IE10 User Agent String Update


The
IE10 user agent string
was originally introduced in the first platform preview
of IE10. In Windows 8 Release Preview we made two additions to aid server-side feature
detection.

The first addition enables detecting whether a machine has touch-capable
hardware via a new Touch token. Using this token you can present a
touch-friendly version of your site to users with touch-capable hardware (typically,
in the case where your normal site is not touch-friendly). Keep in mind that users
with touch-capable hardware may also use a mouse and keyboard. You can see where
this token fits into the user agent string in the examples below.

  • IE10 on a machine without touch-capable hardware:
    Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2;
    Trident/6.0)
  • IE10 on a machine with touch-capable hardware:
    Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2;
    Trident/6.0; Touch)

If you’re detecting touch support client-side, do not sniff the user agent string;
use
navigator.msMaxTouchPoints
instead. If the property exists and returns
a value greater than zero, the user’s PC has touch capability. For example:

var hasTouch = navigator.msMaxTouchPoints >
0;

The second addition to the IE10 user agent string is a new architecture token
for ARM devices running Windows RT. This complements the existing values for other
architectures. The examples below show how this compares to a few other configurations.

  • 32-bit IE10 on 32-bit Windows:
    Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2;
    Trident/6.0)
  • 32-bit IE10 on 64-bit Windows:
    Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2;
    WOW64; Trident/6.0)
  • 64-bit IE10 on 64-bit Windows:
    Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2;
    Win64; x64; Trident/6.0)
  • IE10 on Windows RT:
    Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2;
    ARM; Trident/6.0)

These additions apply to both desktop and Metro style IE10 since they expose the
same platform capabilities. We recommend optimizing your site’s experience around
the capabilities of the system (like Touch) through feature detection, rather than
browser or environment detection. Where content depends on plug-ins, use the

requiresActiveX prompt
to help users switch to desktop IE10 until the content
can be made plug-in free.

—Tony Ross, Program Manager, Internet Explorer

Comments (70)

  1. BartS says:

    That's great, but please explain why on earth the 64-bit Windows indicator changes based on the the version of IE. In other words, why is the user-agent string indicating 64-bit Windows "WOW64" with 32-bit IE but its "Win64" with 64-bit IE.

    That makes no sense to me.

  2. Arieta says:

    It's so you can make a difference between 32bit IE, 32bit IE running on 64bit windows, and 64bit IE running on 64bit windows.

  3. BartS says:

    @Arieta, if having "WOW64" vs "Win64" is to enable detection of the different versions of IE (32-bit vs 64-bit) then why do they bother adding "x64" to indicate its a 64 bit version of IE?

    To me, they should have set it up this way:

    32-bit IE10 on 64-bit Windows:

    Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Win64; Trident/6.0)

    64-bit IE10 on 64-bit Windows:

    Mozilla/5.0 (compatible; MSIE 10.0; Windows NT 6.2; Win64; x64; Trident/6.0)

    This allows server side detection of the version of Windows by checking for "Win64" only. If "Win64" then its 64-bit Windows.

    Base on their current proposal, I have to do If "WOW64" or "Win64" to see if its 64-bit Windows.

  4. Sam says:

    Which of these userAgents or similar JavaScript properties will tell us if we are in REAL IE10 or the crippled Metro IE version?

    Do we just check in JavaScript…

    if(navigator.userAgent.indexOf('MSIE') != -1 && typeof(window.prompt) == 'undefined')){

     /*

       We are in IE 10(+) in Metro Mode!

       Thus we can't use Flash, the prompt() function, resize windows etc.

       All HTML5 Audio and HTML5 Video must be served up in Microsoft's non-open formats or be turned off/disabled.

     */

    }

  5. Jrd says:

    Are CSS Media Queries Level 4 supported for this? That seems like the "right" way to do it:

    @media (pointer:coarse) {

     …

    }

    I understand that will only work client-side, but I'm not sure that's a bad thing.

