You can improve IE.next

By Aaron Gustafson | November 4th, 2006 | Filed in Action, Browsers, Bugs, DOM, DOM Scripting TF, Microsoft, Microsoft TF, Outreach

If you’ve ever wanted the opportunity to tell Microsoft what they should do with IE next, now is the time.

Skip to comment form

The Microsoft Task Force, DOM Scripting Task Force, and the JS Ninjas have been approached to help give the IE team some direction for improvements needed in IE.next, focusing mainly on JavaScript and the DOM. Together, we’ve begun assembling a list of things we think need addressing (bugs & implementation issues, enhancements in language support, etc.). We’ve tried to keep it balanced, with some things for the seasoned JavaScript developer and some for the folks who are just starting out in the world of DOM Scripting, but we need your help to make sure we aren’t missing anything.

In the interest of time and keeping the process streamlined and organized, we’ve opted to make the wiki invitation-only, but we do need your input. Please have a look at what we’ve put together so far and leave your thoughts/ideas/recommendations in the comments below. We don’t know how soon we’ll need to get this list over to the IE team, so please don’t wait to long.

We will work to incorporate any relevant ideas into the final list, prioritize it, and pass it along to the IE team. Once the list is out the door, we will need to develop test cases for each item on it (some of which we’ve already begun), so if you are interested, please let us know that as well.

Note: We’re not guaranteed to get everything we ask for, but they are listening.

Your Replies

#1 On November 4th, 2006 2:47 pm Gérard Talbot replied:

Hello,

IE 7 bug collections
http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/

I do not understand why we need **again** to create list(s) of bugs and implementation isues, enhancements, etc.. when all this had been already done in 2004 and 2005 :

Internet Explorer Standards Support
with a list of HTML 4.01 failures
http://channel9.msdn.com/wiki/default.aspx/Channel9.InternetExplorerStandardsSupport

Programming bugs
http://channel9.msdn.com/wiki/default.aspx/Channel9.InternetExplorerProgrammingBugs

Feature requests
http://channel9.msdn.com/wiki/default.aspx/Channel9.InternetExplorerFeatureRequests

IE Bugs
http://channel9.msdn.com/wiki/default.aspx/Channel9.InternetExplorerBugs

I have my own bug lists on IE 7 and together with several specific IE 7 bug lists all here:

http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/

Gérard Talbot

#2 On November 4th, 2006 6:20 pm Nicholas C. Zakas replied:

A few things that seem to be missing from the list:

1) Implement DOM 2 Events (event bubbling and capturing, not just bubbling).
2) Implement E4X. It’s a great little ECMAScript extension that is so much easier to use than the XML DOM. I’d almost prefer this over the JS 1.7 stuff.
3) Better custom exception support. If I create my own type of Error object that has a message property, that property should be displayed when an error is thrown (now it just gives a generic message, only using an Error object shows the actual error message in IE).

A couple things I disagree with:
1) I don’t think IE should recognize white space in between DOM elements as text nodes. White space has no relevance between DOM elements (I’d also like Firefox to get rid of this too).
2) I don’t think arguments should inherit from Array. It doesn’t make sense to want to change the order of arguments that were passed in or to change their contents. I’d be more in favor of a base Collection type from which Array and arguments could inherit useful methods, but Array should be read-write and arguments should be read-only. NodeList doesn’t inherit from Array, I don’t see why arguments should. That being said, perhaps access to Arguments.prototype would solve everyone’s issues.

#3 On November 4th, 2006 6:43 pm En webbplats på svenska om xhtml » Medverka till att IE.Next blir bra! replied:

[...] “IE.Next” – alltså som nästa generationens MSIE ! Och enligt vad The Web Standards Project skrev tidigare ikväll kan även du medverka genom att ge dem konstruktiva synpunkter inför den vidare utvecklingen. Så vad gör vi nu då? Klagar vi ihjäl alla deras ambitioner bara för att de är Microsoft och att det tagit ohemult lång tid att komma hit? [...]

#4 On November 4th, 2006 7:10 pm Small Paul replied:

I do not understand why we need **again** to create list(s) of bugs and implementation isues, enhancements, etc.. when all this had been already done in 2004 and 2005

They don’t need a list of everything that’s broken/missing, they need to know what the important ones are. It’s the prioritisation that’s required.

