MS crippling IE
Well, they're certainly experts at that - been doing it for years!
The forthcoming Windows Phone 8.1 Update will include plenty of tweaks and fixes for Internet Explorer 11, and while many of them are designed to improve the browser's web standards compliance, Redmond says others are designed to do exactly the opposite. The problem, to hear Microsoft tell it, is that IE11 on Windows Phone is …
While I agree 100% with them crippling it...well more like just saying to hell with the standards we are going to do it our way and screw what the standards actually ARE, I cannot help but agree with them on this. I don't like that they are adding yet one more thing to go wrong with the browser when there are so many other attack vectors already in the bloody thing, truth be told I cant fault them for this. Why not do what they need to do so that people can actually view the website they happen to be visiting and have it actually work. In this I have to place, at least, some of the blame on the web developers who cannot be assed to design according to standards and test against at least the top 3-4 browsers, at the most current version available anyway, on the market.
Im no M$ shill but the blame has to go both ways in some extent.
Now if M$ was developing all of the webpages then I would be not so much on the wagon train as driving it. Then again I wish someone would develop a phone, be it land line or cell, that I could reach through and strangle the person on the other end due to their utter incompetence and lack to bloody listen if I ask them to do something for me so that I can fix their damn problem instead of repeating myself over and over until I am blue in the face and at serious risk of an aneurysm.
Then again screw it, you know what I don't care as I don't have a phone with Mobile Exploder 11 on it.
People can't view the website on IE and have it actually work. Websites serve up the IE version because MS spent years going their own special way, now they've gone and removed years of compatibility cruft in IE 11 Mobile the lipstick has come off the pig. The only way MS can fix it is abandon their user agent ID altogether.
Exactly - it's not that IE11 is more standards compliant than the competition (most of them have been scoring 100% on the ACID tests for several years), it's that every website on the planet has had to commit all sorts of internal atrocities every time it sees a user agent belonging to IE. The fact that it looks fine when they change the user agent to Gecko is proof of that.
It's possible that some of this mess can be fixed by running a newer jquery, once jquery supports IE11 fully anyway (apparently right now it doesn't detect it quite right and it falls back to a quirks mode for an earlier IE).
Websites don't have to "commit all sorts of internal atrocities" when they see a user agent belonging to IE. That's entirely the author's choice. And any author who wrote code for a generic "IE" user-agent, assuming that future versions would continue to support the same non-standard quirks as current ones - deserves all the s**t they get.
In 2011, when I rewrote our company's website, I found I needed 4 different stylesheets: one each for IE6, IE7, IE8, and one for (every other browser including IE9 and above). There was no IE-specific code in the last of these; if there had been, I'd have put it in a separate sheet for IE9, on the assumption that IE10+ probably wouldn't need it. What there were, were prefixed style rules, e.g.:
-moz-border-radius: 8px; -webkit-border-radius: 8px; border-radius: 8px;
But all those proprietary prefixes were only ever meant to be stopgaps. It's always been understood that, when a browser learned to support 'border-radius', it would use that rule and ignore the '-moz-border-radius' one.
So there's a perfectly clear, smooth upgrade path. Now as I understand this story, MS has decided that if there's a '-webkit-border-radius' rule, but no plain 'border-radius', then it will support the -webkit version.
As a developer, that strikes me as - bloody annoying, frankly. But from MS's point of view, I can see the attraction: it makes their browser look more like what people are used to.
Next time try reading the article - many of the fixes are due to lazy web developer's using safari-specific elements rather than the standardised elements, not due to older IE compatibility.
It seems that Chrome/Safari are becoming the new IE - implementing features ahead of standardisation in an attempt to define what is actually standard.
Microsoft made it damned difficult (nigh on impossible if you weren't technically astute) to test against the various flavors of IE. And that's been like that since the beginning of the web. One reason why I liked, and heartily recommended to webdevs, VMWare from the beginning was that it allowed me to have multiple testing targets. I could even, in the really offbeat corner cases, compensate for OS versions as well. And that's true still today and not just Microsoft although they get a huge amount of the blame in this game.
Some are well thought out, cutting down on the baroque riff raff the regular page throws at you.
On the other hand, I've encountered quite a few that are plain awful.
Some, such as ons**pe, take forever to load. No excuse for being so incompetent.
Others disable pinch to zoom. Really? You've lost my pair of eyeballs, because I can't see what you have to show me!
Mobile device screens are small to begin with, don't assume that everyone has 20/10 vision.
</rant>
Can't zoom? Blame Apple for this nonsense which originated on early iPhones:
<meta name="viewport" content="initial-scale=1.0, user-scalable=no">
You need that if you want a mobile site to fill the screen width and only scroll vertically (because 4-way scrolling on a phone is pretty horrible). You can say user-scalable=yes, but users will zoom in by accident and then wonder why everything's broken. What a trainwreck.
