The May 2006 Internet Explorer developer chat is over, and this time every one of my questions got through. You might also want to take a look at the previous chat. Here my new questions and their responses:

Dave Massy (Moderator):

Q: [21] The public betas of IE7 have introduced a few new possible CSS hacks, such as *+html{}. Are these likely to change before the final IE7 release?

A: See the team blog post at http://blogs.msdn.com/ie/archive/2006/03/23/559409.aspx where it was announced that IE7 is layout complete for CSS. Please give feedback through the mechanisms outlined at http://www.microsoft.com/windows/ie/support/default.mspx so we can continue to improve IE in future versions.

So I guess that's another firm confirmation that what we see in the current beta is what we'll get in the final release.

Dave Massy (Moderator):

Q: [36] In addition to Quirks and Standards modes, Firefox, Opera, Safari, and Konqueror have also implemented an Almost Standards mode for certain doctypes to mimic some IE behavior. As this behavior is fixed in IE, will IE also offer an Almost Standards mode?

A: I'm not sure what a mode of operation to mimic what IE already does would accomplish :)

Dave Massy (Moderator):

Q: [46] Dave Massy: The Almost Standards mode is to mimic *bugs* that IE has in certain situations. My question is will an Almost Standards mode be introduced in IE as those bugs are fixed?

A: there is no plans to increas the number of modes between "quirks" and "Strict". I think a third mode would only lead to even more confusion than we have today and not really benefit developers.

Dave Massy (Moderator):

Q: [54] Some things like mixtures of integer and percentage values in rgb() are accepted in IE, but should be illegal according to the current CSS standards. Will IE ever have picky issues like this fixed, or are they considered not worthwhile fixes?

A: Hi Nanobot, Limiting these formats can easily break existing web content so we have no immediate plans to change this at this time.

This question sparked an endless argument in the public discussion channel with CDarklock (another visitor in the chat), who apparently believes that strict error handling, an essential feature of CSS that helps future growth, is stupid and Internet Explorer shouldn't follow it. It should be noted that Internet Explorer, as well as all other major browsers, does generally follow these error handling rules, but happens to support a nonstandard implementation of the rgb() value. So the rgb() issue is a minority inconsistency in Internet Explorer and the rationale Microsoft has generally accepted for the rest of its CSS implementation (as well as the one in the standard) is contrary to CDarklock's rationale.

Dave Massy (Moderator):

Q: [70] Can we expect support for alternate stylesheets in the version of IE following IE7? It's a conformance requirement for CSS 2.1 and already supported by Firefox, Opera, and Konqueror.

A: This is somethign that may have to wait for the next version of IE when we expect to continue to improve our CSS support.

Dave Massy (Moderator):

Q: [86] From my testing, it appears that the only changes to DOM support in IE7 were adding minWidth, maxWidth, and maxHeight to Element.style (minHeight was already there in IE6) and XMLHttpRequest changes. Where there other changes I overlooked?

A: For IE7 the main platform focus was addressing key CSS pain points that developers were feeling. This included fixing nearly every issue on www.positioniseverything.net . Improvements to DOM and JScript are definitely on our list for future versions. We'd appreciate any feedback on what you think we shoudl address first and why so we can prioritise that work.

Dave Massy (Moderator):

Q: [98] Dave Massy: Well I think top priorities should be Dom Level 3 Core and the standard event model. (I don't need a response to this)

A: Standard event mode being DOM level 2 or DOM level 3 events. Unfortunately there are compatibility issues between the two that probably need to be investigated.

Update 2006-05-22: I probably should have mentioned prototyping of DOM objects and the various bugs with getElementById and such, but they slipped my mind at the time.

EricLaw [MSFT] (Expert):

Q: [100] MS states that its product development is based on customer demands. Is it generally regarded that web devs demand standards compliance as much as possible, or do you still believe there is justification in branching away from or ahead of the standards?

A: IE has a lot of "customers" including web developers and normal users who like to surf the web. Some web developers demand standards compliance, and some demand that their pages continue to work exactly the same from version to version of IE. It's always a balance.

Uche (Expert):

Q: [110] If you were stranded on a desert island with only one kind of junk food to survive on for the rest of your life, what would it be?

A: donuts!!! :)

Considering that the rest of the IE7 development will be aimed at basic performance improvements, I doubt I'll have much to ask during the next chat, if anything. They still seem pretty vague about development for the version after IE7, but it seems it will continue to focus on CSS but also touch on DOM support.