#5 On November 4th, 2006 9:21 pm WaSP Member agustafson replied:

As Paul said, it is the prioritization that matters as well as seperating the wheat from the chaff. I can only imagine how daunting it must be to look at the huge lists of change requests out there and not know where to begin. Part of our task is to figure out which bits are really key.

@Nicholas: Those are some great suggestions. And as for your disagreements with things on the list currently, these are the same discussions we are currently having. We’ll see what happens to these over the next week or two and how the votes go once we begin the prioritization process.

#6 On November 4th, 2006 10:28 pm Georg replied:

Support for CSS table.
Only IE is holding us back now.

#7 On November 4th, 2006 10:39 pm Lachlan Hunt replied:

1. Make sure the DOM is actually properly structured tree, even when the markup has badly nested elements. See this test case.

2. DOM Events. Particularly addEventListener, and please do something to help improve keyboard events.

3. Support for the DOM interfaces Element, Node, etc. so we can use things Node.prototype to extend them and access constants like Node.ELEMENT_NODE.

4. Begin implementing XBL!. The sooner that becomes supported, the sooner it can become a recommendation. But I know this isn’t as important as all the other bug fixes.

#8 On November 5th, 2006 12:11 am dave replied:

They should just give it up, obviously its too difficult for them to get right & I personally don’t feel like waiting till css 99.0 before m$ can have css 2.0 figured out.

I’d say they could just buy flock or opera but they would ruin that as fast as they could buy it.

#9 On November 5th, 2006 1:57 am Gérard Talbot replied:

Ok. Here’s my priority list.

#1 Fix what’s broken before adding anything new. Fix CSS bugs: known, reproducible bugs, testcased bugs reported bugs by web designers.

#2 Fix bugs that creates important problems like crash and cpu maxed due to infinite loop: those are serious bugs which should be addressed first. Yes, there are bugs like that! The MADD (Magical Act of Disapearing Divs) bug reported by
John A. Bilicki III (http://www.jabcreations.com/) on September 2nd 2006 at IE Blog does that in IE7 final!

#3 Fix the many collapsing adjoining margins (see section 8.3.1 of CSS 2.1): it’s now crystal clear that this stops several columnar layout from working as expected and those bugs are major headaches for web developers because other excellent-compliant browsers don’t behave like IE7.

#4 Fix float implementation: float bugs are numerous, well reported and well documented by a lot of people now. Many people use or rely on float to create columnar layouts.

#5 Fix the CSS 1 tests that IE7 fails. That’s right: CSS 1 tests.

#6 Fix z-index: visit http://www.aplus.co.yu/css/z-pos/index2.php
Wrong z-index implementation bugs by Aleksandar Vacić

#7 Fix CSS position-related bugs: there are still bugs related to position in IE7.

#8 Fix CSS inline box model bugs: e.g. a parent inline element should not grow, should not stretch because of its inline nested child.

#9 Fix all of the background (background-image, background-color) related bugs

#10 Fix the inherit and inheritance related bugs

Fix known, reproducible, already documented bugs. And don’t implement something new before fixing known, reproducible and already documented bugs.

Gérard Talbot

#10 On November 5th, 2006 3:00 am zooplah replied:

I’m glad that MS is committed to creating a next version. To me, implementing the event model is one of the most important things now. Also, CSS2 adjacent sibling selectors seem to not work for me in IE 7. I’d love to see that work.

#11 On November 5th, 2006 3:25 am Peter replied:

Yes, they should obviously give up, right after releasing the first major version in what, five years. They’re obviously not getting anywhere.

Obviously.

I mean, obviously.

(Blood pressure… rising…)

#12 On November 5th, 2006 7:17 am Lachlan Hunt replied:

The test case I linked to in comment 7 was the wrong one, since it doesn’t demonstrate the bug. See the correct test case.

#13 On November 5th, 2006 6:50 pm Duncan Ponting replied:

“And don’t implement something new before fixing known, reproducible and already documented bugs”

here, here, let’s hope this message gets through!

#14 On November 5th, 2006 9:22 pm figgy replied:

1. Make it CSS 2.1 standards compliant.

DON’T ADD ANYTHING until you fix all the broken and unfinished stuff FIRST. IE is a half-baked, defective product. It doesn’t need “new stuff”. It needs to work right first.

#15 On November 5th, 2006 10:58 pm Coward replied:

All I want for Christmas is the Q Element.

#16 On November 6th, 2006 12:31 am Akaash Maharaj replied:

I believe that while MSIE’s failure to fully implement CSS2.1 is a grave failing unto itself, it is more worrying still as it suggests a continuing corrosive attitude within MS towards universal standards.

#17 On November 6th, 2006 4:53 am Peter Davison replied:

Internet Explorer is still the only major web browser not to fully support ‘CSS Tables’.

Adding this will make life easier for every Web Designer and Web Developer in known existence.

This should be a priority finish CSS feature for IE.next.

#18 On November 6th, 2006 4:55 am Asbjørn Ulsberg replied:

Full CSS2.1 (includes CSS1) compliance. When that’s set, full DOM2 (includes DOM1) compliance would be nice. That should take care of all CSS and scripting related bugs too, hopefully. Full PNG compliance is also needed. When that’s set, start implementing new stuff like SVG, CSS3, DOM3, etc. It’s still much too early for IE to implement any new features. They have to learn how to crawl before they can maneuver a space shuttle.

#19 On November 6th, 2006 6:33 am Ivo Emanuel Gonçalves replied:

Actual support of XHTML is a MUST NEED. First PNG, then CSS, now XHTML. IE is the cancer of the web. It is slowing implementation of every great web technology there is and there will be.

But hey, how can they support XHTML semantics, when they (the IE team) can’t even figure how to make the tag work?

I’m sounding way too harsh here, but it’s the truth. One can’t even say they are listening, because they aren’t. They simply do not want to fail rendering the Javascript behind so-called Web 2.0 sites like GMail, because then, we’d see a massive exile of users to other browsers.

#20 On November 6th, 2006 9:09 am Jeremy French replied:

Better support for the object tag would be nice.

These fail even with activeX on.
http://www.w3.org/QA/2004/02/object/test-case-001
http://www.w3.org/QA/2004/02/object/test-case-005
http://www.w3.org/QA/2004/02/object/test-case-007

IE should also allow images in the object tag without having to enable active X.

#21 On November 6th, 2006 11:19 am Trails replied:

XML MIME type support is one I would love to see.

Specifically, trying to render xml mime types and not treating xml declarations as a reason to flip into quirks mode.

#22 On November 6th, 2006 12:34 pm Thomas Tallyce replied:

My number one request would be that the IE team spend the time required to refactor out hasLayout.

While the IE7 engine seems to have got rid of a lot of problems, it’s clear that engine still contains a log of float bugs etc. From all I’ve read, until hasLayout is refactored out, these bugs are going to continue to surface. This situation likely to get increasingly problematic as more complex layouts, as well as increasingly demanding specs from the W3C and others, come forward. Piling hack after hack onto a 5-year old engine simply will cease to work.

#23 On November 6th, 2006 2:41 pm Milo van der Leij replied:

What I would like to see fixed, are all the bugs I already reported (with testcases) using the IE7 feedback mechanism. I would love to re-state that list of bugs, but since the feedback site is “temporarily” closed, I can’t. Here are the ones I remember off the top of my head:

* input and textarea fields using a style of width:100% become too large when they have a large value for the value attribute.
* the select object’s add() method takes a number (index) for the second parameter, while the DOM states this should be an object (“before”).
* querying the value of the offsetWidth attribute can change the display/layout of items on the page (new in IE7)

Fix bugs first. Add features later.

#24 On November 6th, 2006 5:59 pm Ned Baldessin replied:

All my improvements are in CSS :
1. display: table/table-row/table-cell/inline-block (all the other designers I talk to rave about the table box model, we love it!),
2. if possible, implement the “flexible” box model of XUL too (Gecko + WebKit have it), very useful to vertically center boxes, an ongoing pain in CSS 2.1,
3. A selection of CSS 3 stuff (no need to implement it all guys!):
– multiple backgrounds,
– multi-columns (god, I’ve been waiting for this since the last century),
– opacity (browser sniffing and applying “filters” doesn’t cut it),
– downloadable typography: if all the other players haven’t implemented your model, I’m guessing it’s because it’s encumbered by patents. Please do something to unfreeze the situation, we are all *dying* for downloadable fonts! If I have to write “text-indent: -9999px” once again my brain will explode!
4. We all love the WHATWG stuff: yay for datagrids! yay for canvas! and storage, and menus, etc etc.

Thanks you IE team, we don’t love you yet, but we just might get around to it in a release or two ;)

-n.

#25 On November 6th, 2006 6:29 pm Tino Zijdel replied:

(I already posted this on quirksmode.org but obviously it didn’t pass the scrutiny/censorship of PPK)

I have very mixed feelings about this, it looks to me as if MS is *again* asking the webcommunity to do their groundwork for them, and with what guarantee that they will actually listen this time?

The mere fact that ‘prioritizing’ is mentioned implies that IE.next is also not expected to be on the same technological level as other browsers which is a bad position to start with. Basically imo it is crucial that IE.next should (roughly) support what other browsers already support now in various areas including a.o. CSS, (X)HTML, DOM and Javascript. The last thing we need is yet another version of IE that is still the least dominor but has different behavior and different bugs from the previous version, we already got that with IE7. It is simply too late to improve IE gradually, we want what we have in other browsers and we want it yesterday and not smeared out over the next couple of years.

I dug up a small list of JS-related bugs that I kept around somewhere just to be helpfull but I have no intention of writing actual testpages for them; MS should be capable of doing that themselves:

String.substr(-1)
Object.toSource()
Element.style.fontSize = ‘inherit’
‘string’[1]
dont reset lastIndex on RegExp literal outside scope
performance array-methods
performance string-concattenation
XmlHTTPRequest prototyping
ActiveX object instance expandos
set innerHTML on fi TABLE/TBODY/SELECT

#26 On November 6th, 2006 10:39 pm Christy replied:

I’m simply repeating what others have already posted, yet perhaps in the repitition it will get noticed…

CSS2.1 (including CSS1.0) and then DOM2 (including DOM 1). I’m kind of burned by “IE7 has tabs” as if that is just what we were all waiting for. NO, we are dying out here trying to maintain browser compliance due to a lack of fixing what is broken.

Please fix the IE browser.

#27 On November 7th, 2006 6:20 am Dave Hodder replied:

DOM Level 2 Events, XHTML 1.0 with the XHTML media type, and some more time spent on CSS 2.1, please.

If it’s straightforward, support the application/javascript and application/ecmascript media types on the ‘script’ element too. Thanks!

#28 On November 8th, 2006 7:04 am Chris Hester replied:

agustafson wrote:

“I can only imagine how daunting it must be to look at the huge lists of change requests out there and not know where to begin. Part of our task is to figure out which bits are really key.”

Isn’t that Microsoft’s job?

Peter wrote:

“Yes, they should obviously give up, right after releasing the first major version in what, five years. They’re obviously not getting anywhere.

Obviously.

I mean, obviously.

(Blood pressure… rising…)”

There’s no point continuing if the program is simply too buggy. Can they ever catch up with Firefox? It’s like trying to build new floors on a house with shaky foundations. They did it once with IE7, but really, why carry on adding new features when so much remains broken? The whole house must be rebuilt (as other browser vendors have done.) I believe IE7/Vista is a rewrite.

Thomas Tallyce wrote:

“Piling hack after hack onto a 5-year old engine simply will cease to work.”

Isn’t Trident more like 10 years old? Peter (above) said there was a five-year gap between versions! In today’s internet, we really shouldn’t be using a product designed to compete with Netscape 4.

#29 On November 8th, 2006 5:08 pm Cameron Junge replied:

How about something old: fieldsets & legends.

I know that one issue was discussed in a MSDN blog about empty legends, but I’d like to see background colors implemented correctly. How hard can it be to get the size of the background colour to STAY INSIDE THE BORDER??!?!?

The border is correctly position, but the background color overflows to include the legend, which is clearly wrong from a design POV.

This might also fix some javascript positioning errors that are introduced when trying to hack around this mess up with CSS.

Cameron.

#30 On November 9th, 2006 6:31 pm Albin Larsson replied:

My advice for the MS-dudes:

People who writes doctypes generally don’t write their code for IE only.
So write a new render engine (or even better use the Gecko engine, it’s free) but keep and use the old one for Quirks mode (for backwards companility issues). IE8 with dual engines. That way web developers wouldn’t have to hate you.

#31 On November 14th, 2006 9:16 am Charcoal Design replied:

“So write a new render engine (or even better use the Gecko engine, it’s free) but keep and use the old one for Quirks mode (for backwards companility issues). IE8 with dual engines. That way web developers wouldn’t have to hate you.”

This is largely the conclusion of an article I wrote about IE7′s failings:

http://www.charcoaldesign.co.uk/weblog/6

It’s madness to devote countless man hours updating a broken layout engine with bugs that have existed for 10 years to try to make it behave the same way as another renderer, when you could just use that renderer instead.

I can only assume that Microsoft’s irrational fear of open source software has a role to play here.

#32 On November 15th, 2006 4:34 pm Nicolas replied:

- Full XHTML 1.0 support, starting with application/xhtml+xml
- Full CSS 2.1 support.
- Full PNG support.

The rest can wait (*).

The only other priority is real competition in the browser market, which cannot be achieved through lobbying MS, but through lobbying PC MAKERS (1) to provide multiple browsers in their preinstalled OEM Windows and (2) to set up the default Internet application to a browser that is more standards-compliant than IE. Maybe, then, MS would be more compelled to try to catch up.

=> The WASP should start an OEM TF.

(*) even though SVG support would be nice.

#33 On November 16th, 2006 4:58 am CSS3 . info - » IE.next to support CSS3 colours? - Weblog replied:

[...] The Internet Explorer team are asking (again) for input on most-wanted improvements for the next version of IE. Most of the work seems as if it will be focused on Javascript and the DOM, but as you can see on this wishlist there are also requests for CSS updates – including support for RGBA, HSL and HSLA from the CSS3 colour module. [...]

#34 On November 16th, 2006 7:58 pm Julian Bennett Holmes replied:

Fix the CSS bugs?

#35 On November 17th, 2006 1:39 pm Daniel replied:

The Internet Explorer is still missing the CSS1 “inherit” keyword.

In think the development team has to do some fundamental changes in the rendering engine. The concept itself is probably not that bad, but it’s buggy.
As long as some simple bugs are not fixed it’ll be dufficult to improove the system at all.

#36 On November 18th, 2006 4:39 pm Alexandre Roberto Pereira replied:

The Internet Explorer needs to implement CSS correctly. Full CSS 2.1 is enough to avoid webdesigners suicides.

And Internet Explorer needs to be upgraded for all systems, without WGA verification if possible. I live in Brazil and here the piracy is strong.

#37 On November 20th, 2006 11:50 pm Chris Martin replied:

They might be listening, but that’s it.
Why waste our time outlining what’s readily available on the net, when they’ll most likely spend more time on some blog answering posts about how they’re listening, then actually fixing these issues. (see IE 7)

Before anyone wastes their time thinking about this, read this perfectly accurate outline of the progress they’ve made listening for the past two years.
http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/CWilsonMSIE7AndCSSCompliance.html

#38 On November 23rd, 2006 1:06 pm MJHolmes replied:

I’ll add to the outcry, FWIW.

Just support CSS up through the current CSS3 draft. That should not be too much to ask one of the richest software companies in the world to do this when open source effors can do it.

Especially opacity and RGBA colors.

I don’t even care what wonky proprietary stuff you support, but support the standards.

#39 On November 25th, 2006 10:36 am Web Dev Bros » Blog Archive » IE7 known bugs important for web developers replied:

[...] IE.next < back home [...]

#40 On November 25th, 2006 1:17 pm Charles S. Cavanaugh replied:

Well, it seems that all of this has totally left the in-expert in the lurch. I utilize E-Mail and Instant Messenger for a variety of reasons.

It is not my intent to become a computer expert. Frankly I have no idea what most of what has been posted here means, nor do I want to know.

What I want is, to be able to interact with E-Sites on The Net, without hassels. That is becoming more and more difficult. Although my computer is older and is programmed with Windows 98-II, I am finding that this stuff is too complex for my poor old ‘Puter to understand and then translate it for my edification.

Why should I have to discard a perfectly good Computer and it’s programming, with the incredible difficulties of trying to keep my already installed programs and files and folders, because someone thinks they have a better way? If all of this Fol-dee-rol IS necessary then it must be made available, FOR FREE, with SIMPLE availability to all of us who are barely computer literate. My illiteracy is not hurting anyone else. So let me be illiterate. Just supply the necessary upgrades in a simplistic manner and for free. I only want them if they are necessary for my accessing sites on the Net. Apparently they are, or I wouldn’t be here in the first place.

Thank you,

Charles S. Cavanaugh

#41 On November 28th, 2006 6:34 pm Leslie replied:

When styling a select element in a form, if width is applied, the options are cut off if the text string in an option is longer than the select box is wide. For example, if I say:

select {width:50px;}

and one of the options in the select is “This is a very very very very very very long string”, all you’ll see when you open the select list is “This is a very”. Firefox handles this by expanding the options window.

#42 On November 30th, 2006 4:22 pm Scott replied:

I’ve noticed that the accept attribute on the input element isn’t supported yet (not just in IE), but it’s part of the W3C specification. Here’s a quick example of it’s usage:

As I understand it, the intended functionality is that the file dialog which appears when you click the Browse button is supposed to filter on the mime type so that files with other mime types are not shown.

Since it’s not on the list yet, I figured I should speak up. I also sent an e-mail to the IE team, but if the keepers of the wishlist think it’s worthy there’s a better chance it will get implemented sooner than later.

#43 On November 30th, 2006 4:23 pm Scott replied:

Woops, HTML gets stripped out. I’ll convert the angle brackets to square brackets:

[input type="file" name="myFile" accept="image/jpeg" /]

#44 On December 1st, 2006 5:45 pm JonnyT replied:

I’d like to see the ability to adjust the browser to render settings of previous versions of IE for testing and compatibility.

#45 On December 12th, 2006 7:22 am Carta a los reyes - Scriptia replied:

[...] Aaron Gustafson, PPK y otros ninjas han anunciado la creación de una «carta a los reyes» que describe, en lo que hace a javascript, las mejoras que deberían incluirse en la próxima versión de Internet Explorer (corrección de bugs y características aún no implementadas): IE.next Wishlist. [...]

#46 On December 13th, 2006 3:04 pm Ted replied:

I don’t think MS is capable of fixing their rendering engine. They are on version 7 of their product and it is still horribly broken. It’s time for them to admit defeat and ditch the core rendering engine and replace it either with WebKit, Gecko, or buy/license Operas. Any move that doesn’t either completely fix the outstanding rendering bugs or replaces their engine entirely proves to me that Microsoft doesn’t really give a crap about standards, and is just paying lip service everytime that ask for feedback.

#47 On December 14th, 2006 5:02 am anonymous replied:

IE really needs XHTML support. without it it’s pretty much useless to me because i would have to use a different browser for about 40% of my web browsing…
support for an XSLTProcessor object would be nice, too… that would mean i could eliminate the last tiny bit of browser-specific javascript in a web app i’ve been developing…

#48 On December 14th, 2006 5:10 am » IE.next Wishlist — cne _LOG Archiv replied:

[...] So wie es schon Roger Johansson, die Microsoft Task Force, die DOM Scripting Task Force und die JS Ninjas mit ihrer IE.next Wishlist getan haben, sollten auch wir uns Gedanken über einen wirklich guten IE machen. [...]

#49 On December 19th, 2006 1:02 pm Isaac Yuen replied:

Make Ctrl+L directly highlights the url textbox instead of a stupid “Open” pop-up. Why provide both ways (Ctrl+O and Ctrl+L) to open a stupid pop-up while you can use one way more effectively?

#50 On December 19th, 2006 1:59 pm Jon replied:

@ Charles S. Cavanaugh (Comment #40)

In regards to your comment about why all the things in this comment thread don’t apply to you the non-computer savvy individual; the fact is that all of these comments effect you. No, you may not care about CSS Standards compliance, or JavaScript bugs in IE, but it still effects you. You say all you want is for the websites you visit to better serve you, but in order to do that they need these technologies in place and working seamlessly. Personally, I’d rather spend my time making the end user happy by adding functionality than worrying about making sure something works at all or looks good in Internet Explorer. So much time is wasted when you have to hack things together to make it work in IE.

Alright, now to get back on topic. As it has been said a million times before, CSS compliance would be my biggest complaint. Maybe add the ability to use min/max width?

#51 On December 20th, 2006 10:06 pm Nick Stimpson replied:

So, I’ve just been fixing a web site layout that’s OK in IE6, correct in Firefox and Opera 9, but out of kilter in IE7. Aaaargh!

Please don’t let M$ use Gecko, the thought of them putting their sticky mitts all over it are too much to bear. I need a rendering engine that works properly even if other users choose to stick with rubbish.

Completing CSS2 implementation is what we need first. It’s been a recommendation for eight years for goodness sake.

#52 On December 21st, 2006 10:24 am nick holmes replied:

Tooltips. So we all know that tooltips are supposed to be what is in the title on the link right? And we all know that IE shows the alt tag as the tooltip. While I don’t have a problem that IE shows the alt, I do have a problem that it is inconsistent with other browsers in that it gives the alt as a tooltip precedence over the title (the link destination) as a tooltip, i.e. if you have both the alt is show and the title isn’t.

Is there an argument not to change this to be consistent from Microsoft? I’m not saying dont show an alt tooltip, only that if there is a title attribute on the link that this takes precedence.

thoughts?

#53 On December 24th, 2006 8:59 am Daniel replied:

One more thing that I wanted to mention:
The IE developers stated some time ago that implementing standards is difficult because CSS and HTML got some parts that define certain things differently.

Thats actually the reason why display: table-*; can’t be implemented, or so they say.

In my opinion it is far more important to make todays browsers CSS-browsers (with knowledge of HTML), because CSS is used for XML as well (which can also use display: table-*;).
So in my opinion when HTML and CSS are contrary, implementing it the CSS way is the right way. This should be the ultimate goal of IE.Next.

On the other hand, it could make sense to implement the CSS-Tables into standard-compilant Mode (as this is definitely the Mode good desingers are using) and keep the HTML-Tables for QuirksMode (which is actually the way IE developers did it in the IE6->IE7 jump).

And again: the CSS1 inherit keyword is very important. It’d make my life way easier even if they just implemented it for color and background-color.

#54 On December 26th, 2006 12:07 pm Alex replied:

<OPTION> tags with the ‘disabled’ attribute should be disabled. In IE7, you can freely select it.

#55 On December 30th, 2006 12:33 pm Alimadzi - by Patrick Lee » Firefox unofficially passes the Acid2 test replied:

[...] As for Microsoft, they’re apparently working on something called IE.next. I wouldn’t hold my breath for it to pass Acid2, but they are soliciting recommendations for changes that should be made. That’s certainly a step in the right direction. [...]

#56 On January 3rd, 2007 11:32 pm Sandy B replied:

CSS height for deus’s sake. HEIGHT, i mean damn people, cmon. Possibly some sort of button that allows you to download firefox directly , deleting IE in the process. Personally I would like to see the ability to add background pictures to a select box. That would be cool. Basically just get it together, you’ve been doing this for years and you should be on top of the game….not way behind it. Call me a noob if you wish but im tearing my hair out here trying to find work arounds for what I consider just laziness. But im sure your doing your best guys. We all are.

#57 On January 6th, 2007 8:39 am Alan & Julie replied:

comment We with our research has found IE to be the most popular of browsers and rescnt advances, although at first seem more sensitively adopted we believe still the first choice by the majority

#58 On January 8th, 2007 11:52 am Nathan Chase replied:

I second Leslie’s request of improving IE’s rendering of the options within a fixed-width styled SELECT box.

quote: “When styling a select element in a form, if width is applied, the options are cut off if the text string in an option is longer than the select box is wide. For example, if I say:
select {width:50px;}
and one of the options in the select is “This is a very very very very very very long string”, all you’ll see when you open the select list is “This is a very”. Firefox handles this by expanding the options window.”

#59 On January 11th, 2007 7:07 pm Asad Hasan replied:

DADesigners. is compiling a list of css errors found in internet explorer 7. As of now we have more than 20 bugs with css fixed but these are needed to be addressed universally in the whole ie7 release. I will post the link to the project page as soon as it is done. We can all work to make ie7 a wonderful browser

#60 On January 13th, 2007 6:03 pm Rao Nagesware replied:

You guys are doing an awesome job, i m not one of those who just simply criticize microsoft since they are good, instead i believe in an effort to make it better. I would love to see it as stable as mozzile, and yet more speed efficient

#61 On January 15th, 2007 12:08 pm That Voodoo You Do · DOM Support in IE7 replied:

[...] Hurray! The IE team are now shifting their focus to fixing up the ol’ DOM support. Excellent news! If you have anything to contribute, or can provide and excellent test case, head on over to the Web Standards Project post on the topic, or have a look at their growing wishlist (invitation only). Great work, guys! [...]

#62 On January 16th, 2007 6:29 pm George replied:

OK, the tabs are nice, but they could be made a lot more useful. I’d like to be able to drag a tag out of one window into another. Most importantly: if there was a ‘Save state’ button which stored the IE window with tabs and their addresses so that reopening all the tabs and their URLs to continue work on something complicated was as simple as clicking on the saved state, I would begin to love working with IE.

It’s not a technical request, but website consumers make up the majority of the userbase of IE and this would help its cause immensely.

#63 On January 23rd, 2007 9:23 pm Daniel replied:

Is the Wiki linked in the article still maintained? It’s last update was over two months ago.

I discovered a bug, but I don’t know where to submit a testcase so it I wondered wether there are any news about this?
Thank you.

#64 On January 24th, 2007 1:11 am Candace replied:

Possibly some sort of button that allows you to download firefox directly , deleting IE in the process.

That made me giggle ;) But yes, while I spend hours trying to make my CSS site look right in IE when it already functions perfectly in all the others.. I start to wonder just how much microsoft wants my business.

I would like to see full CSS support. But seeing how long it took for the tabbed browsing, I guess I’m still in for a wait.

#65 On February 4th, 2007 3:49 pm Talking with Microsoft about IE.next - The Web Standards Project replied:

[...] You may recall that the DOM Scripting and Microsoft task forces, in collaboration with JS Ninjas, had been compiling a list of issues, needs, and wants for IE.next over the last few months (a list many of you contributed to as well, via your feedback). The list was to focus on what we wanted to see happen in terms of JavaScript support (as IE7 didn’t get much of an update in that area), but when it came down to it, there were other areas we really felt needed some love. [...]

#66 On February 4th, 2007 3:50 pm Talking with Microsoft about IE.next ¬ Easy Reader replied:

#67 On February 5th, 2007 9:25 am Painfully Obvious » Blog Archive » Link: Talking with Microsoft about IE.next replied:

[...] Talking with Microsoft about IE.next. The things he asked for (mostly JS-related) were determined (by vote) to be the most critical things for IE to catch up on. I’m optimistic. [...]

#68 On February 5th, 2007 12:54 pm Jack Sleight replied:

I’ve run into this CSS bug a number of times, unfortunately it was never fixed in IE7. A fix would be fantastic:

http://rowanw.com/bugs/overflow_relative.htm
http://mt-olympus.com/emmett/bug_overflow_positionrelative.php

#69 On February 5th, 2007 4:39 pm John Lascurettes replied:

I would like to second Thomas Tallyce’s comment (#22) that says hasLayout needs to be refactored. A majority of Explorer’s more persistent bugs and shortcomings are directly tied to this behavior.

#70 On February 6th, 2007 10:13 am Juergen Riemer replied:

>>> native support of SVG

#71 On February 6th, 2007 7:53 pm Dave Golding replied:

It shouldn’t be all that difficult to fix CSS measures, given that Firefox is open source and all the web standards are visible to everyone. Just copy and paste the rendering engine from Firefox into IE, and we’re good. Heck, you do that with everything else (e.g. Word => WordPerfect, Excel => Lotus, Windows => Mac, Internet Explorer => Netscape Navigator, Zune => iPod, X-Box => Playstation).

#72 On February 7th, 2007 9:09 am sanyodenki replied:

1. Make it CSS 2.1 standards compliant.
DON’T ADD ANYTHING until you fix all the broken and unfinished stuff FIRST. IE is a half-baked, defective product. It doesn’t need “new stuff”. It needs to work right first.

#73 On March 11th, 2007 10:23 pm Alex replied:

What I’d like to see is visually see a td with a colspan or rowspan attribute (kind of like how Firefox displays td tags with “rowspans” and “colspans”). Otherwise, a table might look ugly and seem like there was nothing filling its spot.

#74 On February 19th, 2008 7:19 pm Allen Garner replied:

Security and Support for css Standards is definately at the top of the list for example the select element in Internet Explorer 7 still does not support many css implementaions. Also, we need to start thinking about the web more in terms of programming. The internet has been and is increasingly more so being used as an alternative solution for applications in intranets and in webpages on the internet. We need to add native elements to html such as datagrid combo boxes , tab structures, color pickers, date/time pickers, hierarchical tree structures and many other useful elements that will aid in the improvement and usability of the internet.

Working together for standards The Web Standards Project

You can improve IE.next

Your Replies

Post a Reply