Internet Explorer and Accessibility


Since this is my first post on the IE Blog, I wanted to introduce myself.  My name is Kelly Ford and I’m part of the test team responsible for testing the user experience in IE.  I also head up our accessibility testing efforts.

Today I wanted to talk about three aspects of accessibility as they relate to IE and Windows in general.  First is access to the Windows OS for individuals with disabilities, second are a couple of hints for users of screen readers using IE in XPSP2 and finally is a request for feedback to help guide our development in IE7 and beyond.

Accessibility Starting Points

Some blog readers may already be familiar with the wide range of technologies and issues around software accessibility.  For those that are not I wanted to provide a few starting points to learn more.  Whether you are an individual with a disability, know someone who has a disability or accessibility isn’t something you have much experience with, there are many resources available.  The key point is that the computer and web can open doors to people of many different abilities.  Through accessibility features in Windows or a category of software known as assistive technology, people with mild to severe impairments can find software and hardware solutions to help in getting access to the computer and web.  Some resources to learn more about this topic are:

Microsoft Accessibility Home – http://www.microsoft.com/enable
Assistive Technology Industry Association – http://www.atia.org
W3C Web Accessibility Initiative – http://www.w3.org/wai
MSDN Accessibility Resources – http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnanchor/html/accessibility.asp

Hints for Users of Screen Readers with IE in Windows XPSP2

For those unfamiliar with screen reading technology, a screen reader is an application that is used by individuals who are blind to access the computer.  Text appearing on the screen is verbalized and optionally displayed in Braille with the appropriate hardware.  Inside of IE, screen readers use a combination of Microsoft Active Accessibility (MSAA) and the document object model to present a screen reader friendly view of web pages.  Briefly, items such as tables are presented in a more linear fashion, structural information such as which text is a link, a heading, alternative text for a graphic  or part of a list is inserted, and a general overview about web pages such as the number of links, form elements and alike is provided.  There is much more that today’s screen readers do inside IE and if blog readers want more details around this technology let me know.

In monitoring feedback around screen readers and IE with XPSP2, I’ve noticed a couple of recurring issues for which I wanted to offer screen reading users hints.

Accessing the Information Bar

The switch from dialogs to the Information Bar for things such as automatic file downloads, ActiveX installs and related notifications was one of the larger user experience changes introduced in IE for Windows XPSP2.  Although we offer a first time alert dialog notifying users of this change and detail how to use the Information Bar in help, I’ve seen many screen reading users ask about the Information Bar on various discussion forums.  The most important question deals with how to get access to the bar with the keyboard.  The hotkey for setting focus to the Information Bar is alt+n.  You can then press space to receive a menu of choices related to the notification being displayed.  The Escape key will close the Information Bar.

If you prefer notifications about blocked activities through dialogs, you can revert to this behavior by turning off Information Bar notifications for certain activities.  The help under Internet Explorer’s Information Bar has full details on the procedures for making these changes.  That said we recommend leaving the Information Bar on and using alt+n to access the Information Bar with the keyboard when it appears.

Pop-Up Blocking and Screen Readers

IE’s pop-up blocking technology attempts to allow a class of pop-ups that we consider “user initiated”.  Briefly this means that if the pop-up is the result of pressing Enter or clicking on a link, IE will generally allow the pop-up.

As mentioned earlier, inside IE most screen readers present a modified view of web pages by default.  One result of this is that when in this modified view, even though you may be pressing Enter to activate a link, the link itself is being activated through one of several programmatic methods.  In some instances you may find that you press Enter on a link and the result is still a blocked pop-up.  In such cases you can use a feature common to most screen readers and typically known as Pass Key Through or some variation on this name.  Using your screen reader’s hot key for this feature and then pressing Enter on the link you want to activate will send the standard Enter key to IE.  In most cases if the pop-up should be allowed by IE’s pop-up blocker after this, the new window will open.

Accessibility in IE7 and Beyond

Accessibility is something Microsoft, the IE team and I take very seriously.  We want to continue building software that can be accessed by everyone as well as ensure that the wide range of assistive technology products for Windows and IE work well with what we develop.  We’d like your feedback on what’s working well for you today, where you’d like us to improve and in general hear your thoughts around accessibility and Internet Explorer.  We look forward to your thoughts on the topic of accessibility and Internet Explorer here on IE’s blog.

Kelly

Comments (48)

  1. Anonymous says:

    Your web documentation is hardly full of best accessibility coding pratices. And how many Microsoft pages pass the W3 and Bobby accessibility tests?

    Please put your own house in order.

  2. Anonymous says:

    Perhaps you can answer a question that always puzzled me.

    Why did you guys pick Alt to be the access key when it so obviously conflicts with existing keyboard accelerators? Why couldn’t it have been Ctrl + Alt or some other key combination?

    Other browsers have made similar choices, and these conflicts effectively rule out use of access keys where they would otherwise be useful and assistive to disabled people.

  3. Anonymous says:

    for speech accessibility, i’m wondering if you are considering the speech add-in for IE to view multimodal web applications? for users, i’m considering visually impaired, tablet pc, and media center.

  4. Anonymous says:

    Too bad that IE-based screen readers don’t support aural stylesheets.

  5. Anonymous says:

    My concerns are less with Internet Explorer (although more compatibility with CSS would be nice) but with the rubbish the .Net server controls generate. What we need are DDA (Disability Discrimination Act) – compliant controls that generate accessible HTML. We’ve had a nightmare of a time suppressing the invalid attributes .Net throws out, adding attributes to generate labels with for attributes for controls, removing the garbage generated on the FORM tag and validation summary controls etc. Until Microsoft get their development environment more "accessibility friendly" it’s hard to see how more accessible web sites are going to be generated.

  6. Anonymous says:

    I think the best options for IE Accessibility are better support for CSS Media Types, especially Media Queries. While the blind population and screen readers are important targets for accessibility-based design, another major problem is low-vision users, as I learned when teaching my grandfather to use his first computer in 40 years, and watched him squint at most web pages.

    Full keyboard support is also important, because his hands often had trouble aiming the mouse on small form widgets: _all_ form elements should be able to accessed by a keyboard. By default. There are currently problems with seeing select list group labels using the optgroup element, and selecting multiple, non-consecutive items in a <select multiple="multiple"> is impossible.

    The Web Hypertext Application Working Group has done some great work on web forms, and implementing some of their standards first might give you folks a leg up on other web browsers.

    Also, if I made an alternate stylesheet for low-vision readers, it would be great if IE7 had a built-in widget that let users switch to it.

  7. Anonymous says:

    Find-as-you-type like in Firefox/Mozilla is a MUST-have feature for IE7. It’s just incredible for people who can use some kind of keyboard but not a mouse, and for people who just prefer the keyboard.

  8. Anonymous says:

    I think supporting standards is probably your first bet — the akward hacks for me to make em’s work for font sizes is one that comes to mind. Text Zooming (though in IE for Mac) is limited in PC IE. Here’s an example:

    http://diveintomark.org/archives/2002/07/15/day_26_using_relative_font_sizes

    Again, as I boast much on this site, having to pitch us standard issue PR marketing phrases "Accessibility is something Microsoft, the IE team and I take very seriously." may cover your legal requirements, but the language lacks tone and sincerity. Please, talk with us candidly, I read these posts and all I see are sales pitches, not developers reaching out into the community.

  9. Anonymous says:

    There is clearly some work that needs to be done on new CSS features for accessibility.

    According to what I’ve heard, fully configurable keyboard shortcuts and something like find-as-you type would also be appreciated by many people.

    If the script debugger used the information bar instead of dialogs, I think it’d be an improvement.

    Another change I’d appreciate is to integrate the old "Add to trusted zone" web accessory to the default interface.

  10. Anonymous says:

    My company makes MathPlayer, an ActiveX control for displaying math in IE (free download from http://www.dessci.com). It can speak the math and integrates with screen readers and LD tools via MSAA and custom interfaces.

    Due to security concerns, SP2 makes it harder to use MathPlayer. Eventually, we found a check box in the "advanced" tab of "internet settings" to prevent the warning messages for each page containing math ("Allow active content to run on files on My Computer"). We posted a note on this on our website, but I doubt many people will find the note or figure out the problem on their own — SP2 puts up an obstacle for those who wish to have access to math in web pages.

    We are very concerned about what IE7 will do to ActiveX controls in the name of security. Please remember that there are legitimate ActiveX controls, and throwing obstacles in their way makes it hard for users to take advantages of their functionality, especially if the users face accessibility issues. I understand that security concerns are very important and it is sometimes necessary to reduce functionality in the name of security. Perhaps you can set up a program similar to what Adobe does for Acrobat Reader plug-ins whereby developers certify their control meets certain criteria. Adobe vets the organization and issues a certificate that allows the plug-in to run in Reader. Doing something similar for IE7 would be a win-win for security and functionality.

  11. Anonymous says:

    There’s a lot you can do to make accessibility better for everyone including non-disabled people too, like having some sort of support for rel="" on hyperlinks and link tags. Catching up to mozilla 1.0’s featureset would be a great thing.

    And get rid of the bug that displays alt="" attributes in tooltips. All it does is encourage people to abuse them.

  12. Anonymous says:

    Neil Soiffer, it sounds like users are using your ActiveX locally, with the HTML coming off the local drive. That runs afoul of the SP2 Local Machine Lockdown. If you control the pages, you can insert a "mark of the web" to move the page into a different zone.

    http://www.microsoft.com/technet/prodtechnol/winxppro/maintain/sp2brows.mspx

    "Perhaps you can set up a program similar to what Adobe does for Acrobat Reader plug-ins whereby developers certify their control meets certain criteria."

    To some extent, that is what Authenticode was supposed to do–create some accountability. However, Verisign is the fox in charge of that henhouse; they have issued certificates to real scumballs and don’t have any way to allow consumers to petition for cert revocation.

    http://www.benedelman.org/news/020305-1.html

  13. Anonymous says:

    Why not just drop the IE rendering engine and use Geko.. make everyone happy!

  14. Anonymous says:

    I have to agree with Ian Smith. I know for a fact that VB.Net generates some pretty ‘interesting’ markup when using design mode.

    I’ve seen my clean, valid markup get turned into a nested table mess with some unintuitive ID attribute values as well. (ie: table1, table2, etc.)

    Don’t even get me started on the Tab control.

  15. Anonymous says:

    Before I forget too, there’s a plethora of information regarding web accessibility at:

    http://www.accessify.com

    That may be useful to add to your list.

  16. Anonymous says:

    Fix numerous rendering bugs that force developers to avoid good techniques and sometimes use nasty hacks instead.

    Catch up on CSS2 support. Display: table-cell makes transistion from tagsoup to semantic markup easier.

    Support basic HTML like <link rel="next", prev, up, home, etc..

    Support alternative stylesheets.

    Fix text-zooming issues.

    If you want more – see Opera. Besides decent support for typical technologies it has XML+V support, media queries, spatial navigation, nice set of user stylesheets, easily configurable interface and keyboard shortcuts, accesskey support for all keys, etc.

    And for completness: get PNG working :)

  17. Anonymous says:

    >>Why not just drop the IE rendering engine and use Geko.. make everyone happy!

    Dont use Gecko, it make me not happy.

  18. Anonymous says:

    Dear IE team.

    It’s great that you are trying to keep people up to date with what you are doing, but can i suggest you go to Opera’s forum and just look at how their staff interact with their customers, you could really learn something about customer service.

    If you are not going to take the time to answer people, your customers is who they are, then turn the comments section off. And just post your progress.

    Neil, I read a lot people saying SP2 has caused problems with the software their company makes. But I would rather have an operating system that is more secure, and does cause such problems, than have an operating system that isn’t as secure so companies own programs work.

  19. Anonymous says:

    I have to agree that Microsoft and MSN pages do not validate with Bobby as accessible content. There is a lot that MSN and Microsoft must do in order for their pages to be accessible. For example, where is the Skip to main content link? It should be at the top of all pages.