Oh, apparently Android browsers ignore this... but web designers all use Apple.
Unfortunately pinch to zoom does more than just enlarge things - it starts mangling coordinate systems and triggering events and overflows. So people sometimes disable it to prevent a site from breaking in other ways. Blame this one on the 'pinch-to-zoom' inventors not caring that it breaks stuff that was already working fine in other circumstances.
What is one person's solution is another person's problem -
It's like building a house. A house with stairs is not 'compliant' for wheelchair users. A house with elevated shelves is not 'compliant' for short people. A house with windows... heliophobia!! etc etc.
Unfortunately pinch to zoom does more than just enlarge things - it starts mangling coordinate systems and triggering events and overflows. So people sometimes disable it to prevent a site from breaking in other ways.
Rather than redesigning their sites to take into account the fact that HTML was designed from the beginning to flow.
Even El Reg is guilty of this. See those giant empty bands on either side of your widescreen display? Those are there because web designers are too stupid to make the content pane dynamically resize to fit the screen. So readers have to scroll more and read less at a time, because most sites are still based on a horribly ill-fitting fixed-width model of the antiquated print industry.
The only reason so many web designers still have work is because they keep having to come up with more shite hacks to make their original shite hacks compatible with new standards.
See those giant empty bands on either side of your widescreen display? Those are there because web designers are too stupid to make the content pane dynamically resize to fit the screen.
No, the whitespace was there in the early days of the web because everyone used tables with fixed sizes for layout. This meant that regardless of what sized screen the content was viewed on, the size of the content remained static.
Content on modern websites does scale appropriately with the size of the viewport. However you're unlikely to find many websites that are wider than they are tall, just like you won't find any magazines or newspapers with that layout either. Magazine editors developed what's known as the grid layout and it has been almost universally adopted by web designers. It's a system that helps maintain proportions that are pleasing to the eye. If you have lines of text that are too long, they become unreadable. Likewise, if you have too many separate images horizontally the eye has trouble keeping the rows separate and it appears as a cluttered mess.
"Rather than redesigning their sites to take into account the fact that HTML was designed from the beginning to flow."...
...And there is the problem - It was designed in the previous millennium.
Clients do not hire designers who spend their time worrying about W3C nonsense like 'flowing'. The reason is because most of their portfolios look like shite.
Also, the designers are not the devs unless you work in a tiny company. Designers make stuff look good, Devs try to make it work on the target browsers (which is never all of them, nor should it be).
"Even El Reg is guilty of this. See those giant empty bands on either side of your widescreen display? Those are there because web designers are too stupid to make the content pane dynamically resize to fit the screen. So readers have to scroll more and read less at a time, because most sites are still based on a horribly ill-fitting fixed-width model of the antiquated print industry."
there has to be an upper limit to how wide a text column will go. have you tried reading text that goes from one edge of a the screen to the other? it's a nightmare. it's what column were invented for in the first place - so the eyes can track where you are in the text easier.
>Others disable pinch to zoom. Really? You've lost my pair of eyeballs, because I can't see what you have to show me!
Android > Chrome > Settings > Accessability > Force Pinch to Zoom (override a website's request to disable Pinch to Zoom)
> Be the biggest cause for "optimizations" and assorted cancers on websites for over a decade
> Suddenly take the moral high ground, whiteknight the industry and emit a sickening ooze of outrageous statements of self-praise
Pretty sure many cow-orkers from the Only Democracy in the Middle East are working over there. It's the only causal explanation that is possible.
So, to recap: Micros~1 has for years...no decades, been end-running web standards, foisting its own home-brew garbage on lazy and/or ignorant webbies who only wanted to use their broken tools and methods. Now the shoe is on the other foot, and they get a taste of what the rest of the world has had to put up with.
/me can't stop laughing.
But please, Neil, stop painting Micors~1 as the injured party here. Microsoft didn't "[hold] its collective nose and come up with a solution", the did what Mozilla, and Opera, and the long departed Netscape, and others have had to do: Adapt to what the real world is doing.
Once upon a time, if you wanted to see what browser someone was running in a CGI script, you inspected HTTP_USER_AGENT.
It was a string of the form: RendererName/Version (other detail)
Dead simple. So Lynx was Lynx/whatever, Netscape (with the Mozilla engine) was Mozilla/whatever, NCSA Mosaic was NCSA_Mosaic/whatever… etc. The "other detail" was usually platform-specific information such as OS.
Netscape became dominant, and so everyone wanted to be like them. So they started announcing themselves as "Mozilla". Internet Explorer has been doing this for years. In fact from day one, despite being, based on NCSA Mosaic.
