Why IE9 is a Web Designer’s Nightmare | Hacker News Reader

Why IE9 is a Web Designer’s Nightmare

sixrevisions.com

40 pointsrodh25731 comments

A very forceful title that is unfortunately backed up by very little in the article content.

The article made one valid observation: the historically extended update cycle for Internet Explorer could be a problem when combined with the incomplete implementations of CSS3 and HTML5.

Anything bad in the title or comments about Microsoft or IE results in automatic upvotes regardless of content.

Do you have a reason for that observation? I agree with it, but I don't have a good reason. People of pro-microsoft opinion often cite the relatively tiny numbers of Mac, Linux, etc users all the time, especially when giving rationale for staying on Windows, or giving a reason for why Windows attracts the most malware.

But you can't have it both ways: if there's only a tiny minority of non-Windows people, they won't have the juice to generally influence Microsoft-related article's ratings. It you want to have it both ways (Windows dominates to the exclusion of everything else AND those pesky Linux/Mac/whatever users are everywhere, upvoting critical articles), you've got a world view with a very basic contradiction at its core.

So: are there more Mac/Linux/Minix/LoseThos users, or do even the Windows users generally dislike Microsoft?

Just because there are far more Windows users than Linux users (not that that has ANYTHING to do with this, lots of Windows users use Firefox for example) doesn't mean the minority group won't be able to massively upvote something out of control. It's not like everyone votes, only a small minority vote - and guess who are more likely to vote? People who feel strongly (for or against). So if there are more haters than there are people who care to fight the haters, guess who "wins" the vote?

I suppose that I'm assuming some kind of proprotionality in upvoting or downvoting, maybe even two varieties of proportionality.

First, I'm assuming that Windows-users are represented proportionally with the general population in the Hacker News population.

Second, I'm assuming that Windows-users and Linux-users would be equally inclined to up- or down-vote any given story. That is, the proportion of voters (your "small minority" that votes) is the same in Windows-users population and Linux-users population.

Your point about Haters vs Haters-fighters is well taken.

I would imagine the average hacker on HN would be more anti-Microsoft than your average man on the street.

I stopped reading this as soon as I got to the "negative point" that Microsoft thanked JQuery for supporting IE9.

Libraries such as JQuery of course have to support major versions of all browsers. In particular in this case much of the work was disabling workarounds required in earlier versions of IE!

Mozilla actually run the unit tests of most major libraries as part of their regular test suite to make sure that the Internet will by-and-large still work on any new release. So I think the author has a valid point here, new browsers shouldn't break existing libraries this badly.

Webkit (recently) and Opera (last year) also incorporated jQuery's test suite into their own.

Perhaps, but as the parent post had stated: the changes to JQuery were to get rid of workarounds for past versions of IE. If Microsoft kept bug-for-bug compatible with older versions of IE, they'd be lambasted for that. Now you're lambasting them for actually fixing things. Rock, meet hard place.

This article is trash. In one breath, the writer lambasts IE9 for supporting a compatibility mode (while not mentioning that other browsers also have a quirks mode) --- and then goes on to criticize the decision not to support XP, the most insecure operating system on earth. The article is a classic example of what results when you start with a conclusion and work back to the evidence.

Windows XP is certainly not the most insecure operating system on Earth.

…still in mainstream use.

That would probably be Damn Vulnerable Linux (http://distrowatch.com/table.php?distribution=dvl).

>lambasts IE9 for supporting a compatibility mode (while not mentioning that other browsers also have a quirks mode)

I thought that quirks mode was to accommodate non-standards compliant markup. Also, isn't compatibility mode to fudge things to work by accommodating errors in the rendering engine?

You comment is also trash.

I have a Win XP desktop (besides my Ubuntu one, on which I spend most of my time lately). It's legit (paid money for it), all the software I care about runs on it and I never have had it infected with a virus (in 3 years since I bought it). So tell me, why the hell should I upgrade from the most popular OS on this planet and PAY for that upgrade? Because it doesn't conform to the perfect world you want?

And if IExplorer 9 won't run on it, that's fine. There's always Chrome, Firefox, Safari and Opera.

And about compatibility-mode, having that mode in IExplorer 9, but disregarding compatibility for Windows XP is inconsistent. Not to mention their competition is doing the exact opposite.

While I don't particularly agree with him on the compatibility mode, I don't see the contradiction. His argument is that getting rid of compatibility mode provides an incentive for innovation and eliminating old unsupported code from the web. The more people that use IE9 (without a compatibility mode) would put pressure on site owners to make sure their code met current standards. Supporting IE9 on XP would increase that number.