  6. Jrd says:

    See also the "hover" media query.

  7. Martijn says:

    Why stick with the Mozilla/5.0 piece? That's probably the most untrue part about the whole string.

  8. Martijn says:

    One more ting:

    var hasTouch = navigator.msMaxTouchPoints > 0;

    DON'T ACTUALLY DO THAT. It's IE-only and therefor deemed bad practice. Instead, check for window.ontouchstart (whose typeof result will be "undefined" if no touch support).

  9. @Martijn: It's come up before, lot's of websites break if the user agent doesn't start with the Mozilla bit and leaving it there doesn't do any real harm.

    What I don't understand is why touch is included in the UA for detecting touch, when UA sniffing is a bad way to do it. Isn't that just encouraging a behaviour nobody wants to see continue?

  10. IE10 & WebGL says:

    Does that mean that IE10 will finally support HTML5 WebGL?

  11. Echo 8 says:

    @AndyCadley:

    When it's server side, all you can do is check the UA string unless you use an intermediate page to ask the user what capabilities their browser has.

    @IE10 & WebGL:

    Um, where on Earth did you get the idea that MS has reversed their position on WebGL?

  12. Jonathan says:

    Why the bloat for 64-bit Windows? Do you expect anyone to be browsing with an Itanium based browser?

    Wouldn't just Win64 OR X64 be acceptable?

  13. Mario says:

    you stupid little robbers at Microsoft.. Linux and Apple had touch screen stuff before you! Give them credit instead of taking their ideas! Think of something of your own for once. like dang! Aren't you your own company or just a rip-off!

  14. WebGL is not HTML 5 says:

    does not matter if you think it is….W3C says its not!

    nuff said….

  15. sad says:

    This post is another sad confirmation that IE10 will not be available for Win 7.

  16. WebGL is part of HTML5 says:

    The W3C isn't the only organization that creates open specifications for the web.

    The WebGL specification is published by Kronos, the same group that is also responsible for OpenGL.

    WebGL is related to HTML5 though and listed as one of the HTML5 technologies on the W3C HTML5 logo page.

    Jump here: http://www.w3.org/…/logo

    "Between SVG, Canvas, WebGL, and CSS3 3D features, you're sure to amaze your users with stunning visuals natively rendered in the browser."

  17. RP says:

    @sad, msdn.microsoft.com/…/hh673549(v=vs.85).aspx states that IE 10 will be available for Win 7.  Let's hope Microsoft haven't changed their minds…

  18. Microsofts Friday July 13th Agenda says:

    Microsofts Friday July 13th Agenda

    1.) Fix the IE Blog Comment system

    2.) Post an update about full flash support in Metro without the censorship list

    3.) Windows 7 version of IE10 beta so that developers can actually test IE10 – something they can't easily do now

    4.) Publish a post regarding Metro support of other browsers – will we be able to upgrade IE10 Metro to a Metro version of Chrome or Firefox?! Especially if that is the only way businesses and users can get complete flash support

    All of this should be accomplish-able by lunchtime.

  19. jader3rd says:

    So since there are different ways of detecting touch, what's the purpose of adding the touch value for the user agent string? How would anyone use this, besides detecting touch support?

  20. Not enough says:

    We also need to be able to detect full blown IE10 from a desktop vs IE10 running on a phone, tablet, or in the Metro mode from the SERVER SIDE.

    I realize that "user agent sniffing" is dirty and all, but we often need from the server side to serve up different files for desktop vs. tablet.

    Detecting this at the JavaScript level is too late… please provide an indicator as to whether the user is in Metro mode or not… e.g. add the keyword "Metro" to the user agent string please.

    Just because a users new laptop might support touch… running in "Real Desktop Mode" I don't want to serve the user up a less than ideal  content experience because I "think" they can't handle it… (or visa versa)

  21. Sarah K. says:

    @Sam, always use feature detection. Try using Modernizr to detect the features. Adobe, Google, Microsoft and everyone else recommend using Moderizr library in web projects for feature detection, rather the poor and notorious practice of sniffing UA.

