Latest Update from Molly

Hey everybody! Molly Holzschlag here. As some folks might be aware, I’ve been visiting Microsoft, refining our work goals in relation to standards, and meeting some really great people in the process. One thing that’s really got me excited is how many people from around the company reached out to me with great enthusiasm regarding product evolution in relation to Web standards and interoperability. This did come as a surprise, frankly, I was confident that the IE team is more than interested in doing the good work, but learning that other product teams around Microsoft are paying attention to your feedback and concerns and are beginning to reach out to form strategic plans makes me feel very optimistic. Rome wasn’t built in a day, of course, and there’s a lot of working to do over the coming days, weeks, years and decades. But one thing is very clear, people are interested, listening, and best of all, enthusiastic.

Here’s a synopsis of the conversations and issues we discussed, along with details as to how some of the challenges Microsoft is facing are being prioritized and addressed.

My visit began with a social get together with Pete LePage on Sunday. After enjoying a visit to downtown Seattle, we settled in over garlic fries at Gordon Biersch and began to brainstorm focus for the visit, and what we wanted to accomplish individually and together. Both of us felt strongly that we should each come up with a clear description of role, responsibilities and task priorities in detail. I spent Sunday evening sorting through the main points of our discussion and coming up with top goals.

Monday morning was spent fleshing out goals and deliverables with Pete. The afternoon found me clarifying four activity areas of my role, drawing out a general plan. Then, I mapped each activity area to delivery mechanisms and measurable results. These were reviewed by Brian Goldfarb and Pete, who provided guidance and editorial feedback.

Activity Areas and Related Actions:

Developer and designer community liaison. This focus area is a continuation of ensuring that the discussion between web developers and designers and Microsoft at large remain open. I will continue to attend and participate at events and conferences focusing on standards and interoperability concerns as they relate to all tools and browsers. A monthly resource roundup will be posted here on the IE Blog to ensure that strategic updates like this one, news of importance, articles, podcasts, screencasts and videocasts of interest to the community are published on a consistent and regular basis.

Vendor Interoperability Liaison. In this activity area I will work to achieve the following:

  • Acquire and develop solid relationships for browser and related software vendors and general interoperability solutions
  • Encourage more productive discussions - to quote Chris Wilson from the Browser Panel at the recent South by Southwest event, “The talking smack years are over”
  • Assist with standards and long-term strategy and planning for interoperasbility concerns as they relate to other vendors and Microsoft products and services

Strategic consulting for standards implementation in Microsoft Web-related products. Priority products at this time include:

  • IE
  • Visual Studio  / ASP.NET
  • Expression line of Web design and development tools
  • Additional products as I am able (Office, etc.)

Acquiring and producing educational content related to aspects of working with standards: Accessibility, Markup, CSS, JavaScript, DOM and ASP.NET resources will be acquired, produced and offered to the community at large for both ongoing and internal education.

Wshew, that was a big step. Now that Pete and I were able to get that formalized, the days continued into some interesting areas of actually doing some strategic planning. I met with Chris Wilson and members of his team to address developer concerns regarding feedback and bug reporting mechanisms. This is understood to be a priority concern for both developers and for Microsoft, who really do want to provide decent bug reporting so as to improve developer input on IE and other products, too.

A bit later I met with Saloni Mira Rai. Saloni is tasked (among other things) with solving problems in IE CSS support for printing. We discussed specification and implementation oddities as they currently relate to the way people use and write CSS. This was a highlight for me, as I’ve been studying how ambiguous language in specs influences browser differences. Another meeting, this time with Channel 9, gave Pete and I a lot of ideas and resources for where and how to publish and promote developer and designer resources.

Dean Hachamovitch, General Manager for IE, met me for an early morning coffee Wednesday at Café 9. Dean is always energized when I see him, supportive, honest and filled with questions and great ideas. I have some video clips of our discussion, which ran the gamut from standards to security to browsers to commenting on the gorgeous morning. I’ll be publishing those soon. Late morning saw a meeting with Andrew Jewsbury, Program Manager of Expression Web. We talked about the product’s current weaknesses and strengths, and discussed creating an advisory board of external experts to assist with the direction of Expression product growth.

Here’s an exciting bit for those developers and designers who’ve been clamoring for internal training. Well, it’s going to happen. I’m working with Markus Mielke and Cyra Richardson planning internal standards and best practices on-site training events across Microsoft.

