards that Safari can?There are no “Safari sites”. Sites that render well in safari render well in any other browser. The internet is NOT limited to Safari’s development as is witnessed by the number of chrome only sites.
You also keep saying that developers should not aim at any particular browser, but then insist they should aim at Safari. So I ask you again, why Safari? Why not Internet Explorer 11, which is much worse than Safari? Why not Vivaldi? Why not Netscape? Why not Chrome? It's like you bought a car that can't get to freeway speeds, and are complaining that civil engineers are designing roads for modern cars instead of your Peel 550.
If your browser can't handle the modern standards used by current developers, and if the sites the build to those standards are ones you want to use, you should just use a better developed browser.
Are there any recent differences between the rendering engines? Or is it the JavaScript engine? I can't find a board which lists all the "step over the norm" regarding to chrome ?.SOME rendering engine has to be the “following the norm” one.
Devs want to have a quick “out” for anyone that asks them why their site doesn’t work in Safari. And “because I designed it just so it would work in chrome” raises questions (like “Why?”). “Because Safari is the new IE” doesn’t!
Another big hit I think is the browser version binding to the OS. We must deal with old Safari versions back to 11 because people are not concerned of, or willing to update. Just forcing them to update the OS to the latest version is not our job. With this in mind, it indeed remembers me to IE times, where you must support IE x. Microsoft came with Edge, they sticked to IE 11. Safari has in the older version many CSS bugs, some I know are connected to flexbox. The render IS broken without a hack. Of course, "following the norm" and introducing bugs are different shoes, but this is the salt and pepper with Safari.
#1 Apple bring the newest version to all operating systems, after that, I can imagine a change of things. Now it's just a chat in a roundabout ?
What he’s saying is that Safari has pretty good standards support, even if it doesn’t run every wizzbang new standard. Maybe you don’t remember the bad old days of IE, but stuff written for IE was horrendously broken. Web developers had to put in all sorts of fixes to maintain support for IE (which was critical, since it was still the most popular web browser at the time). However, a page that’s written for the set of standards that Safari supports will run well on modern versions of Chrome and Firefox, even slightly older pre-Chromium versions of Edge.
I do think you’re being deliberately obtuse to an extent, though, or at least overly literal. No one is suggesting that targeting Safari will support every browser that’s ever existed, but what he is suggesting is that targeting Safari (or the set of standards Safari supports) will give you coverage of every browser with a share greater than 1% or 0.1%, basically for free. While targeting for Chrome lets you shoot yourself in the foot with technologies that aren’t well supported in Firefox and Safari.
The big thing missing in all this is to what extent adding Safari support breaks an existing web page. Back in the IE 6 and 7 days, it often took just as long to get a page working in IE as it did to get it working in all other browsers combined. That doesn’t seem to be the case with Safari, it performs to standard, but it doesn’t support every latest standard. Which is a slightly silly thing to complain about, at least in terms of JS, since most people are still using things like babel in their project boilerplates. It’s also somewhat silly to see how eager web developers are to live on the edge, in terms of supported technologies. (Here’s a hint, most developers don’t get the privilege of working with the latest and greatest technology. I still have to write in Python 2 compatible code on occasion and maintain VBA code!)
Edit: And lazy devs who just want to target Chrome are a legitimate issue for the longevity of the WWW. Basically, you’re complicit to the act of giving one vendor (and a big one, at that) control over the web again, just like back in the Netscape 4/IE 5 days. Do you legitimately trust Google any more than you trust Microsoft? I’d suggest that you shouldn’t. You shouldn’t be an enabler assisting one firm to have a 90%+ market share. Oh sure, Google may not let Chrome stagnate like Microsoft let IE, but giving Google control over every major browser lets them do some big self serving things, re web standards. And, if Google bullies the Mozilla project into siding with it (by threatening the Mozilla project’s main revenue source, for instance), then it can artificially manufacture consensus among two of the three web browser developers.
Last edited:
Are YOU seriously suggesting that Netscape Navigator 2.0 could properly handle all the HTML5 standards that chrome can?