    The most annoying thing however is that currently MSN have started a habbit of using images in the place of form buttons. For example, on music.msn.com there is a search box and a Go image, not a Go button. Well as you know pressing an image is not exactly accessible. In fact it is completely inaccessible.

  20. Anonymous says:

    I have to disagree with almost everyone who advocates the addition of a find as you type feature. I would advise everyone who pushes for the addition of such a feature to use a screen-reader first. Let me explain:

    Screen-readers, as has already been said, present an aultered view of web pages. In order to do this, screen-readers are forced to take control of the keyboard extensively. For example, the cursor keys are no longer used to scroll the page but are instead used as navigation keys for reading the page content line by line. Every imaginable key has a use. For example the letter h might be use to jump to the next header. Even obscure key-combinations like alt+shift+f11 have a use.

    Adding Find as you type on top of this would certainly create many compatibility issues for our screen-readers. You should also know that screen-readers are not updated immediately when new software, such as IE7, come out. It takes time, maybe more than a year. And updates are not always free of charge. On the contrary, updates usually cost a lot.

    Given the above, I would suggest as a new accessibility feature of IE7 to add a new text window which will show the contents of a web page in a Notepad-like format. Check out Webformator for ideas: http://www.webformator.com/ There are many times when I am forced to Select all and Copy the page from IE into Notepad in order to read it. That is why a text-only view of web pages in a special window would be very useful.

    Concerning the Find as you type feature, add it where it belongs, ie. as an extention to the Find dialog box or as a new toolbar that will hilight text as you type in a box on it. Do not add it as an IE main window feature please.

    Having mentioned toolbars I also wish to express my desire to make them accessible. In Office applications accessing toolbars with alt and then ctrl+tab is simple. Why not in Internet Explorer? Please make toolbars accessible.

    Note: My screen-reader of choice is Supernova http://www.dolphinusa.com. It combines magnification, speech and Braille output. If IE7 is not accessible with it, it is guaranteed that I will not be updating to it soon. Plus everybody else who is blind or partially sighted and who uses the same or similar screen-reading technologies.

  21. Anonymous says:

    Neil Soiffer: I think your company should cooperate more closely with MS and become more involved about security issues. There is work to be done on MathML support in IE, especially if XHTML support is considered. I have to point out that I’ve found *no* information on your website on MathPlayer security issues, nor a system for security update. For a piece of software that is as complex, as intricately liked to the browser, and as exposed, this looks **very** bad.

    Since concurrent browsers have opted to integrate MathML support in the mainstream program, what about co-developing a built-in MathML support for some future version of IE, so that security fixes could be distributed via Windows Update? You would gain MS’ collaboration on security issues, and MS would gain your help and experience on MathML rendering.

  22. Anonymous says:

    Nektar – you can turn Find as You Type off and use a normal dialog. I don’t see the problem.

  23. Anonymous says:

    > I have to disagree with almost everyone who advocates the addition of a find as you type feature. I would advise everyone who pushes for the addition of such a feature to use a screen-reader first.

    Accessibility isn’t just about catering to visually impaired visitors. Find-as-you-type is useful to people who have difficulty using a mouse.

  24. Anonymous says:

    Please consider developing more powerful and efficient keyboard shortcuts. I use opera for its absolutely brilliant use of shortcuts. I can browse the web for hours without ever touching a mouse–while being much faster than a mouse user. This is certainly not possible with any other browser. I don’t think firefox has done anything here. The 3 big things Opera has done well are website link navigation, proper focusing, and building a complete list of shortcuts for tasks–many of which are single keys. Also, make sure there is a very good way to navigate tabs if you offer them. I like 1 for previous, 2 for next.

    Also, the fact that Opera instantly loads pages when going back is huge–not checking with the server should at least be a user option. I have never seen a downside. It probably saves me hours a week.

  25. Anonymous says:

    I think Internet Explorer is perfect as it is. And nobody really complains, except a bunch of computer geeks without life. Don’t break IE!

  26. Anonymous says:

    I have one serious nag with IE currently. I like to keep the popup-blocker at high setting. Infact, I would like to have 0 popups. However when I press CTRL to allow a popup – it opens it in new window! If I have the High setting, I obviously want 0 popups, so that should be respected and open the next page into the same window.