So the industry settled on using Mozilla/whatever, then in brackets, announcing their true identity. Now I see, we're going to be lied to yet again, MSDN reports the following user agent string for IE11:
Mozilla/5.0 (Windows NT 6.3; Trident/7.0; rv:11.0) like Gecko
"like Gecko" but not Gecko. What's wrong with InternetExplorer/11.0 (Windows NT 6.3; Trident/7.0), Microsoft? Scared to use your real name in public?
You've got it completely wrong. It's not that MSFT doesn't want to use the correct user-agent for IE11, it's that if they do, the modern web goes "der, what's that Chauncey?" and fails to detect that this is an actual standards compliant IE (snicker), as contrasted to every other IE.
Hmmm...
If changing the User Agent string "fixes" things, then it's not that IE11 is "so perfect", but rather that Web sites haven't got a clue what it is, so send something basic
And, as for the 3 example image shown, am I the only person who actually prefers the left-hand one? It contains more information on the screen! (I don't need "pretty backgrounds" to distract me.)
Shall we talk about the real problem now? The crucial cause of Microsoft's current browser issues? It is, of course, Internet Explorer's long and disgraceful history of weirdness and incompatibility and the effect this history has had on the way web developers work and think. No-one with even the slightest web-design clue ever trusts a Microsoft browser to behave in a sensible or obvious way. Anyone with any experience or expertise automatically expects to code work-arounds and weird proprietary hacks to accommodate the Browser from Hell. Often when even the weirdo hacks fail to tame it, we fall back on giving the Microsoft browser degraded content while we hand Firefox and Safari and Opera and Chrome and Seamonkey and all the others the real thing with all its features. This has become almost reflex, an automatic, habitual practice any web designer follows to greater or lesser degree, largely depending on the amount of time and money available for any given project.
So now that Microsoft has (so they say) finally released a browser that works properly (do we believe them? there is another question), it has discovered that millions of web sites don't trust their product and do all the weird stuff that Microsoft forced them to learn to do with past products, and this latest IE can't cope with all the weird stuff. How sad. It has to pretend to be Firefox, sort of, and put itself through hoops. That's fine by me. Anyone who has had to waste years of his life hacking and mangling perfectly good code just so that the Browser from Hell wouldn't fail as miserably as it mostly did most of the time will shed not one single tear for the bastards.
Note also the other reason for their current practice: their barely-higher-than-zero mobile market share. Now that they are (in this segment) a global irrelevancy, no-one does special code for their latest product anymore and it has to stand on its own two feet, if it can.
What, so now they are actually doing the right thing(tm) you are beating them around the head for it ? If so why should they ever comply with a standard again ?
Unlike the "good old days" I do not see any "hacks for getting around IE10 or 11 problems on the more influential web design blogs- its now all how to build a better responsive site etc etc
Provide some links to the tests, and we'll do so, and remember that feature checklist scores are not the same thing as having standards-compliant rendering.
I consider IE11's rendering to be standards compliant, in that I've never had to take any action beyond using standards-compliant CSS rules to get IE11 to render my HTML content exactly as Safari/Chrome or Firefox.
My own web browsing is done 50:50 between IE11 and Safari, and in the rare cases that I see something broken on IE11 it's because someone blindly cut-n-pasted "-webkit-:" CSS rules into their stylesheet. The old problem of sites erroneously serving an IE6 document are actually quite rare, because generating different content based on UserAgent strings requires deeper knowledge of HTTP than most current web developers have, and it's something that a lot of web frameworks don't make easy. (most sites I've seen prefer to use Microsoft's own <!--[if IE]> macros for doing IE-specific CSS; macros which IE11, correctly, ignores)
The problem today is that a lot of designers just assume that "renders okay in WebKit" is the same as "complies with the standards". Sure, competent and diligent web developers know how to do CSS rules that don't require a specific browser, but why should website design be the one profession where every single practitioner is competent and diligent?
"Microsoft is saying the other major browsers are all wrong and not rendering web pages correctly but their browser is the only one in the right. Okay ..."
No, they're saying they have to pretend to be another browser because Web authors are failing to use the standards correctly, and instead coding to a particular browser.
Try reading the article next time.
I understood this article to say that MS developed a browser which is 'fully standards compliant', which I take to mean HTML5, and then they realised that few web pages actually use HTML5, so they had to put all the other code for other versions of HTML back, just like every other browser on the planet.
So in actual fact, this is a non-story?
No, the story is about how the Microsoft team, despite being late to the party of browser standards compliance, still insists on spinning it as though it was the fault of other browsers.
Goebbels would be proud of the tech evangelists at Microsoft. Then again, astroturfing is a big part of Microsoft's marketing activities. Hardly surprising.