  22. Prior Semblance says:

    Do all you guys begging for IE10 on Win 7 never read the responses?  It will be ready for Win 7 when the last prerelease version comes out, constantly whining about it isn't going to change the plan they've had for months.

  23. What good is any of the features of windows 8 when Microsoft security essentials is not compatible with it? I was going to install the preview of Windows 8 to get a good idea of what is good and bad about it, and when I got the warning that I would have to uninstall MSE, that was the end of it. I stopped the installation. This is an ongoing problem with a company like microsoft, they have too many butt kissers in different depts and the depts dont talk to each other.

  24. @iRocket says:

    MSE = Windows Defender in Windows 8, which is included in all versions of windows 8 without a separate installation.

  25. L e n n y says:

    @iRocket now you can go back where you came from and kiss their iButt ! Btw, how much did they pay for this cheap act ? An old saying: "He who try to buttkiss the man in iGrave, gets his sorry ass kick'kethd" so GTFOH willya…  😉

  26. PhistucK says:

    @iRocket –

    This is completely off topic.

    However, if I recall correctly, Windows 8 has a built in improved Windows Defender (the features of the old one combined with Security Essentials), or at least an option to download it, so, no worries.

  27. Kosovari says:

    ANY NEWS WHEN IE10 WILL BE RELEASED IN WIN 7 ?

  28. alvatrus says:

    @kosovari

    end of october

  29. Todd says:

    @Sarah K – feature detection is preferred for sure – however when you are dealing with enterprise level content detecting the device at the server is critical!

    More importantly why would I waste a single millisecond to handle IE bugs if I can tell up front that the browser accessing the content doesn't have these bugs!

    The only problem with user agent sniffing is when you expect the worst out of the future versions of a browser… Eg to expect that IE would never fix getElementById(); just because they failed to do so in IE6 & IE7.

    One of the key things I use user agent sniffing for now is if the device is a modern phone or tablet vs. a legacy IE terminal running in kiosk mode on a touch screen.

    For desktop IE on a touch screen I provide a software keyboard because the winXP or win7 keyboard is near useless in terms of being appropriate for a web app… However for a modern tablet like the iPad using the native keyboard makes perfect sense.

    This is easily detected by the user agent and thus no wated http requests for the HTML, CSS, JavaScript for the soft keyboard are made – Awesome!

    Trying to future proof this is where its get hard now because Microsoft has not provided any usefull information in the IE10 user agent string.

    I don't want to serve up extra, useless content to IE10 when it is running in Metro on a tablet but I HAVE NO WAY TO DETECT THIS from the user agent!!!

    Are you listening Microsoft?!?! Feature detection is NOT perfect! And your BUSINESS APPS DEPEND on the userAgent!

    And for the love of God please fix this comment form!!! A blog shouldn't fail trying to submit a comment!!!

  30. Todd says:

    In that last comment "wated" was supposed to be "wasted".

    Omg! How many times do we have to submit content to post successfully! This comment form is horrible!

  31. @ Todd says:

    "Omg! How many times do we have to submit content to post successfully! This comment form is horrible!"

    How right you are! It even contains an intelligence test, and guess who has failed it!

  32. @@Todd says:

    Just what exactly do you mean "@Todd"?!?!

    My comments were right on point, talking about Enterprise Web Apps – where most of the money on the Internet is made.

    If you have an insight to share – do so. If you just came to troll – go elsewhere please.

    If you've followed this blog on a weekly basis since it started you'd know full well there is a problem with comment submission, it has to do with the extremely old ASP postback technology that this blog uses and that Microsoft has been informed hundreds of times about this issue and has even been provided a patch on numerous occasions but has failed to listen to their developers that keep pointing it out.

    If you think I'm being childish by requesting that someone from Microsoft finally pay attention and fix it then you have no idea how the world works.