    So PLEASE, consider that some may want to browse web in the "old" way, opening the content in the same window. If you go to MSN Spaces etc, you have hundreds of windows open eventually if you do not keep closing them as the f*c*i*g MSN Spaces loves to open new window on every click.

    FIX: Open in current window even when script wants to open a new window if I want so!

  27. Anonymous says:

    > I don’t think firefox has done anything here.

    You can switch between tabs using Ctrl + Page Up and Ctrl + Page Down, or you can go directly to the first, second, third, etc tabs with Alt + 1, Alt + 2, Alt + 3, etc.

    > Also, the fact that Opera instantly loads pages when going back is huge–not checking with the server should at least be a user option.

    Not checking with the server should be the default behaviour. RFC 2616 is very clear on the matter:

    "In particular history mechanisms SHOULD NOT try to show a semantically transparent view of the current state of a resource. Rather, a history mechanism is meant to show exactly what the user saw at the time when the resource was retrieved."

  28. Anonymous says:

    Your article did not occur in the RSS feed.

    Regards,

    Dick Boogaers

  29. Anonymous says:

    Please make new keyboard shortcuts compatible with Firefox.

    For example, ctrl+enter, shift+enter, ctrl+shift+enter should have the same behaviors as in FF.

  30. Anonymous says:

    So, to ask a random question…

    What makes IE7 on Win2K so hard anyway?

  31. Anonymous says:

    The best thing IE can do to enhance the accessibility of web sites is to enable web developers to write accessible web sites. The CSS spec spells out a variety of accessibilty enhancing features. I’d consider implementing all the CSS accessibility features a baseline before other possible accessibility features are considered.

  32. Anonymous says:

    There are some inconsistency with a HTML controls behavior regarding the accessibility events. The VALUE_CHANGE event fired only by combos and list box because they have their own window, what about the rest?

  33. Anonymous says:

    Accessibility without webstandards eh? :)

    You guys wont reach your goal because you CAN’T!

    Support standards, get out of the stone-age

  34. Anonymous says:

    zzz: Holding down Ctrl bypasses the popup blocker, hence you get popups. The fact that the web sites are coded to use popups/new windows is their fault.

  35. Anonymous says:

    How about implementing the DOM natively so I don’t have to load a jackass ActiveX object to parse a tree? You do understand that EVERY OTHER BROWSER puts this in their standard DOM model. I don’t have to load jack doo-doo to call "document.evaluate" and run a simple XPath in Mozilla browsers. IE however, lives up to it’s vast history of getting this exactly wrong. Let’s just spew the laundry list, shall we?

    <b>Things IE 7 Needs To Remain Relevant The 21st Century</b>

    <ul>

    <li>Real CSS 1 support (font-size, I’m looking right at you)</li>

    <li>Real CSS 2 support, but only after you finish your CSS 1 implementation</li>

    <li>Native DOM/XML implementation that requires no library or object loading</li>

    <li>Real png alpha support, like you had in IE for Mac before you killed it</li>

    <li>Native SVG parsing, display, and scripting</li>

    </ul>

    Ironically, all this could be yours for free if you’d just sign on to the Mozilla project and relese some code back to the base. Then you could use a REAL engine, instead of the crumbling pile of (probably) Mosaic era crap you’ve been claiming as an integral part of the operating system since 1998. In a word, grow up and join the rest of the human race, here, in the fabulous 21st century.

  36. Anonymous says:

    Jonathan: Indeed, but given that in .NET I can inherit a class and override a method, why the * its not possible to override the behaviour of the method that opens always new window/popup?