Clearly, a lot of energy, time and money is being put behind the commitment to standards and interoperability within a variety of Microsoft products. Finally, there’s but one goal, and that’s to ensure that Microsoft’s commitment to standards and interoperability as well as all people who use and work with Microsoft products will continue as a priority goal for both the short and long-term evolution of the Web.


Comments (46)
  1. Adam A says:

    Thanks for the update, Molly.

    It looks like you are going to have quite a busy schedule. I am sure that I am not the only one who is really looking forward to the results!

  2. 1 topic to solve them all says:

    This will sound a bit far fetched for some that haven’t worked intimately with IE, but I would like to see prototyping on the HTML Elements working in IE(next).

    For those of us in a controlled environment (e.g. we can ensure users have javascript turned on) Intranet, Internet service, etc. this would be a godsend.

    With the above, we could safely workaround missing/broken functionality at the DOM level, with minimum interference and browser specific custom code.

    and I for one, would no longer need to complain about anything the IE team does, because whats broken, I can fix.


  3. This is quite a task! Is there really that much of a disconnect with standards that there requires on staff training of the subject? I’ve always trained my staff internally, but there’s a level of understanding that you hone your skills on your own time as well. So much of what you’re doing seems like 101 education on the subject of standards, semantics, etc – backend changes to the engines that run these products will be a while down the road I’d assume then.

    On a side note: has there been any feedback about merging Offices’ rendering engine with IE 7?

    Regardless, all of this sounds positive to me, and it’s an enjoyable surprise. Although I haven’t seen as much change as I’ve been hearing (and we are always IE reluctant), it’s a pleasure to know that these subjects are transparent to the community and moving forward.

  4. Just a clarification regarding training – even well-established web designers and developers require ongoing training, not to mention plenty of what I term "spackling" (or polyfilling) the vast gaps that exist in our education.

    What’s more, these aren’t just the designers and developers who work directly on the Web. In many cases it’s simply a matter of informing software engineers about the standards and best practices for web developers. It’s imperative to understand that someone building complex software is going to be focused on the technologies he or she is using – programming languages that I certainly don’t know. So in that case, it’s simply a sharing of what it is we do as designers and developers for subject matter experts who work in other areas.

    Regarding Outlook, I’ve been supported in helping with it but as a priority, no, because it’s not within my department. Microsoft is a big company. Like, really big. So it’s not as simple as it looks. What we’ve decided is that I am free to pursue it on Microsoft time if I am actively addressing other deliverables first, and I can also work on that on my own time.

    So, some community assistance in this would be great. I need to find a good contact who is interested in this within the Outlook team and begin a conversation. Your encouragement and help will be much appreciated.

    Ever yours,


  5. Anonymous Coward says:

    SVG support would be appreciated. But whatever standards you implement, please try your best to do them right the first time.

  6. Ian Muir says:


    I agree that web standards should be web 101 kind of stuff, but unfortunately they aren’t. Also keep in mind that standards compliance goes way beyong just using CSS and having markup that validates. I’ve actually met very few developers who really understand the more subtle aspects of writing good markup.


    As you already know, I’m always happy to help out. Whether it’s providing guides for standards focused development or just be available for Q&A, I’m available. On another note, I’ve worked extensively with Visual Studio and I’ve been giving Expression Web a pretty good workout since SXSW.

    I’m not going to get into the whole debate about combining the IE7 rendering engine into Office. I think it’s a bad idea and people are pushing it for the wrong reasons, but that’s a whole other topic.

  7. Ian:

    I’m aware that standards compliance isn’t simply tableless and CSS designs – if my notation came off abrupt, it wasn’t intended too. I am simply taken aback by how old a technology this is now, and how much there seems to do.

    As a note, what are the downsides of combining the IE7 rendering engine? My only gripe is how poor MS Office renders it’s content – I already have fearful clients coming out of the woodworks for their evil (albeit somewhat of a necessity) html newsletter. I agree that fixing IE is a larger priority – outlook on the other hand has even my PC clients in dread right now.