Anyone remember the hoopla when IE8 was initially going to follow strict standards mode? They had to backtrack because site owners were up in arms about changing millions of sites to render properly. IE9 uses a lot of technology and APIs that don't exist on XP, recoding all that for a 10 year old OS so that people can continue to keep running it beyond its lifespan(and holding back progress in other areas) doesn't seem to be a good thing.

While I think Microsoft's decision not to support XP makes sense from their perspective, I also think it's perfectly fair for others to complain about it (I don't think people should be obliged to adopt the POV of platform vendors, whether it's Microsoft or Apple. We saw this before with some of the defenses of Apple's language restrictions in their App Store developer agreement. I also think PC users should be free to upgrade, or not, on their own time).

But I wish, instead of helplessly complaining about a Microsoft decision that's unlikely to change, they'd focus more on the opportunity this presents to promote other browsers.

I am not a fan of IE (6 in particular) but all these bashing of the latest IE 9 seems to not have valid arguments. Modern browsers by whose standards?

If you disable JS in all browsers, IE9, Firefox, Chrome and Opera render the page the same. This leads me to believe the web designer is writing brittle javascript somewhere. If you look at the JS, is uses conditional comments based on the version of JS, so I would probably start looking there.

In the end I don't think we can expect to push the web development field and browsers forward without dropping old cruft. From using IE9, it seems they've made a big jump with this version. Unfortunately some old code may no longer work the same, and for that there is the X-UA-Compatible meta tag.

Let me answer this as someone who writes non-brittle Javascript code on a 1 year old web app (e.g. I have more unit tests per LOC in my Javascript than most of the Java code that's in the underlying app. AND I run my test suite on all major browsers regularly).

Even if you write solid code, using good architecture, and thorough testing you will find that if it passes on FF then there's a 98% chance it also passes on Chrome and Opera... but maybe a 60% chance that it passes on IE8 (requiring HOURS of work to investigate and fix), and there's some abysmally low rate of success on IE9 from having run these tests against the Beta earlier (we don't plan to announce IE9 support for a while).

So while you are correct old cruft will have to go, brand new solid code shouldn't have to go as well... and that's unfortunately the consequence of IE9. :'-(

There has been a notable amount of hate on IE9's lack of XP support, so I thought I'd ask. Has anyone happened to read ~why~ IE 9 isn't supported in Windows XP? People continue to be angry that it's not, but has anyone actually checked why it's not supported?

Don't quote me on this, but I thought it was because IE9 made heavy use of Direct2D from DirectX10 or 11 or some such which XP doesn't have access to.

Firefox 4 uses the same Direct2D stuff but has a software fallback. It could be argued that IE9 should also have a software fallback, but they don't.

There must be some kind of fallback because IE9 runs fine inside a virtual machine with disabled graphics acceleration.

IE9 does indeed have software fallback. You can toggle the setting from inetcpl.cpl -> advanced to disable hardware acceleration if you wish.

The single paragraph that's actually worth reading from this article:

> I quickly downloaded it and began the installation process. That went well. I then performed the obligatory Microsoft reboot — it’s 2011 and the software still needs a computer restart, but whatever, I’ll live — and then I opened up the browser.

If someone has a TLDR version of this article, sharing that is much appreciated. I tried reading this twice, but I am failing to understand the author's point(s) ... What exactly are the problems with IE9?

Anyone know a good IE 9 emulator?

Keep an eye on IETester: http://www.my-debugbar.com/wiki/IETester/HomePage

Right now they just have IE9 beta, but I'd imagine they'll have full IE9 support soon.

I run XP on my machines, so here's what I did:

I downloaded the free (and legal) Microsoft Windows 7 enterprise trial ISO:

http://wb.dlservice.microsoft.com/dl/download/release/Win7/3...

I installed the free VMWare Player (actually I already had it but needed to update to support Windows 7 more natively)

http://download3.vmware.com/software/vmplayer/VMware-player-... ( or http://www.vmware.com/download/player/download.html )

I used one of the free vmx generator tools to make an image ready for Windows 7 and set the ISO as the CD drive

http://www.easyvmx.com/supersimple.shtml

Installed in less than 30 minutes (including VMWare stuff) and then I can test in a pure Windows 7 default environment.

The great thing about the Enterprise trial is it lasts for 60 days, then you can "rearm" (google it) for a couple more times, but even when it's expired, you can use it for an hour before auto-shutdown which is plenty of time to test pages.

Reply on news.ycombinator.com