Here’s another one, the Blackberry browser can’t support code written for the IE version released in 1827.I didn’t say that developers should not aim at any browser. I day they DO aim specifically at chrome (and developers in the thread has confirmed that). AND, I say, that there’s nothing wrong with that, they’re free to do what they like…
But, YOU are saying that users should bend to the will of web developers. I’m not saying that. I’m saying that a user’s free to download Chrome if the site’s important to them OR not if they just want to use Safari. They just have to understand that some developers exclude Safari in their testing and their experience will show that.
I think that’s what the developer in the story the thread is related to is asking for. There’s a lot of devs saying “safari is the new IE” (and we know why) but the webkit developer wants to know “in what way”. Most web developers having problems likely don’t even know because they ONLY code to chrome and then just tell their users to use chrome, anyway.
This is probably the closest thing that relates to the idea of being like IE, but, it’s not related to the work the webkit developer is doing today to deliver a renderer (again, the topic of the thread). There’s nothing they’ll be able to do to get someone to update or use a different browser, because, like you stated, the browser is tied to the OS in a lot of cases. But they CAN work to ensure future versions of Safari can do more. And, I’d imagine the number of aged non-Safari browsers dwarfs the number of aged Safari browsers, right?
Certainly I was deliberately choosing a extreme case when I asked why we should not be developing for Netscape, jus to show the absurdity of his approach. But IE 11 holds a 1.3% marketshare, which exceeds the " 1% or 0.1%" you suggest should be in the target range. Samsung Internet is at 3.5%. Both of these are behind Safari in terms of compliance (IE 11 is miles behind). So again, I ask why should Safari be targeted and not IE.
I remember the bad days of IE very well. Hell, I remember when Mosaic released for Windows, and my first time accessing the World Wide Web. There is no IE equivalent today. IE was problematic for several reasons. Yes, developers tested against it first and foremost, because it had market dominance. But that is not the core of the problems. It was significantly worse than the competition, but had massive market share due entirely to the market share of the systems it was the default browser on. Chrome also dominates the market, and is the first and most important browser to test against. But its position comes from people actively choosing it over the default browsers of their devices. Like Firefox before it, Chrome's rise was due to quality.
Safari's market share isn't what IE's was, as Apple doesn't dominate the computing space the way Windows did. And Safari does not fair as poorly as IE did in comparison to its competitors. But its market share is almost entirely due to it being Apple's default browser, and it is behind other options. In that way, it is like an IE junior. If Apple's market share were higher, it would be IE.
I'll say it again. Devs should be targeting web standards. This is true to web devs AND broswer devs. Right now, that means that Chrome is going to give the best experience, because Chrome best supports those standards. You can't claim to worry about the future of the WWW while also arguing that web development be limited to the pace of Safari (a browser chosen because it holds an artificially high market share).
The WebGL example I gave earlier is illustrative. WebGL was supported by all the major browsers from 2012 to 2017. In Jan 2017, Safari dropped support. You seem to be saying that web devs should have followed suit, removing WebGL code from the applications. What would have happened then? How would the WWW benefit? In that world, every browser would eventually drop support of WebGL, since no site could use it any more. Safari would get to exclusively decide the shape of the web. There's your nightmare scenario.
In reality, devs continues to design their applications according to web standards, even when not every browser supported them. People who wanted to use those sites used browsers capable of rendering them properly. The put pressure on browser devs to improve their products. As of late last year, Safari began supporting WebGL again. Let's remember that IE's problems didn't result in the fall of the WWW. It resulted in the death of IE. If Chrome ever becomes IE, people will simply move to other options.
Are YOU seriously suggesting that Netscape Navigator 2.0 could properly handle all the HTML5 standards that chrome can?
I didn’t say that developers should not aim at any browser. I day they DO aim specifically at chrome (and developers in the thread has confirmed that). AND, I say, that there’s nothing wrong with that, they’re free to do what they like…
But, YOU are saying that users should bend to the will of web developers. I’m not saying that. I’m saying that a user’s free to download Chrome if the site’s important to them OR not if they just want to use Safari. They just have to understand that some developers exclude Safari in their testing and their experience will show that.
Huh? You claimed that sites made for Safari would render in any browser. I gave examples where that was not true, including Netscape.
Developers in this thread did not say they developed specifically for Chrome. One said that it can be hard to test against Safari because of the lack of available tools. In your response, you literally rewrote what they actually said into what you wanted to hear. And now you are using that fictional quote as evidence that developers agree with you.
