Notes: Please add your name, if you contribute, to the 'Authors' section at the bottom of this file.
DO NOT COPY AND PASTE THIS FILE WHOLESALE. IT MAKES IT SLOW. Export it or save it as text.
15/06/2006
12:00 - 13:00
IE: 7 and Beyond
by
Chris Wilson
Who am I?
An unknown member of Flock of Seagulls?
* Browser guy since 1993
* NCSA Mosaic for Windows
* SPRY Mosaic
* Internet Explorer
* 2.0, 3.0, 4.0, 5.0, 5.5, 6.0, 7.0!
Internet Explorer 7
* Amazing User Experience
* Secure and Trustworthy Browsing
* Powerful Web Developer Platform
Amazing User Experience
* Streamlined UI
* Tabbed Browsing with Quick Tabs
* Extensible search in UI via OpenSearch
* OpenSearch 1.1
An open way to describe search providers
* Script API prompts user to add provider: window.external.AddSearchProvider([url])
* Or use a link: fallback
* É
* Alpha channel support in PNG images
* Native XMLHTTPRequest
* Better enables DHTML/Atlas applications
* No longer subject to ActiveX being enabled
* element now windowless
Better standards supportÉ
É means some pages break
The big point is that pages do break. This is a big deal to us. Being user and developer focussed, we have to comply with standards, but be compatible with live sites. We've decided to stick with Quirks/Standards modes. Quirks mode stays the same, many changes are only in strict mode
This is where you come in.
Pages will be broken if the standards issues aren't fixed.
Common breakages
And how you can easily fix them
*User agen strings and browser detection
* Overflow - use min-height/width or correct box size
* prolog doesn't prevent strict mode in IE7, which affects the CSS box model
Community Workarounds
aka CSS Hacks
* Rely on parser bugs or not yet implemented features
* * html (Target: IE only)
* _height:
* height/**/
* html > body
Ideally no workarounds are needed - but reality intrudes
* If you use CSS hacks - only target obsoleted UAs
* KNow what version your hack will stop working
* IE alternative: conditional comments to target browser version
We are here to help
Tools to identify and fix breaks
* IE7 Readiness toolkit
* Web developer toolbar - IE6+
* Rich toolset for exploring DHTML + CSS
* ExpressionFinder - identifies CSS hacks
* Application compat. toolkit
* Identifies features blocking app functionality
* Fiddler HTTP monitor
* And other tools
The future of Internet Explorer
* Yes, IE has a future
* We're planning the next two releases now
* Continuing to put the user in control of their security and privacy
* Building on, improving, and adding to the powerful standards-based platform
* Providing the best web user experience
IE beyond IE7 - the FAQ
* When is IE7 shipping?
* Second half 2006
* Side by side versions of IE?
* Technically hard, IE provides the OS web platform. We've heard the request for site testing, and we're actively working on it.
* Why not XHTML?
* Not until we can do it right - wellformed-ness-testing parser, integrating multiple schema, etc. We will support this in the future
* Acid2?
* In the future, yes; but not all features and bugs sorted to top of the list in IE7
What's next for the Web?
* The Mash-Up pattern will continue to grow
This is very much the idea that you're not just getting data, but you're getting components. Tons of great components from Google, Yahoo, MS, frequently free and can be mashed into some really powerful apps.
*RSS will help stimulate that growth
A great way to pass data and interoperate with a bunch of data sources that may not be rich, but have a continuous stream
* So will semantic tagging of data
* Microformats is helping to stimulate that
* LiveClipboard stuff is a good example
* More capabilities will become standardised
* XMLHTTPRequest, for example (!)
* Need application semantics, not just pages
e.g, back button going back to last change, not last page
* The interoperable platform will grow
Key takeaways - call to action
What should you do?
* Please prepare your pages to be ready for IE7
* We are dependent on each other
* Standards improvements mean that behaviour will change
* End users don't like broken pages, so we need your help!
* Make sure your IE components are prepared for opt-in & Protected Mode
* Take the Leap
* RSSify Now
* Use OpenSearch for search exposure
Community resources
* IE Dev Center
* IE Team Blog
* RSS Team Blog
* Chris's blog: http://blogs.msdn.com/cwilso/
Questions
---------
We're concerned about the way some of the security has been approachedÉ Asking the user for permission, you very quickly press 'ok' to anything you seeÉ
CW: One of the largest problems is getting informed user consent is that users don't want to be informed. Changes are targeted to try to avoid that problem. I fyou get a site with a security certificate taht doesn't match the site, we used to put a bar on the site. We now navigate you away from teh site and give you options. Still a problem, and this is the biggest, hardest thing to crack.
When IE7 launches, how is it going to be distributed? Windows Update, so they don't have a choice?
CW: When we release IE7? Will we make it a required update? The short version is no, we can't for them to install it - we get in trouble for that. We will strongly encourage them to do that, though. 'How long before I can start ignoring IE6?' It's in our best interests to get to that stage, tooÉ
How do you, as a developer, feel restricted by the long lead time in IE updates? Doesn't the fact that there are 5 years between releases annoy you?
CW: How do we feel about the fact it's taken us a long time to ship? The change to integration with the OS, it made IE development difficult. On one side, I'm frustrated 'cos I want my team's work to get out there. On the other hand, there are reasons it's taken this long (5 years aside). The bar for code is high, and so we run tons of tools on code. It's a more involved process, because the quality needs are higher. There's also lots more code. I feel badly, but I understand why it's taking this long, too.
In terms of DOM support, do you intend full support of things like element prototypes? That would be a tremendous boon to scripters, as we can only really work in Fx at the moment.
CW: What are our plans for DOM support? That's the next big thing for me, really. We focussed on CSS because those pain points were more painful. We know about the OM. The problem is that some changes to the DOM will break applications or cause them to crash. We are trying to come up with solutions, and there will be more work on the object model in the next release. The biggest problem today is our JavaScript performance. We're trying to do this today, but features will be next release.
Vista doesn't run users as Admin by defualtÉ How is that going to affect old applications?
CW: People are doing things that are just plain wrong for a standard user. The number of applications taht do this are minimal. You can still elevate to admin in Vista, and run in protected mode even when you are an adminÉ You'll get prompts to do admin stuff.
Stats can be misleadingÉ It does seem that IE's market share has been sliding. People think this is a good thing 'cos it's driving IE development. If, as a result of launching IE7, market share grows again, what assurances can you give that this won't happen?
CW: Personally feel strongly about this. Next time this happens, I don't see what I want to work on. From a professional standpoint: 1) I think IE3 was better than NN4 - we had a really good browser, IE won the war - in my opinion - because we had a better product. However, that's not necessarily true today. There's not really a need for us to own the space, and I don't htink we could. Default install isnt' as powerful as you might think, and yes, MS works better with competition. We can't simply take open source marketshare and assume it'll go away, due to open source nature. People - BillG included - have said they plan to reinvest in IE long term.
What do you think about WHATWG? Do you think MS will adopt their standards?
CW: They've done some interesting stuff. I agree with nearly all their goals. Not comfortable with the way they're working, in particular patent policy. At this point, all the things they've set out to do - evolution rather than revolution - will be adopted because that's now happening within the W3C. They're looking at how to evolve the reality. That will cover pretty much everything, and everyone in WHATWG are getting involved in the W3C.
Should we discourage people supporting older versions? And what d'you think will be fixed by IE7 release compared to beta?
CW: Second question first: what's the final release going to look like, relative to b2. We don't plan to do more features. If we regressed something from IE6 or a major issue with a new feature, we'd fix those, but not implementing new features. We have a system at microsoft where we say you have x years of supportÉ Anything older than that, we won't fix generally. If someone wants to build a site that works on those older stuff, go for it, but that's up to them.
--------------------------------
Authors:
Steve Marshall: http://nascentguruism.com/