    When they decided to allow windows being opened to context not related to the original source it allowed all the moronic tricks of pop-here-and-there. The correct behaviour would have been to show the popup window only when the underlying page which opened the popup is still visible. Some consideration should have been put into the scenario where script can drain the machine out of resources by opening ton of windows. Perhaps its time to determine _what_ the code does _before_ executing it. I do not run random .exe’s without atleast quick peek at them for suspicious strings and api calls. If exe is exe-packed, I unpack it first and if that fails I delete it. The computer should be equally smart to figure out if the code has bad intents, such as opening popups even after I have shouted loudly at it for doing such silly things. Simple configuration option would do for now though :-)

    It’s nice as dev to have the power and features, but clearly its time to make the browser _USER FRIENDLY_ – I do not see all the Firefox/Mozilla/.. users give rat about the security issues constantly reported – they care about blocking popups, ads and usability.

    People still use older crap than XP SP2, if they cared more about security than usability (in terms of being too lazy to upgrade over some under the hood improvements and stuff that they already get in 3rd party apps like tabbed browsing) theyd be flocking to XP SP2 without the automatic update. I do not see this happening what comes to some people still using Win2K even when they have P4 with HT (my argument that for HT’s benefits you need XP fall to deaf ears – their concerns are elsewhere)…

  37. Anonymous says:

    Jonathan: Besides, there is the "Reuse windows for launching shortcuts", not to mention "Enable page transitions" "Enable visual styles…" etc The IE advanced menu had a load of rather cosmetic stuff never used. There are like <3 options I’ve ever needed to change.

    The "Reuse window when clicking link" would be welcome addition to the "Reuse windows for launching shortcuts".

  38. Anonymous says:

    I’d also like to ask why TLS is disabled by default? SSL2/3 are enabled. I wonder whose server breaks down if TLS is enabled.

    It’d also rock if the "Reuse window when clicking link" had the possibility to ignoring pragma:nocache and like, so if I want to go back in the history of certain window/thought space, it would just go back without reloading the pages, I am not interested what the page currently has if I’m going back to look what it _had_!

    This theme could be taken bit further in the design:

    <tab1> <tab2> <tab3>

    When clicking on <tab> you’d get list of the links previously opened in that tab :

    <currentpage-tab1> <tab2> <tab3>

    [tab1hist-1]

    [tab1hist-2]

    [tab1hist-3]

    And by always opening popups in the same tab history instead of a new tab, this would make the flow of thoughs between the tabs easier without having to open multitude of IE windows with multitude of tabs.

    Of course there are 0.1% of sites which need to show multiple windows at once, like stock quote info etc. These are very minority of the web! You should NOT design the browser in their terms. They are supposed to create smart clients and applets for that "cool" functionality – not bring billion crappy popups.

  39. Anonymous says:

    Dave: we were unaware of the "mark of the web" (MOW) trick. Thanks. We looked into adding that into MathType/MathPage, our math editor/xml+MathML generator. Unfortunately, IE seems to restrict MOW to .html, .mht, and .xml files. Pages that work in both Mozilla and IE that contain MathML in them must be XML files, and so the .xml suffix works except… MS Office 2003 registers a mime filter for that suffix and MathPlayer never gets invoked (MathPlayer has its own mime filter to invisibly rewrite the Mozilla compatible page into something that invokes the MathPlayer behavior). In any case, .xml has been deprecated for XHTML+MathML, and the MOW doesn’t seem to work with .xht or .xhtml files. We don’t know why IE (apparently) restricted its use.

    Lionel Fourquaux: we would be happy to work with MS on any and all issues surrounding math support in IE. We have expressed our interests and problems to the IE team more than once.

  40. Anonymous says:

    First of all IE should respect current Web Standards.

  41. Anonymous says:

    Hi Kelly and everyone that is working on the new version of IE7,

    I am delighted to hear that Microsoft has started work on IE7 and is calling developers to air their opinions. As you might have noticed there is a lot of issues that has caused the Web design and development community to feel a little ‘rough’ towards MSIE. I personally think that now that you are finally giving us space to suggest what should be included in your next version of Internet Explorer, the community has to take advantage of that professionally as many have been doing here.

    Although I admire the fact that you guys are working with very little overheads, the organisation and pace of your work has to be more user-centred and both governmental bodies and assistance institutions should be included/asked to take part into the discussion as well. Please note that as a UK resident we are since october 2004 under a much more stricter flavour of the Discrimination Act than the US: every website providing services in the UK has to be in level 2 compliance with WACG. Needless to say that means more necessary work to be done for a just and fair cause.