For what it is worth, I have some basic web apps on the website I run, built with my meager skills. I don't develop specifically for Chrome. I develop to some specific, simple standards. I also don't test against Safari, because I don't have the tools to do so. If Safari can't handle my code, that's a problem with Safari, not a conspiracy between Google and myself.
And yes, users SHOULD move to browsers that better support web standards. Why would Safari, or any browser, ever improve if developers had to ensure their products worked well on Safari?
We’re closer in agreement than perhaps you think. I largely recommend coding to standards, too.
I do take issue a little with the specific example of WebGL, for two reasons. 1) That’s around the time Apple was deprecating OpenGL (and OpenGL ES) on their platform in favor of Metal, and Apple’s removal of WebGL was a side effect of that (the reintroduction no doubt implements an OpenGL ES compatible mode for Metal). It’s understandable, given the larger platform moves, why Apple dropped it. 2) I don’t see much use for WebGL (or much use of WebGL) among the general population of web developers, it’s a niche technology that would mostly be of use to game developers and I’m not sure how compatible the major 3D engines (like Unity or Unreal) are with WebGL. (And I wouldn’t bother coding a game in raw WebGL, WebGL was probably my least favorite section of the class. It’s too low level to really create anything in by hand, even the older version of OpenGL felt easier to use.) If Safari had dropped support for a more significant standard, I’d probably concede your point, though.
Did Safari actually drop WebGL support? I definitely used it on an up to date iPad Pro in November 2017 in Safari, I can date it to then because that’s when I took my college computer graphics class. I may have been using it in a canvas tag, I don’t remember for sure. But I definitely used WebGL on an up to date iPad Pro in late 2017. Perhaps they announced it in January 2017 but removal didn’t occur until June 2018 with the next major iOS release after the one I used in my class?
And, I ALSO gave an example where that wasn’t true. And, to go one better, I chose an even MORE extreme case!
What is the outcome from not testing against non-chrome browsers? The outcome is that the developer has no idea if it will render in Safari. If a customer emails indicating they’re having a problem with the website and the developer says “Thanks for the input, we’ll work on it” that’s one thing. If the developer says, “Use chrome”, then as I indicated, they may not want to say ‘I coded specifically for chrome’ but what they’ve done, as witnessed by the fact that the site doesn’t load properly in Safari, is code specifically for chrome. Which is also what anyone coding in chrome in such a way that it yields a site that doesn’t work in Safari. And, THAT IS NOT A BAD THING. It’s their sites, they have total freedom to do what they want to do with them.
Some folks like to dictate what others should do, some folks would rather others make their own decisions based on what they think is important. It’s a big world and enough websites for everyone! If Amazon thinks it’s important to THEM that their site loads properly in unsupported versions of Safari, they’re free to have their devs do so.
I’d say it’s no excuse that Google makes developing for chrome easy.
They have a vested interest in as many devs pushing users to Chrome as possible. And, like you’ve said, the success they have at making the web a chrome only place is up to the developers desire to want that to be a thing (and, in this thread it’s been mentioned several times that Apple should just use chrome’s engine).
Oh man, I knew rapid versioning was kinda dumb, but I didn’t realize it was THAT dumb! I hadn’t heard about that, I’ll have to look into it.
Edit: Looked up the article. I’d hope most developers parsing user agent strings are using a regex something like “\d+\.” for major version number parsing instead of “\d{2}\.” (Well, the latter would break on single digit version numbers, too.) If you make too many assumptions about about the data your regex is supposed to parse (especially on something as relatively well-formed as user agent strings), then that’s on you. That’s also one of those situations where your custom bespoke solution likely won’t be as good at handling edge cases as a solid third party library. (Time Zone math, especially daylight savings time aware Time Zone math, is another one of those.)
Last edited:
Safari is the worst for conformance that's why a lot of sites don't render properly. Apple needs to drop WebKit for Blink.
https://html5test.com/results/desktop.html
https://html5test.com/results/desktop.html
If there are a lot of sites that don’t render properly, I’m sure you can provide a link to one that doesn’t.
When one looks at the list of things that Chrome supports that Safari doesn’t…
The HTML5 test - How well does your browser support HTML5?
The HTML5 test score is an indication of how well your browser supports the upcoming HTML5 standard and related specifications. How well does your browser support HTML5?
html5test.com
Last edited: