Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
IE 8 is the new IE 6 (infrequently.org)
98 points by functional-tree on Oct 11, 2010 | hide | past | favorite | 52 comments


This complaint is as old as the hills. You may as well slap an obnoxious "Designed for HTML5" button on your site. The cost of having the world's largest addressable audience of any development platform is that web developers have to deal with the fact that it varies a lot. Them's the breaks.

Edit: because I hate comments that are purely negative, my suggestion for getting users to the latest technologies is to use libraries (like YUI and jQuery) which do the hard work of implementing newer features in backwards compatible ways, but use the modern feature if it's available. That way the difference between IE8 and a more recent browser becomes one of performance rather than the binary absence of functionality. Users can that way see what they're missing and have a reason to upgrade that's more compelling than the designer petulantly insisting their browser sucks.


The thing about all this I don't get is that IE8 is not anywhere near as bad as IE6 was. Like you mentioned, using libraries like YUI, or jQuery really mitigate a lot of the problems.

If you also plan a bit up front and have a competent designer to help out with the css definitions and strategizing (like avoiding gradients for now, fun/cool as they be), it's really not that big a deal.

I'm working on a new site and I've literally done nothing other than have a designer help out with the CSS and the only difference between Firefox, Safari, Chrome and IE8 are css drop-shadows, and rounded corners. Purely cosmetic, unlike IE6 which if you didn't account for it could stop your app in it's tracks.

To me IE8, as much as I would never personally use it, is a huge improvement from IE6. Even over IE7 it's hugely better, easier to deal with.


IE6 was a gigantic improvement over IE5.5, too. The problem is not there, the problem is that people are still stuck with IE6 today, close to ten years later...

IE8 seems not so bad today, but without any support for html5, css3, canvas, video or websockets, what will you think 10 years from now ? That it sucks: it will force you to have Flash/Silverlight fallbacks to do something as trivial as plotting a graph or displaying a video. Yet many XP users will still be using it, so you will have to support it...


MS with IE6 had a cavalier, take-it-or-leave-it approach to features/standards and sought to subvert/control the web/web browser-market.

MS with IE8 (an IE9) are actually trying to make their browser standards compliant and while we might not like the pace of the behemoth, it's moving in the right direction and this is a good thing.

2010 is a much different landscape than 2005 as far as the browser market goes as well; MS/IE are no longer the dominant juggernaut that can attempt to dictate standards.

More than anything IE releases are tied to MS's OS releases - IE8 has become the largest slice of the IE pie because of Win7, not because of features or any push by MS. As MS increases their release cycle of both browsers and more importantly, OS's, the likelihood of a IE6 repeat with IE8 is extremely thin. MS can't sit around on any version of a major product like they did with both XP and IE6.


Considering the pace of browser development, I would say that's true of any browser that doesn't auto-update. Ten years from now, if we still have this bizarre thing called the internet, IE9 will be the dinosaur that everyone hates. Or maybe even IE10.


Yes, it's an improvement. But the killer quote for me is " in 2015, you’ll still be thinking about a browser that doesn’t support <canvas> or <video> and doesn’t even have a JITing JS engine.". I don't care how much better IE8 is than 6 and 7 - that sucks.



How embarrassing. On the other hand, the only clue that he's one of the authors of dojo is that he links to dojo projects in the sidebar a lot, but he also links to chromium and webkit. Should I assume he's one of the authors of those too? (He could be, I've no idea)


Yes, he's the principle developer of Google Chrome Frame and works on the Chrome team at Google.

His most recent post is worth reading, as it expands on his position: http://infrequently.org/2010/10/story-time/


A better way would be to make the basic functionality work in browsers, then make it pretty and really nice to use for the rest.


> I’m developing for the future exclusively

Weird, I'm developing for my employer, and by extension for our clients.


We should let the OP "develop for the future" and keep our sites providing the content for everybody who has any browser.

After some time we can compare the notes.

(I wonder how he imagines that I would even try to access his site with anything else but with Opera Mobile from my phone).


In my personal time, I goof around with all the new stuff that I'll get to use in 12 years when my clients that I work for catch up.


I'm developing for the employer and clients exclusively.


Yes, but IE9 will also be the new IE8 eventually. Perhaps automatic embedding of unicorns becomes the new css-gradient and textshadow in 10 years. Designing for the minority isn't a very good solution.

A better solution:

embedded SVG rendering within <img> and <htmlelements> cross browser. Then you can assemble ANY trendy graphical element without having to wait for people to continually agree on, release, and implement css4, css5, css6, etc.

Other benefits:

- it's an XML format like HTML

- Machines can now scrape understand and generate pictures too

- Vector graphics will scale great with different zoom\resolution settings

- Designers can use tools like Adobe Illustrator, a more logical approach than css-versionX hacking that fits the problem domain better


> Yes, but IE9 will also be the new IE8 eventually.

not necessarily. IE6 wasn't the new IE4, and IE7 wasn't the new IE8.

IE9 probably won't be the new IE8 eventually for a number of reasons:

* Unless MS stops the IE project again (doubtful at this point), IE10 should arrive pretty close on the heels of IE9, say two years (similar to the delay between IE7 and IE8, or IE8 and IE9)

* IE10 will very likely be compatible with Windows 7 (and Vista, but that's not as big an issue as it's been far less popular than 7)

* Considering the two-steps improvements done in IE8 (CSS) and IE9 (more CSS, JS) the differences between IE9 and IE10 will very likely be much smaller than between IE8 and IE9, or IE7 and IE8.

The problem here is stagnancy. IE8 is the last browser of its line compatible with Windows XP, and Windows XP has a huge install base and huge former popularity. This means the (MSIE, Windows XP) platform is now completely stagnant and it will not disappears until Windows XP does.


  >  IE7 wasn't the new IE8
Is there an ActiveX control for time travel?


Bah, I of course meant "the new IE6".


>Yes, but IE9 will also be the new IE8 eventually.

I think the problem is that microsoft is tying the browser version to the OS version. Will you be able to install IE9 on XP?


I don't think the underlying problem has to do with specific companies and applies to older versions of Firefox as well. The underlying problem is:

(Demand\Rate of requests for specification changes) >> (rate of widespread implementation)

...ensuring there will always be a very narrow audience for the latest features.

While it's good to increase the rate at which new specifications are implemented, it's largely impossible to do without authoritarian measures to A) control the programmers working on every web browser B) control specification committees to agree faster. edit: or C) make everyone use the same web browser

Another approach is to DECREASE the rate of [edit: demand for] specification change. I think a large part of the problem is our expectations from the CSS specification. We are trying to expand its problem domain to include the construction of graphics and graphical effects, something which it was clearly not built for on a syntactical level! It might be better to convince browser teams with limited resources to better support long existing open standards to handle the dynamic creation of graphics with more logical syntax. Because the rate at which specification revisions are required will be drastically reduced. Whereas the process for creating new graphical features to your website via CSS is to submit it to a committee and wait 5years, it is done within the syntax of SVG. Making it a better allocation of resources.


> Will you be able to install IE9 on XP?

No.


The best way to help is to port Chrome to Windows 2000.

Firefox is too slow to be usable on most Windows 2000 boxes, IE newer than 6 isn't supported, and Chrome isn't supported, which makes IE6 the best option.


Windows 2000 doesn't appear to have any significant market share.

http://www.w3schools.com/browsers/browsers_os.asp


Much higher that win98. I wonder if my 12yr old daughters machine is the last win98 machine on the internet?


Who uses IE6 on XP? Pirated copies of XP? That's bizarre.


How about half the population of China? http://gs.statcounter.com/#browser_version-CN-monthly-200909... Plus folks at many organizations with conservative IT policies.


I know I had to do some voodoo to make a website we setup work right for a client because all of their internal windows computers were still running IE6


Chrome is hardly stable on pages with a ton of flash content ...


Opera runs well on Windows 2000


Boxes still on Win2k won't have much memory, and Chrome is really memory-hungry, so it isn't going to help.


This will always be a problem. The corporates and most of the business will always use whatever comes with their OS, for a simple reason that they don't have administrative rights to upgrade their browser, or simply because its just an extra overhead for the corporation to upgrade thousands of their workstations.

I can't think of any alternatives to this problem. I am pretty sure I will still be considering IE6 while developing websites in 2015, let alone IE8.


There is a lot of pressure mounting for even the most slothful corporations to upgrade, both their browsers and their legacy web apps that [currently] only run in IE6. This is partially to do with the growing prevalence of enterprise-ready web apps & services, but also with Microsoft's insistence that they really will kill XP and IE6 in 2014. The same thing recently happened when they finally killed support for XP SP2 (just last month) and there was a big run to upgrade millions of corporate PCs to SP3 (which forces an IE upgrade unless you do all sorts of not recommended funky hacks to prevent it).

As an aside, most companies for whom IE6 is an elephant-scale problem are using both AD and MS SCCM for user configuration management & group policy controls. CM makes it trivial to deploy stuff across the board, including complex software like OS service packs [and browser upgrades]. I posted before that my own org is in the process of upgrading everyone to IE8 right now, but we switched to Firefox as our standard browser about a year ago and are currently testing the new Chrome MSI (woohoo for GPO support), since even though Google's interpretation of the HTML5 standard includes a few things that aren't actually in the draft yet, it works much better with Google Apps.

You won't have to worry about IE6 in 2015.


Not to mention that IE's turn-around time will almost always be longer than, for instance, Firefox's, or Chrome's, etc., since Microsoft is inherently less flexible than the others.

Oh, and don't forget that a lot of proprietary corporate apps are written to work ONLY in IE6. I wonder if a "legacy" mode for new versions of IE would alleviate part of this problem. (And yeah, I know that there's a big fat bag of poisonous acid-spitting kitten-eating snakes that this would open up.)


The legacy mode is exactly the problem. They have an IE7 legacy mode in IE9, so upgrading once you are on IE7 or up is generally not that big of a deal. I still fail to recognize any other reason than corporate decision making for why IE8 didn't have an IE6 legacy mode. I can understand they didn't understand the issues when they released IE7, but by the time IE8 got shipped they should have known better.


Firefox and Chrome would have the same fate, if they are as widely used as Windows/IE and comes default with an OS(Windows killer), like IE.


Which fate? Slow release cycles, or apps written specifically for a single browser?


Fate as in, if Mozilla came up with a Windows-killer OS and all the Windows user moved to the new Mozilla OS. Than people will be stuck with FF 3.6, even after 10 years from now.


There is no evidence to support this assertion.

The new auto-upgrade model Mozilla has adopted for FF 4.0 (and that Chrome has had for ages) makes upgrade cycles, for all practical purposes, negligible.

Chrome 6.0 was released on September 2. Since then, Chrome 5.0's market share has dropped to nearly 0%, with Chrome 6.0 picking up all of 5.0's share while also absorbing a good amount of IE's share.

If Microsoft instituted an auto-upgrade path for IE and released updates with more regularity, this would slowly become a thing of the past.


This presupposes that a corporation doesn't turn this auto-upgrade feature off. Some companies like to have cut-throat control over their employees computers, despite known security benefits from the alternatives. "It's working now, so it's safer than upgrading and breaking our other stuff."


I haven't tried it but can you upgrade those browsers without administrative rights? I am assuming you can't do it on corporate settings.


On Windows, if the user installs Chrome, it's installed into the user's home directory (local application data or something). This means the user does not need admin rights to either install or upgrade Chrome.


There are several ways to update programs on Windows without needing admin rights:

- You could do it as Chrome does and have a per-user install. I don't think Mozilla wants to do this.

- You could install a service or a scheduled task to do the work for you.

- MSI updates can be installed without admin rights if they're Authenticode signed, and Mozilla does have a valid Authenticode digital certificate.


In Enterprise Company there is no such thing as "Auto Upgrade"


Over half our Firefox traffic is still pre-3.6. Not IE6 bad, but it lacks CSS gradients and such.


Around the time Firefox 3.0 came out, I dealt with a customer complaint that our web app was really slow. Their browser? Mozilla 1.7. No, not Firefox; Mozilla.


As long as features can be simulated adoption will stagnate. As much as we like to think we (the developers) are in control of browser adoption, the simple reality is that we're not. The end users are. The "normals". They don't give a shit if your box-shadow uses a PNG or is implemented using CSS3. They don't know the difference between canvas and flash. Hey! it works right?

I honestly believe that the way to move the web forward is by letting browsers innovate at the expense of standards, and for developers to adopt non-aesthetic API's that will cause functional differences from the stagnating dominant incumbent if there is one at the time (at the moment IE).

In other words, when developers get pissed, they find solutions. When end users get pissed, the switch browsers, or decide your app is not worth it. Our goal is to get that proposition in front of end users as often as possible, and create enough value that they choose the former.


I don't think I've ever met a web developer who thinks they're in control of browser adoption. We're slaves to dumb corporate IT policies and users who don't even understand the concept of a browser, let alone what it would mean to "upgrade" one.


> I’m developing for the future exclusively and asking users of legacy browsers to adopt a modern browser or install Chrome Frame. It’s an uphill battle from here, but nobody is going to bend this curve but us.

I do exactly the same. Unless generously compensated, I refuse to develop for outdated broswers while an easy plug-in solution exists.


I wonder if Microsoft has to support Win XP forever in the eyes of the author.

I don't know of a single reason why you shouldn't switch from xp to seven aside from evil corporate policies.


What about the system requirements? As sad as it may be, a lot of people in under-funded organizations are stuck using computers that are 5-10 years old, which may not support Windows 7.


I saw several netbooks (which are not famous for horsepower) upgraded to win7 without a single problem.


I think it's safe to generalize this as:

IEn is the new IE(n-1)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: