IE development in China

The information published in this post is now out-of-date.

—IEBlog Editor, 21 August 2012

Tony Chor, Rob Franco, and myself are in Beijing today as we make our way to Kuala Lumpur for the Hack-In-The-Box conference. We have a great team over here in what we call the ATC (Advanced Technology Center). In the past 6 months they’ve gone from just starting to adding serious value to IE 7. The folks over here have already contributed to IE 7 by re-writing the select control which Chris Wilson alluded to in his post from the PDC. Other improvements that we’ll see come out of the ATC for IE include improving our bi-directional font support, font linking and fallback improvements, and increased accessibility support. Watch for more blog posts in the future from our teammates here in China covering the kinds of things they’re working on.

After Kuala Lumpur I’ll be heading down to Sydney and will attend the Ruxcon security conference in early October. Hopefully I’ll get a chance to meet some of you down under.

On the road from China (and see you in Kuala Lumpur & Sydney),

 – Christopher Vaughan

Here’s a picture of some of the folks here in China along with Rob, Tony, and myself:

Comments (19)

  1. Rosyna says:

    Are you sure you didn’t mean to say France when you said Kuala Lumpur?

  2. ieblog says:

    Rosyna – I hope not, because my plane will be taking me to Malaysia 🙂

    -Christopher [MSFT]

  3. Anonymous says:

    Confusion between France and Kuala lumpur!

    Give me a smiles 🙂

  4. Anonymous says:

    Can you tell a little more about "increased accessibility support"? One of the main problems I run into as a web developer when programming accessible websites is that screenreaders often rely on the (yet incomplete) DOM implementation of IE.

    You mentioned in another post that <abbr> will be supported, also nice would be recognizing xml:lang for language changes in a document (might have to do with the lack of support for the :lang CSS pseudo class?), support for the hreflang attribute, and full aural/speech style sheet support. There are certainly more issues, it seems like scope in tables has recently been fixed. How about abbr and axis?

  5. Anonymous says:

    Ewww, the photo is JPG. PNG is obviously superior 😀

  6. Anonymous says:

    Nonsense, ‘png > jpg’. PNG is totally unsuitable for photographic images on the web. In the contrary, JPEG is especially designed for that. Use JPEG for photos, and PNG for everything else :).


  7. Anonymous says:

    Btw, cool that you have such a team. Hopefully IE will get great international support!

    What about Ruby?


  8. Anonymous says:

    Laurens Holst,

    You obviously need to read up on PNG, there are two types of PNG, PNG8 (used for images GIF is best for) and PNG24 (used for images JPEG is intended for).

    Generally speaking both types of PNG have a smaller file size than GIF or JPG at the same exact quality settings.

    Of course the best reason NOT to use PNG on a webpage is IE’s admitted lack of all of PNG’s features. So technically the IE team was right for not using a PNG that their own browser only partially supports. (Of course their photo wouldn’t use any of those unsupported features anymays but that’s another arguement.)

    As this blog has stated, full PNG support was a priority of IE7 that’s already been implimented and finished.

  9. Anonymous says:

    Egg on my face, I decided to read up at WikePedia and depending on your use, PNG may or may not be better than JPG for web images.

    If you want superior quality, use PNG for photos.

    If you want the smallest filesize, use JPG for photos.

    Of course with broadband becoming the norm and the fact we’re on the verge of next gen broadband (fiber to your house) so filesize is becoming less important.

  10. Anonymous says:

    Finally, as far as I’m concerned the issue is closed.

    It’s a blog post with a photo, get over it, this is NOT the place for an ill-informed (myself included) PNG vs JPG discussion.

  11. Anonymous says:

    Better font linking and fallback support would be very much appreciated! Thanks for working on this. The current font linking mechanism is very poorly documented and seems to be application-specific. What documentation does exist for it seems to be for developers of new applications/scripts; base font selection is not set up to really work as well as it should in IE5 and 6. We’ve had some discussion of this in recent months on Wikipedia at and has probably been brought up in other areas of Wikipedia, whenever there is a need to use glyphs not available in the base font, as it is not easy to predict when it will and won’t work in IE.

  12. Anonymous says:

    This is quite interesting this blog has picked up as of late.

    Also does anyone know why Internet Explorer 6 Service Pack 1 download for Win 9X/ME on the download site was updated on 9/21/2005 whats new or updated in this release?

  13. Anonymous says:

    What is meant by "font linking and fallback improvements"?


  14. Anonymous says:

    @ GrauW

    IE’s implementation is conformant with the simple Ruby specifications. We have some issues with complex Ruby. Basically, in complex Ruby, you can provide two Ruby strings. In horizontal layout, the two Ruby strings are supposed to appear on top and bottom, but we put both strings on top. In vertical layout, we correctly put the first string to the right and the second string to the left. A bug is filed on this issue.

    I hope this helps

    — Markus

  15. Anonymous says:

    Hi! I’m a program manager on the IE team working on the DOM.

    Martin – I would like to follow up with you about the problems you’re running into developing against our DOM. We’re trying address bugs that accessibility tools, like screen readers, are running into so I would love some examples of where we can and should be investing.

    Kevin – font linking and fallback are used when we are unable to find the character or glyph in the specified font. Without font linking and fallback characters that can’t be found will display as empty boxes. This is particularly useful in global apps when you need characters from multiple character sets.

  16. Anonymous says:

    Does anyone know why the picture below the blog post IE development in China (IE-ATC-team.jpeg) appears way below the post. When I view it in Mozilla Firefox it seems to display correctly and with a border around the picture. I’m just curious. My IE version on Windows XP is 6.0.2900.2180.xpsp_sp2_gdr.050301-1519 if that information is needed. Thank you.

  17. Anonymous says:

    Travis Owens, I probably know more about PNG than you do :). I do not see how I implied in any way that PNG was not *able* to store photographic quality images. What I said was that PNG is "totally unsuitable for photographic images on the web", and that is a fact.

    With regard to IE’s supposedly lacking support of PNG, that is nonsense, it only lacks support for transparency (and even that it has, to some extent). There would have been no problem using a PNG for that photo, except for the monstrous file size it would have.

    Using PNG ‘just for the sake of it’ is nonsense, JPEG is best suitable for photos, and given good settings, there is no visible quality degradation.

    Tony Chor: oh, that’s cool, I didn’t know that! Is IE’s implementation conformant with the spec ( (I tried the example in section 3.4 and it didn’t render as the image below showed), and if not will you look into that for IE7?


  18. Anonymous says:

    Yes I can also confirm that Chris is going to Kuala Lumpur, since I saw his name in the speaker’s list. 🙂 But it’s a pity I won’t be attending…

    As usual, don’t forget to blog about the conference in KL!

  19. Anonymous says:

    Greetings from China (or should I say "ni hao"?)

    Laurens: It is cool to have such a great team here in Beijing. We also have a super team in India, working on IEAK, group policy, and related features. We’ll blog more about them later.

    With respect to ruby support, IE has had support since IE 5 via the <b>ruby</b> element. See <a href="">the MSDN article</a> for more details. (For those unfamiliar with Ruby, it’s a way to display pronunciation information for a word. It’s most commonly used for Japanese.)

    Rosyna: We really did mean Kuala Lumpur, not France. However, if there’s a security conference in France, we’ll definitely consider it!