W3C SVG Working Group Update for January 2010

In this post, I want to share some examples of the progress going on in the SVG Working Group. Microsoft recently joined the SVG Working Group, and other members (Mozilla, Apple and Opera among others) welcomed us warmly. I'm hopeful about the ways that SVG (both its current direction and future potential) could make the web better. We want the spec to be clear, consistent, and predictable for developers. We’re working out ambiguities such as “Pointer events and clip-paths”, “CSS Selectors <use> and as well as inconsistencies with stroked-dasharray” and “<use> and its interaction with the DOM and rendering” so that web developers can write SVG once and know that it will be interoperable across browsers.

I have to admit I was a little hesitant at first to get guidance and clarity on a dozen or so items we found to be ambiguous (see public SVG WG discussion threads), however the positive response has been overwhelming.   Of course we are not the only members raising these issues, but we are happy to be a part of the process.  The future of SVG is bright.

Additionally, Microsoft looks forward to hosting the next SVG Working Group face-to-face meeting in Brussels this May.

A special thanks to those on the Working Group for their warm welcome and shared goals of creating a specification that will promote standards based interoperable graphics for the web.

Patrick Dengler
Senior Program Manager
Internet Explorer Team

Comments (76)

  1. Anonymous says:

    @sooooo : Please stop being stupid.

    Why do you think Microsoft has joined the SVG working group? Just for fun?

    @hmmm : I’ve just tried your link. It freezes also IE8 on Windows XP SP3.

  2. Anonymous says:

    @L re: "over-deliver" hahahaha I’ve never laughed so hard – you’re obviously not talking about the IE team… their motto is:

    "promise-nada, fail-to-fix, remove-features and then claim-to-have-delivered" motto

  3. virusprotection says:

    IE 8 really is great and i must say there are some more developments we can expect from Microsoft.

  4. Anonymous says:

    @nitro: lol, you don’t know when to quit, do you?

  5. Anonymous says:

    Thanks! I didn’t believe you’d ever implement it.

    (while you’re working on making IE not suck, don’t forget DOM2!)

  6. Anonymous says:

    @John A. Bilicki III – I’ve been to your site and although you keep up with technology in the browser your site is so 1989 looking that it isn’t funny.  I’m not sure what work outside your homepage you do but this doesn’t serve you well as a portfolio site. "Welcome  to Version 2.8 Preview IV, the twenty eighth version of JAB Creations, the home page" – yeah, no thanks.

    Sorry dude.

    As for IE. I don’t use it now and unless IE9 does "rock" I doubt anyone will switch back.

  7. Anonymous says:

    @John Bilicki III

    You’re reply to my question unfortunately failed to teach me anything new.  That IE8 supported this daft xsl stylesheet trick in standards rendering mode is something I knew; it still doesn’t mean that the document is being properly processed as xhtml.

    There are many subtle differences between how javascript and css should be treated for xhtml and for html documents and the DOM should be different for both.  XML should further allow the use of namespace aware methods.

    It is true that this funny application/xml trick causes IE to test for document well-formedness.  However, beyond that, it is to be doubted that the document is treated any further as true xhtml by IE.  That will have to wait for IE9 (or later).

    It is further debatable, to say the least, to say that refusing to render ill-formed xhtml is a clear blessing.  For automated xml processing, it should certainly be a boon, as it prevents garbage being processed into sewage, but for the human-oriented www, it is possibly quite disruptive.  I think you are wrong indeed to suggest that people might learn more quickly of html faults due to browsers not merely providing a warning, but refusing to render them at all.  Have you never heard of html validators?

    I don’t want to be rude, but if that website is the pinnacle of your web development, it is rather hard to take you seriously.  The aesthetic is terrible, the technical construction perhaps even more so.

    A few brief examples:

    a) the document doesn’t even validate due to the presence of two id="news" twice;

    b) the document renders IE6 into a quivering heap, such as I had to put it out of its misery with the Windows Taskbar;

    c) the use of the javascript pseudoprotocol [please read e.g. http://crisp.tweakblogs.net/blog/the-useless-javascript-pseudo-protocol.html] all over the place, what is wrong with onclick – or even better binding event listeners as inline scripting is generally poor style anyway;

    d) the use of non-breaking spaces to indent paragraphs leaped out too – what on Earth is css for???;

    e) why not use rewrites to do something with those dreadful uri’s with their long list of ? parameters;

    f) javascripts dumps users in modal alert boxes if there has been some error by the programmer – how friendly is that???;

    g) javascripts use huge long if var = value else if var = other value else if var = yet another value, etc. instead of switch statements;

    h) javascripts have every more or less declared as globals, no attempt to even namespace stuff;

    I could go on forever, but will stop here for the sake of my own sanity.

  8. Anonymous says:

    Thanks for these articles, I enjoyed them!

  9. Anonymous says:

    @Gord If you the lack insight that others have then it’s not my problem but if you’re going to grief keep in mind when it comes to you and I; I’m the one who has in the past and will in the future continue to be able to demonstrate to great length what I am capable of while at the same time nothing came up for "gord" on Bing or Google that I could associate a real person with.

    IE9 is going to rock and I for one am glad that it is already creating a clear distinction between those with legitimate concerns such as standards and mere internet trolls.

  10. sooooo says:

    Please stop being vague.

    Is SVG going to be in IE9 or not?!

  11. Steven Roussey says:

    Without promising, I think the amount of talk about it is a clear enough sign. Considering IE has VML already, and MS Word has MathML already, I think it will be doable to have SVG and MathML ready by the time IE9 ships. Throw in some CSS3 and some GPU based web page acceleration, and I think developers will be happy. Well, not until they kill IE6.

    Please improve the very slow developer tools. Or open source at least that part.

    Oh, and extension writing via COM objects is so Windows98…

  12. hmmm says:

    unrelated but still….where does one post bug reports?

    this link :


    freezes internet explorer 8 every bloody time…on Windows z (x64) and windows 2008.

    Other browsers have no problem with it.

    I’m actually going to switch JUST because of this problem…and the fact that i can’t seem to find a bug report page πŸ™‚

  13. saitir says:


    Am using Win 7 x64 and IE8 and not getting any lockup with that link.  Its open in another tab right now and is scrollable and its links are clickable.

    I’d suggest you have plugin issues rather than it being a generic IE 8 problem.

  14. infis says:

    SVG Support , Canvas Support and addEventHandler



  15. hAl says:

    @Steve Roussey

    MS Word uses OMML for math. That is a new math xml markup format that can be mixed with other office (layout) xml markup unlike MathML (and SVG) which cannot be mixed.

    MS Office only contains indirect MathML support. The OMML used by MS Office can be transformed into MathML using XLS Transformations (XLST). The layout markup that was mixed in is then lost.  

  16. LenardG says:

    It is so nice that Microsoft joined the SVG group, but quite frankly, what took so long? And when will we see SVG support in IE?

    Is the "lets release a major version every couple of years" really the best philosophy to develop a browser nowadays? Wouldn’t it be better to have a component based browser where you can ship small updates and new features more frequently without breaking the entire beast?

  17. Jon says:

    Great stuff! I’m particular pleased that you’re concentrating on SVG use cases for building graphically rich web apps. Really looking forward to what IE9, and other browsers, deliver in this area. Keep up the good work Patrick!

  18. Mohammed says:

    Guys i think Canvas is also the future, make your decisions wisely so don’t left behind.

  19. blah says:

    You just know the other members couldn’t hold in their snickering.

  20. Fake Al Gore says:

    I can confirm @hmmm’s problem. I am using Internet Explorer 8 on Windows Vista x64, and it freezes. I have even tried it using No Add-ons mode found in the System Tools folder.

  21. 8675309 says:

    currently the the experince really does suck across browsers example i used to play a game called hullbreach online which was based on opera canvas tried it on songbird(ff) didnt work so i gave up

  22. Jeffrey Gilbert says:

    Every canvas/svg use i’ve ever seen is CPU intensive garbage. Whatever it takes to make that change, do that. Or maybe we just shouldn’t be using them. I don’t want illustrator like functionality gobbling up cycles in my browser. Keep it simple.

  23. carlos says:

    Thank you for sharing your workgroup experiences.

    Do you have some public schedule regarding IE SVG support? Can’t wait to see some SVG code working in IE.

  24. 8675309 says:

    flash is one of those other cpu entensive apps that should dissappear

  25. Steven Roussey says:

    @LenardG: From the tea leaves I’ve read, I think the new version of IE will come out with each version of Windows (every three years) with one interim version, putting them at around every 18 months. Just how many version of how many browsers do you want to support, anyhow?

    At least with Firefox/Safari/Chrome you can demand the newest version of each from your users since those users likely went and installed it. But with Chrome getting bundled on machines, etc., that assumption/requirement may have to get changed.

    That said, I agree that it should be faster while IE is so far behind and is in catchup mode. Faster would be better here. Two releases between Win7 and Win8 would be in order.

    @hAl: Thank you for the reference to OMML. My only experience with MathML is making Firebug work with it in Firebug 1.5.

  26. Snark says:

    Yeah, I hate it when things use my CPU!

    I prefer when it sits idle, looking pretty and shiny in the case.

  27. gabe says:

    @Steven Roussey relasing two versions would only slow down devlopment it takes time for quilty assurnce etc and if they relased 2 versions that quick not much would be improved

  28. Steven Roussey says:

    @gabe: Yes and no… it really depends on what happens to be ready and when. If, for example, the GPU and JIT work was ready now, but the SVG and other stuff aren’t, then testing might be improved by not having to test too many changes all at once. Moot point however, as I don’t see once a year happening, though they have done it before…


    1995 – IE 1, IE 2

    1996 – IE 3

    1997 – IE 4

    1998 – IE 4.5

    1999 – IE 5

    2000 – IE 5.5

    2001 – IE 6





    2006 – IE 7



    2009 – IE 8

    Kudus not just to MS for engaging the SVG community, but to the SVG community for engaging MS in meaningful ways (rather than just a lot of rants like on the comments of this blog).

  29. L. says:

    If they are discussing ambiguous items in the specs, it probably means that they have started working on an implementation.  (Remember the start of the work on CSS 2.1 for IE 8?  They were discussing ambiguous items and churning out test cases.)

    On the other hand, this is pure speculation, there is no mention of plans for SVG support in IE.  This probably means that the work is not done yet, so they are not allowed to discuss it publicly.  (The "under-promise and over-deliver" motto.)

    Well, this looks good.

  30. Fred says:


    It says what is says, and only MS know if that "probably" means anything more/less.

  31. John Sausage says:

    Whoa! Slow down, first of all the TinySVG subset has to be implemented, there shouldn’t be that many ambiguities there.

  32. John Sausage says:

    Meeting in Brussels, going over to the EU then for a talk?

  33. John Sausage says:

    @LenardG: No. Huuuge updates were always more successful, see for example Vista… (SCNR).

  34. LenardG says:

    @Steven Roussey: I have been working with and developing enterprise application and I do know that for many enterprises there are policies in place that prevent browser updates. They are too afraid to upgrade, because all the web applications would need to be tested and checked so nothing breaks. Many also rely on IE.

    But I don’t think this should be something to block the development and release of new versions, especially because – as you also pointed out – IE needs to catch up with other browsers. The Chrome update model suits me – it just updates itself when there is something new. I do realize that might not suit everybody, of course πŸ™‚ But it might have prevented such things as IE6 still being majorly present. Just ask our Web Designers opinion of that πŸ™‚

    While I also understand the reasons behind connecting IE versions with OS releases I still think it badly affects new features not getting out fast enough to the existing user base.

    @John Sausage: Well with Vista came IE7, and after IE6, it was about time to give something new to IE! No wonder it was popular when it added some features to IE that were greatly missed. But adding useful and good features one at a time could also be greeted with excitement.

  35. Hi everyone,

    I’m also pleased to see evidence from the IE team regarding will to involve in the SVG community (joining the SVG WG and increasing activity in its mailing list being clear proof). πŸ™‚

    I agree with Steve [1] that its not only the IE team but also the community which must be involved in the process; the initial feedback seems nice, and I hope to see further progress on both. πŸ˜‰

    Disclosing a schedule on the availability of an SVG implementation within the IE roadmap would be great, but not a requisite (i.e., if Microsoft/IE team wants to make a surprise to the world, that’s fine also!). I just recall the past in obscuring this sort of information, giving the impression that standards compliance is being stalled, reason why I’m highlighting this roadmap thing (but I’m convinced you already are aware of that, in the scope of CSS2/3). πŸ˜‰



    [1] http://blogs.msdn.com/ie/archive/2010/02/01/w3c-svg-working-group-update-for-january-2010.aspx#9957208

  36. So far a lot of things I’m hoping to see in IE9 have been confirmed. Honestly as much as I can’t wait to try it out I think if it really only took three weeks of work to add GPU acceleration and CSS3 selectors that I’d rather wait for the RTM by an extra month or two if IE9 does not become the Netscape 4 at the end of this decade.

    I’m pretty sure we might see application/xhtml+xml support in IE9 and after getting application/xml to work in IE8 in standards mode and honestly I do prefer the full page break that IE, Opera, and Gecko execute instead of the half-rendering in WebKit (as much as I do like WebKit for other reasons).

    I’d also really like to see CSS3 multi-column support make it in to IE9. Really, it makes reading pages across my 24 inch screen much nicer. Don’t get me wrong, CSS3 selectors are great but it’s CSS3 properties that will make or break browsers in ten years.

    Shadows and transitions are another things that are pretty much a must-have. Honestly while I think jQuery is nice it’s NOT a replacement for JavaScript. In fact I would go so far as to say if IE9 RTM is covers all the bases that we really need that JavaScript frameworks (not just jQuery) in general will become the next Netscape 4. There are still an absurd number of people surfing with dial-up and frankly I’m really getting irked by the rise of sites that automatically resolve JavaScript issues with jQuery instead of JavaScript directly and then the people asking the questions presuming it’s somehow acceptable to not learn JavaScript and to just accept jQuery or any other framework as somehow an acceptable "replacement" for JavaScript. I’d like to see the web free of JavaScript frameworks by the end of this decade and I think Microsoft could really help the web in this regard should you folks cover all the right bases. I’m not trying to be disrespectful to the folks who work on frameworks (they’re geniuses for making things just work as well as they do) however they primarily exist to enable older browsers to do things supported by very little code in newer specs (e.g. * {-o-transition-property: background-color, background-image, border, color, opacity; -o-transition-duration: 0.8s;}). So again I’ll reiterate my most important point: it’s perfectly fine in my opinion to spend those extra few weeks to get those extra things in to IE9 if it means that we won’t be cursing IE9 at the end of the decade.

    Going back to the blog topic while I don’t work with SVG I’m happy to see it finally getting the support it deserves. I think it’s awesome you guys are working with others in the standards community and look forward to seeing Microsoft bringing it’s own contribution of innovation (e.g. AJAX, favicons, etc) while simultaneously embracing standards. This decade is going to seriously rock for everyone. πŸ™‚

  37. marek says:

    Well, it is good news to the web vector community!!! :-)))

  38. nitro says:

    Hello Mr Patrick Dengler

    We are very interested in technology VML and SVG vector and we want to present the

    application PAGEDITOR PRO V8. this function online in IE 5.5 – 6 – 7 to 8, many websites have

    been made in VML since 1998 to date in France. The VML and HTML + TIME are integrated

    into our platform WYSIWIYG and facilitate the achievement of animated graphics without


    Why reinvent the wheel?

    Here are links for more information on the VML:


    Demo PAGEDITOR PRO V6 (2005):


    Best regards

    Patrick Guinberteau

  39. steve says:

    @nitro – Glad to see that you had success with VML in IE.

    However the problem is that it only works in IE and it is a legacy technology.

    Web Technology these days must not be proprietary and must work in all browsers.

    HTML5, SVG, Canvas, etc. are all open technologies with specs that can be adhered to.

    If I build (and I have) an application that uses SVG or Canvas it works across multiple browsers… Firefox, Chrome, Safari, & Opera.

    Only IE is behind with these technologies and it is hurting IE as a development platform.

    Today (2010) if I was asked to build an application in VML or VBScript or make use of ActiveX calls or proprietary JScript, or only work in IE I would refuse point blank.

    More importantly the best development and debugging tools are only available for Chrome, Firefox etc.  I would hate to have to use inferior development/debug tools to build my application when there are so many better tools available.

    Luckily IE6 is on the way out. Existing apps support it, but new ones will not.  The faster IE can catch up to modern browsers like Firefox and Chrome, the faster the Web and all Web related technologies can flourish.


  40. Mitch 74 says:

    @nitro: why reinvent the wheel? Because even MS deprecated VML in 1999.

    @JAB3: IE8 still doesn’t support application/xml in Strict mode by default. However, you can force it through X-UA-Compatible.

  41. Mike says:

    Great to hear you’re adding SVG….finally…after how many years?  

    What about the rest of the HTML5 specs?

  42. Matthew says:

    @Mike: When you make mistakes like that, you make it obvious that you don’t understand your own question. Troll elsewhere please.

  43. @steve: Canvas and HTML5 are not w3 specs, they are drafts. VML is not a w3 spec either, but a submission:


  44. John Sausage says:

    Quoting John A. Bilicki III:

    »This decade is going to seriously rock for everyone. :-)«

    You mean the following one?

  45. dtrim says:


    SVG has its own specs, the SVG specs is not part of HTML5 specs.

  46. Paul McKeown says:

    @John A Bliecki III>>>I’m pretty sure we might see application/xhtml+xml support in IE9 and after getting application/xml to work in IE8 in standards mode and honestly>>>

    Have I missed something… application/xml and IE8…?

    I know the old declare an xsl stylesheet trick as text/xsl was discovered about 2003 (it’s documented, for instance, at http://www.w3.org/MarkUp/2004/xhtml-faq#ie) and was shown to work at the time with IE 5, IE 5.5 and IE 6.  It even works when the xsl-stylesheet is declared in the html as application/xml, but to my knowledge it doesn’t work if the MIME type for the xsl stylesheet is set to application/xml (set using AddType in an appropriate .htaccess or in the httpd.conf).

    It is true that IE checks for well-formedness errors, but it then proceeds to handle the served document further as html (there are dozens of ways that this is visible).

    I’m not sure what use this particular trick is at all, and  it certainly doesn’t allow xhtml to be served to IE with its correct application/xhtml+xml MIME type.  It almost seems, to be brutally honest, as a sort of sick joke.

    However, you mention IE8 specifically, so perhaps there has been some development that I am not aware of, in which case I would be very happy to hear more…

    I agree with you, btw, that xhtml support in IE is long overdue.


  47. @Paul, IE has supported XML since 5.0; what I was talking about was application/xml support with ***standards mode enabled***. The IE team unlocked standards mode for pages served as application/xml in IE 8.0. So even if you can’t directly served XHTML 1.1 correctly it’s "okay" to serve is as application/xml.

    The main benefit of XHTML and using either application/xml or ideally application/xhtml+xml is if there is an XML error on the page the entire page will break. Some people might think that is about as undesirable as it gets and they couldn’t be any more wrong. In just one of many examples I remember a thread about a guy who checked everything from PHP, his database, everything because his page worked fine in everything except Safari; hours later he discovered he was missing a quote. Had he served his page as XHTML he would have known the issue and had it fixed literally in seconds, not hours. So XHTML is a *MAJOR* must-have just for that one benefit.

    Unfortunately IE does request the doctype URL located on the W3C’s website and add to the fact bots have been mindlessly wasting their bandwidth which all but prevented pages from being served as application/xml in IE 8 in standards mode (and in quirks mode only in older versions). However ironically HTML5’s not-a-doctype doctype (I personally feel the doctype should retain the version number for forward compatibility) will work and IE will *not* make, depend, and thus fail to render the page.

    There is a second thing necessary to making IE8 render a page as XML, include the following line (Dean Edwards gets the credit for this) after your XML declaration…

    <?xml-stylesheet href="themes/xhtml.xsl" type="text/xsl"?>

    IE8 does *NOT* require X-UA-Compatible meta element to render application/xml pages in standards mode.

    However when serving a page as UTF-8 if you need non-Latin characters to display correctly you need to use the following meta element in IE7 or older (I think though I’m a few days from finishing work on IE specific issues on my "nightly builds" so-to-speak)…

    <!–[if lte IE 8]><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><![endif]–>

    Internet Explorer is in a few ways more capable then a lot of people give it credit for though if any one who is blindly biased is never going to be interested in the redeeming points to anything.

    IE9 is going to rock and I hope there are things I can find that few others will. πŸ™‚

  48. steve says:

    @Steven Roussey – quite true, HTML5 and Canvas are still in draft mode… but that didn’t stop browsers working on supporting them.  Sure some of the details need to be worked out… but a "date" element as implemented by Opera is pretty much on par with what was expected.

    Ditto for the Canvas stuff that works just fine in Firefox,Safari,Chrome and Opera… since they all followed the "draft" specs.

    I’m not to worried about whether the specs are in draft, or finalized… its the commitment to implement them within a reasonable timeframe that is the issue.

    I was developing in SVG 8 years ago… have bought printed books on how to do it… but support in IE just isn’t there.

    There are times when I feel that IE is not just behind the curve… more that IE isn’t even on the curve yet.

    I’d like to see a post that commits to "trying" to support SVG in IE9.  So far it has all be very wishy washy non-commitment which doesn’t make me confident in IE/MSFT at all.

  49. steve says:

    @John A. Bilicki III – "IE9 is going to rock".. thats a pretty bold statement seeing that most of us have not seen anything concrete about IE9 other that it is supposed to support CSS rounded corners and reasonable speed JavaScript.

    Granted those 2 things are great – but hardly earth shattering since all the other browsers have had both for years.

  50. @John Sausage This decade if you know what you’re doing, maybe the next decade if you don’t…maybe.

    @steve You don’t actually think I only read the blog do you?

    Different people, different perspectives. Sometimes you will bump in to people who don’t know some things you do and sometimes you’re going to bump in to someone who knows somethings that you don’t. You can visit my site and mess with the functionality…not my latest and greatest (2.9 in March, only a lucky few can see my nightlies) but I bring a lot to the table as far as being able to show I know what I’m talking about. I don’t speak for Microsoft in any way though rest assured the excitement a lot of people will have reasonably soon is the excitement I already have now.

  51. Wow this is interesting! Is it also available in spanish?

  52. Hualong Dong says:

    @Cursus Spaans: No, SVG support will not be available in the Spanish IE versions. :o)

  53. iPhone addict says:

    I’m trying to figure out what SVG is all about. Is it comparable to Flash?

  54. Mark Malone says:

    Fabulous!  This is great news.  Next step HTML5 and CSS Effects!

  55. Useful compatibility layers:

    * Cross-browser VML implementation in Javascript:


    * Implementation of Canvas HTML5 command-based drawing for IE:


  56. If the saints can says:

    If the Saints can win the SuperBowl then surely IE9 can support SVG!

  57. Steve says:

    LenardG is on the money!…  =D  

    IE needs to have the option to update it self…  

    @ IE team… if you had problems with auto updates in the past then that is probably an OS thing too?…

    My Mac updates it self no problem and I have had both mac and windows for over 10 years…

    Or when my mac updates or prompts me to update it is not super annoying like other….  for some reason mircosoft, firefox, and adobe update / security prompts / alerts… I find EXTREMELY annoying and aggravating…  The mac on prompts don’t bother me at all and I actually like them…

    Maybe you should do lots of usability and design research on an IE auto updater…

    I hope this helps.

  58. * promising SVG (partial) implementation (in JavaScript+Flash I think):


    * cross-api text strocking capability for Canvas and VML: http://www.netzgesta.de/dev/text/

  59. Jones111 says:

    @Steve: IE will never update itself. There’s Windowsupdate and it works for all programs by MS. The only thing missing there is the lack of updates for other software. But I agree that there is a need for updates and that there should be a security warning if you haven’t updated your browser and your av-prog.

    By the way: I’m very excited about the beta of IE9.