  33. Xero says:

    I agree with your concern and let me tell you that many (including me) have chased this comment-form-submission issue for a while have discovered the following:

    – Microsoft did not implement this blog system. They are using a third party product since years.

    – Microsoft is using Telligent Blog System on all blogs.msdn.com/*, blogs.office.com & windowsteamblog.com (all different versions of same product).

    – WindowsTeamBlog and blogs.office.com — like most blogging systems on the Internet — require you to sign in before posting the comments.

    – Blog.MSDN.Com, however, have two options; anonymous posting (which comes with the price!!) AND signed posting as well.

    – Anonymous posting would require you to respond within 15minutes after the page is loaded, after that all comments will be discarded as a part of spam protection. The suckiest part is, it won't show any warning and it won't post back the content you tried to post. It will give you no reason whatsoever. Also, it won't notify you whether your post was discarded due to the session-timeout OR the spam filter (use of offensive words).

    – Signed posting would not timeout the session and you can post even after days since the page was loaded (just like other blogging system).

    I contacted Microsoft personnel who quickly directed me to Telligent support. Submitting feedback and even contacting Telligent was just a waste of time and energy. So if anyone reading this happens to personally know someone in Telligent inc., please ask them to implement the notification messages at the very least!!

    After been through all this, in my opinion: Its a bureaucracy that is preventing Microsoft to implement their own blogging system. Otherwise, MSDN tutorials for MVC projects have their own implementation on ASP MVC samples http://www.asp.net/mvc/samples (check out FunnelWeb and Raccon blog), msdn.microsoft.com/…/ee671793.aspx etc…

  34. Steve says:

    @xero thanks for the in-depth explanation.

    I agree with your findings but would also note that the comment form also fails frequently if you post within the 15 min window.  Worse yet if you see the green banner confirming your post you'll need to reload a few times to ensure it was successful as I've found that it doesn't mean for sure that the post succeeded.

    It's been said a few times now though… Why don't they just add the 1 line fix and add an actual HTML submit button?! They could have saved years of frustration simply avoiding using failed-by-design ASP postback techniques.

    The part that kills me the most is the 15min timer… With any decent size article and a dozen comments to read you are all but guarenteed to overshoot that 15min window… Why penalize the reader?! The spam protection would make more sense on the server… Add a time served field to the form… As one of the flags on the backend check the time diff between serving and form submission… If the 2 are within 30 seconds of each other -then you've got a sneaky suspicion that you're being spammed…

  35. You'll regret placing the "10.0" in the UserAgent. When Opera hit the 10 mark of the version several websites that were written clunky had faulty regexes that failed because they did not expect two digits before the dot. Of course, when the sites crashed the visitors did not blame the sites but the browsers…

  36. Stifu says:

    @Andrei Rinea: these sites deserve to break. Web technologies in general have been twisted more than enough for crappy web sites. The ones who'll be sorry are the webmasters of these sites, regardless of who visitors blame.

  37. pmbAustin says:

    If everyone complaining about the comment submission form would just LOG IN, their problems would go away.  Sheesh.

  38. Prior Semblance says:

    Why would a spammer wait 15 minutes to spam anyway?  It seems more like a badly implemented way to force people to refresh and see new comments before they post theirs.

  39. FirefoxFAN says:

    Hey IE I'm a fan of Firefox.. Good to see you are still going on.. but i have a sad feeling IE (The browser share) will fall below 6% soon.. 😛

  40. Andreas says:

    will there be a ie10 for windows7, and if yes when ???

  41. Clark says:

    @pmbAustin – why should we log in on every device we own just to post a comment on the IE Blog?  Why not fix the blog instead of forcing users to login?

  42. Microsofts Tuesday July 17th Agenda says:

    Microsofts Tuesday July 17th Agenda

    1.) Fix the IE Blog Comment system

    2.) Post an update about full flash support in Metro without the censorship list

    3.) Windows 7 version of IE10 beta so that developers can actually test IE10 – something they can't easily do now

    4.) Publish a post regarding Metro support of other browsers – will we be able to upgrade IE10 Metro to a Metro version of Chrome or Firefox?! Especially if that is the only way businesses and users can get complete flash support

    All of this should be accomplish-able by lunchtime.

  43. HTML5 & IE10 says:

    Can someone try the well known and respected HTML5 test suite at html5test.com and give us the score on IE10?

    I'm nervous that HTML5 support in IE10 might not be as great as in Firefox or Chrome…

  44. Name says:

    Please fix this bug:

    If import bookmarks from file: windows crash! All Windows not respond. I force shut down.

  45. Dean says:

    You could hit the URL yourself for the HTML5test.com if Microsoft actually released the Windows7 version they promised but they're so focused on trying to sell Metro UI to customers on desktops that don't need or want it they've failed to address developer needs.

    Likewise they haven't told us how to disable Metro when we are forced by IT to downgrade from Windows 7 to the not-ready-for-prime time windows 8.

    Dean

  46. Browser choice in Metro says:

    Seeing that Microsoft made a "technical error" in complying with the European Commission regarding browsers choice… what is the plan for Windows 8?

    http://www.theverge.com/…/european-commission-investigation-microsoft-browser-choice

    Are we going to see the ballot option across all Windows 8 installations world-wide so that users can install a preferred  browser in Metro and even one with Full Adobe Flash support! (since IE10 will be crippled to a badly controlled whitelist that users, IT depts, businesses and enterprises have ZERO control over)

    Where's that update on Full Flash support Microsoft? We've been waiting for weeks now!

  47. pmbAustin says:

    @Clark, you realize that once you log in, you stay logged in, right?  And in fact, if you are logged out, logging in is just one click.  The wonders of a Microsoft ID.  If you don't have one fo those, one might wonder what your'e doing on this site…

    Seriously, it's ridiculous to *** about.  If you hate the timeout, just log in.  You'll remain logged in.  And you won't have the problem any more.  And if you do get logged out again for some reason, logging back in is just one click.  MUCH faster and easier than whining about the stupid comment form.  Yeah, it's broken, but sheesh.  Grow up a little.

  48. Nathan says:

    @pmbAustin – you're completely missing the point.

    A Blog has 1 intended mode of interaction – Comments.  Depending on the setup, users can either easily add comments (e.g. its an open forum) or they need to login first.

    Microsoft published this blog so that both would work… logged in, or not logged in (the easier way).

    However they published the blog with a bug that would not allow comments to work randomly and unpredictably and failed to fix it for over 6 years. (either by upgrading and getting a fix, or patching it themselves.

    If this was "Joe's plumbing thoughts blog" with 10 comments a year we wouldn't care. However this is a multi-billion dollar software company who ships one of the major web browsers on the market and the blog is about the web browser, standards, and moving the web forward with even more amazing technology.

    Yet here we are… dealing with the absolute worst blogging software available (only because it runs on Legacy ASP).

    What does this lack of interest in fixing the blog / software show about a company that is trying to convince us that they've changed and they are not going to let their browser stagnated for another 6 year stretch?

    What does this show the developer community when the company won't take the time to add a 37 character fix that they've been shown repeatedly, and they know will satisfy the issue permanently?!

    It shows us what we've known all along.  This blog is not a 2-way communication mechanism showing Microsoft's interest in sharing open communication.  Microsoft does NOT care about developers at all.  They only care about their own agenda – and fixing their flagship blog on their flagship blog property where the world comes to see what Microsoft is developing for the Web!!!!  is not something they care about.

    ————————————–

    Truly, Sad Microsoft

    TOTAL, EPIC, FAIL

    ————————————–

  49. Bert Lamb says:

    WTF!? So you don't add a UA indicator or a js feature detection mechanism that IE is running in "Metro" mode, but you do add a UA indicator AND a js feature detection mechanism for touch hardware support.  I am so confused by these decisions.