    It would be advisable that Microsoft normalised its key access combinations to mirror the ones people already use in other browsers. This is important mostly because Opera has become the de facto browser for people with disabilities and if you wish to also encompass that audience you will win much more by keeping it the same way those users are already familiar with.

    A more robust implementation of Web Standards does benefit accessibility and usability and strengthens loyalty amongst developers and designers. A fully natively compliant XHTML, XML, CSS 1 and 2 browser is on order and I am sure this is already what you are working on so I am not going to discuss it here.

    Full PNG support is essential not only for a nicer design but, because through PNG, developers can create dynamic information that can be turned into graphics or back to text in case the user prefers not to load graphics at all. If you let me give an example of that, imagine a scenario in which a statistical chart needs to reflect price growth and decrease of chocolate bars across Europe. The products are to be constantly rearranged to show the cheaper ones first. With full alpha support one would be able to juxtapose the bars located on individual layers without blocking the view of the ones beneath in case two or three products have the same price.

    With real CSS 1 and 2 support one would be able to set a perfect and consistent offset of just a few pixels amongst them in order to make sure that these hypothetical bars can be seen distinctively. This is virtually impossible to be done in IE nowadays.

    That is my contribution for now. I do hope you guys get it right and wish the best of luck and coffee to keep the work going.

    Regards,

    Luis

  42. Anonymous says:

    It always surprises me how little attention the Opera zoom feature receives when talking about accessibility. With droves of aging boomers, the portion of people needing better visual experience online is not a small marginal or to be discounted. It makes sense that they would want an entire page enlarged to join “the rest of us” – not just some select fonts that break the design.

    Good accessibility is something all new developers leave until they have done a few sites (if ever). The scores of people learning Dreamweaver basics means there will be mountains of new sites without proper support. Each time I look at an amateur or amateur pro site and toggle the ctrl mouse wheel, invariably only the heading 4, 5, and 6 text changes or worse. The rest stay the same and the feature is rendered useless.

    Let’s consider another scenario – the fonts are relative but the designer needed an imaged based button for a certain effect. Now the largest size font setting does allow the person to read the content but the button is still small and they are forced to listen to the alt tag for a tool tip. If the same site has any type of layout complexity, the new large fonts will undoubtedly break the flow and make it a mess to look at.

    The idea behind enlarging just the fonts is far too data centric. It assumes the value of the page is in the copy but design and diagrams have a lot to do with the usefulness of the web. Being able to find that data requires being able to use all the tools the designer has employed to make the experience flow. By eliminating the text size feature altogether and replacing it with a zoom that enlarges the entire page you solve both amateur design habits and give the user an improved experience. MS Word, Excel etc have the feature – time for EI to catch up.

  43. Anonymous says:

    From my understanding, the point of this post is to discuss IE as an application (excluding the rendering engine). I’m fortunate enough not to have to rely on browser features to enable a website to be more accessible but in terms of the application, you already know the features people need like text size override, single window modes etc.

    Being a tool to access the internet, it can only come back to the W3 standards which is being fought in the following post:

    http://blogs.msdn.com/ie/archive/2005/03/09/391362.aspx

    I don’t really see the need to enhance the IE shell to cater for every possible accessibility feature. Make it a standards compliant browser and let Microsoft Partners build software using this rendering engine specific to the target market.

    In my experience, even sites that are WAI-AAA and validated are still innaccessible due to the use of technical language, specific dialect/grammar or poorly constructed navigation systems which are all web developer problems, not browser ones.

  44. Anonymous says:

    I think it is imperative to point out that he himself is blind. He probably wouldn’t tell you because he absolutely does not want your sympathy. He lives with the limitations of Microsoft software every day. But after reading some of these responses, I think it would help to direct the conversation in a more constructive manner.

    Each and every computer user, whether sighted or blind or physically limited to mouse or keyboard, should be able to use Internet Explorer to it’s fullest extent. When Internet Explorer delivers the world to a user, no obstruction of accessibility is acceptable.

  45. Anonymous says:

    thanks for ie7