  60. geld lenen says:

    IE 8 doesn’t seem to get along with W3C SVG very well πŸ˜‰

  61. steve_web says:

    @John A. Bilicki III

    MSFT has confirmed that "application/xhtml+xml" websites/applications will NOT be supported in IE9 (MSIE Feedback marked as "postponed")


  62. gabe says:

    firefox all the way id never go back to internet explorer id do without internet before using it

  63. Paul McKeown says:


    >>>MSFT has confirmed that "application/xhtml+xml" websites/applications will NOT be supported in IE9 (MSIE Feedback marked as "postponed") https://connect.microsoft.com/IE/feedback/details/527016/ie-cant-open-application-xhtml-xml-websites>>&gt;

    You’re referring to an IE8 bug log. Postponed might well mean, "not fixed in IE8, available in IE9."

    Think, I’ll wait for MS to give us the story.

  64. Paul McKeown says:


    >>>MSFT has confirmed that "application/xhtml+xml" websites/applications will NOT be supported in IE9 (MSIE Feedback marked as "postponed") https://connect.microsoft.com/IE/feedback/details/527016/ie-cant-open-application-xhtml-xml-websites >>>

    You’re referring to an IE8 bug log. Postponed might well mean, "not fixed in IE8, available in IE9."

    Think, I’ll wait for MS to give us the story.

    [Resent with a readable url.]

  65. EricLaw [MSFT] says:

    Paul is correct in noting that the resolutions in the CONNECT database are from IE8, and do not reflect IE9 plans (one way or another) at this time.

  66. steve_web says:

    @EricLaw, Paul McKeown…

    Since the CONNECT DB is still actively taking in FR – Feature Requests and PR – Problem Reports on IE8 and IE9 (as discussed here on the IE Blog and in the Connect updates) and the issue I linked to was opened JANUARY 25, 2010 (e.g. for consideration in the IE9 development cycle)

    Lets face it, anything that isn’t a high-level security bug report (against IE8) in Connect is automatically pushed to be a bug/feature request against IE9. (If it isn’t, that’s a MAJOR flaw with Connect… but I digress)

    Since the original bug: 527016 was closed as postponed in the last week or so, one can only assume that that is the "freshest" info on the matter.

    It isn’t like the existing bug: 445283 (posted 5/8/2009) is getting any attention.

    All that aside.  If you’ve announced IE9 development is under way – why isn’t Connect opened up for IE9 bug/feature submission?

    I must confess – I absolutely hate the way IE bug/feature submission is handled through Connect.  Totally the wrong tool, totally un-community friendly, totally opaque, totally un-helpful, totally not improving.

    What happened to all the comments about this whole public bug tracking thing getting fixed?

    At this point I’m completely un-motivated to submit any bugs for IE9 if this process isn’t fixed up.

    Sorry I tried to resist ranting but I’m fed up with the lack of transparency and progress with interfacing with the development community.

  67. Tom Stack says:

    Steve_web:  Its not so much connect is the right tool. It is pretty flexible and each team gets lots of options. Its how each team uses connect.

    I think that fact that since the final realse of IE 8 the team rarely posts in the comments or closes bugs untill a year later.

    Are you really telling me none of these bugs could be fixed to qualify for inclusion in a IE cumulative update?

    Also in that year they cant comment on what they are planning, Seems like they are only open when they want to be, the IE team needs to pick up the pace and start talking more in between releases. about what is coming, and about what bugs can qualify for fixes in bimonthly cumulative updates.

  68. Cartelett says:

    W3C: Why did they choose SVG ???

    VML was quite al(l)ready and infinitely better than SVG!



    15 years has been lost…

  69. Rex says:

    @Cartelett – One of the big issues with VML is that although it was an open standard (according to wikipedia) and based on XML – it suffered from all the issues that IE suffered from with interacting with it.

    e.g. you couldn’t use .setAttribute() on elements if the attributes fell in to the pit of attributes that IE chokes on. e.g. ‘style’

    or you would need to use CaMeL cAsE because the original IE developers didn’t have the foresight to implement clean APIs.

    (see z-index issues)


    Then you bundle in the MSO (Microsoft Office) "extensions" to VML and you end up with a non-pure open standard issue.

    On the flip side with SVG you have a format that started open and refused to adopt sloppy non-standard intrusions.

    It was (and enforced valid XML) and worked with the standard open ECMAScript DOM API.

    I haven’t built much with VML… other than to fix IE6/IE7 bugs with PNG Alpha and rounded corners so my perspective on how easy it is slanted – but developing in SVG is simple (as it is with Canvas)

    Also keep in mind that development of the VML format ceased in 1998… thus it is a dead technology.

    SVG on the other hand is active, it actually is a W3C recommendation (for 7 years now) and has a community of developers actively pushing its adoption and growth.

    We’re just waiting for IE to finally jump on board and support the format that every other browser supports so that we can develop for the open web.

    The sad news is that even if IE9 comes with native SVG support there will still be users with IE6,IE7, & IE8 for years to come holding the technology back.

    It all rolls back to the fact that IE holds back the web by not getting on board with standards early and attempting to implement various proprietary extensions and features that we don’t want.

    In this day and age if it isn’t a W3C recommendation (or well on the way to becoming one) developers don’t want it.  The only way to succeed in web development these days is to stick with standards… anything else is career suicide.


  70. Observer says:

    @Rex: You’re contradicting yourself when you say:

    "The only way to succeed in web development these days is to stick with standards… anything else is career suicide."

    On the contrary, the only way to succeed is to deliver what the customer wants, and as developers usually observe, that typically includes support for IE6 and not "standards for standards sake".  Customers want interoperability, which isn’t the same thing as standards.

  71. Rex says:

    Say what you want @Observer but when my clients suggest they want IE6 support we kindly inform them that an 8 year old browser is out of date and doesn’t provide the rich experience our software was designed to offer.  We support IE7 and up or any of the better browsers.

    We haven’t lost a customer yet – and most are quite happy when they get a better browser (IE8, Firefox or Chrome typically) and thank us for pushing them/their IT dept. to upgrade.

    I did my last webapp code checkin for an IE6 bug in March 2009 and on March 31st we dropped support for IE6.

    It was the BEST day of my development career thus far!

  72. Cartelett says:

    It seams that "Standard" is a supreme word for many "regular" people.

    But W3C is not always on the right way…

    The only thing I can see: SVG is to date far to be as complete as VML, and is not able to do what VML can do. Just some minor details are lacking to consider VML as an according standard.

    Finally…SVG is progressively looking for to be like VML (15 years have been lost)

  73. apps says:

    I think VML is much better, too. What is the reason not taking it?

  74. Btw, the issue with Connect is that each product version (say IE9) is considered as a separate product.

    That is they make IE9, and treat IE8 as another produt. This isn’t the case with Mozilla and other non-MS products, they have a steady progress and feature/bug tracking throughout the years, not throught a release and support cycle.

    I really hope MS changes this for flagship products like Internet Explorer and Media Players. Another reason for this treatment is they’re linked to respective Windows releases which just makes it harder for all (why bother to pursuade MS to fix something in WMP if you may need to wait for the next Windows to see this fixed?)

    MS is now copying bug reports to an internal DB, should use Connect instead in my opinion and keep the feedback loop open for every product (e.g. there’s no way now one can send feedback via Connect for say PowerPoint 2007)

  75. Cartelett says:

    VML is the only reason I’m surfing with IE.

    I am sure that SVG is on bad bases to correctly grow. This format has a much too fat and complex language indeed.

    VML is not as dead as you say and it comes back with interest particularly in France and China.

    It’s not too late to change! SVG is seriously unwell done and VML is too unknown by developpers.

    If you examine VML DOM, you will discover many Macromedia tracks. These latter have built the Flash success. (Adove has understood that…)

    I can’t see any serious future in SVG vector system.

    PS: No particular problem with .setAttribute() and VML elements…

  76. Rex:

    VML easily offered to me (2003) a web application that I could not do with SVG:


    I totally support this marvelous format.

    I don’t suffer it only runs on IE but why not VML on every browsers? It is so simple and so accessible.

Skip to main content