  50. Microsofts Wednesday July 18th Agenda says:

    Microsofts Wednesday July 18th Agenda

    1.) Fix the IE Blog Comment system

    2.) Post an update about full flash support in Metro without the censorship list

    3.) Windows 7 version of IE10 beta so that developers can actually test IE10 – something they can't easily do now

    4.) Publish a post regarding Metro support of other browsers – will we be able to upgrade IE10 Metro to a Metro version of Chrome or Firefox?! Especially if that is the only way businesses and users can get complete flash support

    5.) Publish an updated UserAgent string for IE10 that indicates if IE is running in Metro mode or not

    All of this should be accomplish-able by lunchtime.

  51. My site implemented the IE9 pinned site with dynamic jump list.

    Is the code be obsolete in IE10?

    Should I target the jump list just for user agent "MSIE 9" ?

  52. Bill says:

    @mocaxu – do you mean those IE9-only… only if the site is pinned… quick link context menus for your site?

    If so then yes it is time to deprecate them.  Once again when you start coding something that doesn't follow the W3C/ECMAScript standards you are going to be walking into a minefield.  You've discovered that using IE-only technology will only cause you pain in the long run.  All of the IE only technology will die – its just a matter of when.

    BMP Support – Dead

    VML – Dead

    VBScript – Dead

    MS Behaviors – Dead

    Marquee – Dead

    CSS Expressions – Dead

    DX Filters – Dead

    Smart Tags – Dead

    createPopup – Dead

    ActiveX – Dead

    Embedded OpenType – Dead

    Silverlight – Dead

    and yes…

    Pinned Site Context Menus – Dead

  53. The distinction between desktop and Metro IE will be a massive headache in parts of the world. I think you might would like to see this:

    labs.signicat.com/…/on-windows-8-and-plugins

  54. Stanley says:

    @Marcus C Almgren – thanks for the very informative Video explaining all the issues with Metro IE, and Desktop IE in Windows 8.  This was the kind of eye opener Microsoft needed to see.

    @Microsoft – this is clearly a major disaster!… if you're not going to fix all of the issues with Metro IE then you **ABSOLUTELY MUST** provide a key signature in the USER AGENT that lets us detect if the user has accidentally tried to use Metro IE to browse the web and that we need to FORCE them back over to Desktop IE for any site that uses DRM, Special Authentication/Security, Flash, Java Applets, VPN plugins, Portals, SilverLight etc.

    If we can at least *DETECT* Metro IE, we can then provide a message/link for users that indicates they need to swtich to Desktop IE in order to have a usable Web.

    Packaging / promotional material for Windows RT tablets running on ARM must also indicate that many business/enterprise tools will not work as they require full desktop IE with a standard x86 architecture for running desktop apps.

    Finally, @Microsoft you need to respond to developers on this blog about all the issues raised!  If you insist on sticking your head in the sand like an ostrich hoping to ignore all the issues you are going to have a miserable RTM launch and receive tons of well-deserved angst at not being open and listening to developers.

    @MICROSOFT – PLEASE RESPOND!

  55. Edward says:

    For developers that had a free PC to install Windows 8 on it… please take a moment to visit this JS Fiddle in IE10 in Metro Mode.

    http://jsfiddle.net/DZEb8/3/

    We are very interested to know if Metro IE10 gives away any indication that a site is running in Metro Mode so that we can inform users they need to switch to the desktop before logging in to achieve the best user experience.

    Hopefully one of these JavaScript tests will reveal that the browser is running in Metro Mode so that we can alert the user to the problem and switch them to the real browser.

    Please report back here to the IE Blog the results of your findings.

    Much appreciated (TIA) – Worldwide Web Application Developers.

  56. More worries says:

    Microsoft is not only helping users and IT managers disable the failed Metro UI on desktops/laptops… but they are actively shutting down any option to subvert it! WT_!? – Are you serious? Have you not listened to the user feedback so far?

    Did Microsoft Just Give Up on Windows 8 for Businesses?

    http://www.winsupersite.com/…/microsoft-windows-8-businesses-143238

    Microsoft is blocking Boot-To-Desktop usability/performance tweaks:

    http://www.osnews.com/…/26022

    Booting as intended in Windows 8 – straight to the Desktop:

    blog.laptopmag.com/…/2

  57. Jaklyn says:

    I tested the JS Fiddle sample that @Edward provided.

    The bad news is that the prompt() function is still defined in Metro IE10… however any calls to it are silently swallowed.

    This means that with JavaScript (like in the 3rd test) we can detect if the user is running in Metro mode – however in order to do so we need to harass Desktop IE10 users with a bogus prompt().

    Does anyone know which other properties or methods have changed in IE10 (specifically in Metro) so that we can properly detect Metro IE and provide a message to users to run on the Desktop instead…. or push a dummy ActiveX control that causes the warning to pop up so that the user is prompted (and auto linked) into the app in desktop mode instead of needing to copy/paste URLs.

    Is window.external still in Metro? I heard that window.clientInformation might be gone now?

    As another poster mentioned it would be best if there were a clear indication both in JavaScript as well as in the User Agent.

  58. Xero says:

    @Edward,  great post! Please take few minutes and logon to connect.microsoft.com/…/feedback and post feedback in their issue tracker. Give us the link for follow up.  🙂

  59. Ricardo says:

    We still need something to differentiate IE10 running on Desktop and IE10 running on Metro. MS should add "Desktop" or "Metro" indicators in the user-agent.

  60. Microsofts Thursday July 19th Agenda says:

    Microsofts Thursday July 19th Agenda

    1.) Fix the IE Blog Comment system

    2.) Post an update about full flash support in Metro without the censorship list

    3.) Windows 7 version of IE10 beta so that developers can actually test IE10 – something they can't easily do now

    4.) Publish a post regarding Metro support of other browsers – will we be able to upgrade IE10 Metro to a Metro version of Chrome or Firefox?! Especially if that is the only way businesses and users can get complete flash support

    5.) Publish an updated UserAgent string for IE10 that indicates if IE is running in Metro mode or not

    All of this should be accomplish-able by lunchtime.

  61. hAl says:

    Microsoft is activly killing Flash ads in Metro browsers.

    Good work

  62. Eric says:

    i hope ie10 will not be as slow as ie9, because my pc is warm and it working all the time when ie9 is on, and i hope it won't be the same with ie10.

  63. Richard Wessels says:

    I will gladly use IE 10 on Windows 7..  However, what I have seen of Windows 8 is beyond rediculous..  Even Apple is smart enough to know that you do not put a mobile / tablet UI on your Desktop OS or vice versa..  This is a terrible move and I fear that it will be seen as such only after this is released to the public.

    Office 2013 = Good, IE 10 = Good, Office 365 for home users = GOOD!!!,  Windows 8 = BAD!!!!

  64. Steve says:

    @Xero – no one uses Connect anymore.  We got sick and tired of Microsoft messing us around, not responding to reports, not keeping reports when changing IE versions, not opening up attachments/test cases so other developers could verify reports etc. etc.

    Until Microsoft makes a written commitment to actually supporting an open bug tracker – We will have no part of it… since it will be pointless.

  65. Kenny says:

    we tried running Windows 8 on a PC here (3 monitor setup) and the whole thing was a disaster.  Major UI issues.  Metro doesn't do multimonitor support and switching from desktop to metro (and back) often causes horrible screen flicker.

    Metro IE has several issues.  There is no flash support (because there are no sites/webapps in the stupid whitelist).  Opening a site in desktop mode from Metro IE doesn't pass cookies, session, etc. so this forces a new login every time (User Experience Failure)

    We can't detect Metro properly so we can't cater our content to the touch interface vs the desktop.

    UI rendering freezes when switching between screens so you get odd trapazoids of screens trying to resize

    Alt-Tab in Metro is useless.

    Alt-Tab in Desktop is ok, and includes your metro app.

    Apps in Metro can self-destruct but still stay open (technically running, but not visible)

    Overall usability to switch back and forth between fake windows and real windows is painful (even for a Windows expert) I think this new paradigm is seriously going to fail hard.

  66. Jacob Rossi [MSFT] says:

    @Martijn

    Detecting support for a particular event type (like ontouchstart) does not necessarily detect that a user has the hardware to produce that event. For example, detecting the presence of window.onmousewheel does not indicate that a user actually has a mouse wheel.

    msMaxTouchPoints detects that user actually has touch hardware.  It also goes further to provide you with the actual number of touches the user's hardware can support (which detecting ontouchstart does not); useful for deciding what gestures a user might be able to perform.  

  67. Marcus C Almgren says:

    Commenting here is like standing in desolate Alaska, where no one can hear you scream. I used to love you guys, what kind of a sick mind makes these kinds of decisions? You're extremely talented people with tons of resources; could someone please explain the Windows 8 master plan to a poor, retarded, M.Sc developer trying to provide the best possible experience for all users on all platforms? Because I don't get it. What's going on in those redmondy conference rooms; do you all have to suck the glass pipe before entering? 🙂

  68. Greg says:

    @Jacob Rossi [MSFT] – please get someone from Microsoft to read all the comments on this blog post and start providing answers/responses!

    Its a disgrace that we get ZERO information from Microsoft about the CRITICAL decisions you are making without ANY developer feedback!

  69. Clint says:

    Unacceptable.

    You can't create 2 versions of a browser and then use 1 user agent string!  Developers need to know which scenario they are dealing with and in this case there is actually 6 browsers.

    1.) IE10 Metro on a Tablet (no physical keyboard/mouse)

    2.) IE10 Metro on a Laptop/Desktop (identical to 1 except that it does have a keyboard/mouse)

    3.) IE10 Metro on an MS Slate (similar to 2 but not quite)

    4.) IE10 Standard Browser on a Tablet (no physical keyboard/mouse)

    5.) IE10 Standard Browser on a Laptop/Desktop (identical to 4 except that it does have a keyboard/mouse)

    6.) IE10 Standard Browser on an MS Slate (similar to 5 but not quite)

    These are 6 very DISTINCT browser scenarios.

    Internet Explorer is used WORLDWIDE in touch screen Kiosk scenarios (running in the traditional desktop mode).  To overcome the lack of a physical keyboard a software based (e.g. JavaScript) one is provided when needed.

    Many of these "kiosk"-type apps are now available on an iPad or Android tablet… and the apps are smart enough to present a software based keyboard when needed (e.g. IE on the Desktop) and not even load it on a tablet… all because the USER AGENT clearly defines what kind of browser it is.

    With IE10 and the current single user agent strategy you've completely killed any chance of these existing systems from using Window 8 – period.

    How does my app know if it is running in the Desktop Browser or the Metro Browser?

    a.) Do I need to supply a software keyboard?

    b.) Can I use popups?

    c.) Can I resize a window?

    d.) Are hover interactions realistically possible?

    e.) Can the user pick precision points? (e.g. not possible in tablet Metro, but fully possible in Desktop Metro)

    f.) Do I need to account for ugly over-scroll (e.g. typical of webkit browser) that ruins the asthetics of an app-like site

    We're not asking for everything here… just some basic info…

    Add "Metro" to the UA if running in Metro…

    – and –

    Add "Tablet" or to the UA if running on a Tablet…

    …optionally add "MS Slate" if running on an MS Slate…

    With this basic info we can provide ALL IE USERS – regardless of their device and mode… with an optimal user experience.  Currently if Windows 8 does not ship with this, many vendors will simply have to warn their customers that Windows 8/IE10 is NOT SUPPORTED.

    Clint