    Molly: thank you for the response – as I noted above, my opinions maybe abrupt at such an old topic, and that’s not my intention. I can understand having to ‘spackle’ development training… I had a similar discussion with the developers for Lightroom, who’s 1.0 had a web gallery that chose an XHTML 1.1 doctype, and then made every tag uppercase without a closing declaration. But that is a desktop app with web aspirations, I find it interesting when the web is their focus. Regardless, I’d be happy to help though I have little experience in desktop app development save bug submissions, tests and the like, if you need a web coder/designer for support, happy to ablige.

  8. Melianor says:

    Hi Molly,

    Great to hear about this update and indeed it sounds like alot of work. I looking forward to hear more and meanwhile we’ll build a little shrine for you 😉

  9. anony.muos says:

    Dont forget to add back the ability to customize toolbars, buttons and their layout 😉

  10. c says:

    Wow, that post sounded like bureaucrat-in-a-blender.

    On a more serious note, all the work to get developers to "modernize" on a div+CSS layout won’t help international sites that try to do things unsupported by CSS, like right-to-left layout.  Table-based layouts flip perfectly and automatically.  Div+CSS… doesn’t.

  11. Steve says:

    I have no idea where to post this question exactly, but here it goes:

    I switched to IE7 but now there are certain webpages I can’t see, and I get an error that says, "Error: object expected".

    So I look online for help, and what I find is this (in IE7 Tools->IE Options->Advanced):

    1. Check the box: Disable Script Debugging, IE

    2. Check the box: Disable Script Debugging, other

    3. clear the box: notify me …

    So I make these settings changes and restart IE7 and restart my PC entirely just to make sure, but still get the same error.


    My wife’s PC still has IE6 and I the page pulls up just fine.

    ALSO …

    When I open the page in Firefox, the page pulls up just fine as well.

    This is very frustrating.

    It seems like every time I update to something new, I get problems.

    A few weeks ago, I did a Windows Update (automatically) and next thing I know, my sound is disabled on my computer!

    (BTW, it took another update – thank God – to fix the sound issue.)

    Can anyone help with the IE7 scripting problem?




  12. Matt C says:

    Hi, I’m in the same boat as Steve. I’d like to debug some scripts without having to shell out for Office, so I’m using the free "Script Debugger" instead of "Script Editor".

    It’s okay with me if this tool isn’t as useful. All I really need is for the line with the "error" (though the whole site works perfectly in other browsers) to be highlighted. That’s all I need, and about 1/3rd of the time the debugger actually does that!

    But could you possibly get it so that IE (6 and 7) doesn’t crash every time it opens?

  13. crazzeto says:

    I’m just curious if there is some sort of time table for MSIE’s many CSS + printing issues. I’ve found that IE7 is particularly bad at properly handling the page-break-after/before style. About the only think you can guarantee is that it won’t work correctly.

    For instance you might (by some mirical) actually get the page layed out correctly for default print margins, but if the end user changes the print margins (in my specific case to make them bigger) IE will all of the sudden decide to start ignoring the style. This particular bug is *REALLY REALLY REALLY BAD!!!!* Bad enough that I’m surprised IE shipped with the bug! Very frustrating because it’s taking what has the potential to be a great web-browser and runing it.

  14. thacker says:

    In complete agreement that the IE engine should not be a basis for Outlook. The Word engine is ideally suited for an e-Mail client. The major argument supporting standards is the benefit across the board for all users and developers. The want of a browser engine for an e-Mail client simply to allow easier delivery of HTML branded content to accommodate a very small percentage of the product’s customer base [web developers, et al] is hypocritical at best. Other and more efficient technologies are available for delivery of branded content. E-mail is text [ASCII] based communication.

    Would very much like to see an update to the HTML Help compiler, CHM, updated for standards and for security.  The 1.4 SDK was released in 2000, if not mistaken.

  15. thacker says:


    Regarding your issue, or at least one of them, with IE7– can you provide links to the sites that are causing you problems?

    Am not so sure that my knowledge is sufficient enough to provide reliable input as to those rendering issues.  However, this blog is read some pretty damn knowledgeable people, that if links are provided, they could very well answer your question.

  16. usability says:

    much agreed that usability is important, thus on the MS task list for IE8 I see:

    * Fix customizable toolbars

    * Fix dragability of links in/out/and all around IE/Tabs/Form elements

    * Fix :pseudo CSS events, so that we don’t need to do JS hacks to style something when it has/gets focus, or is active

    * Fix :hover/:focus on select elements to not stop propogation of events

    * Fix <button/> element to follow spec on submit, so that it can be used properly on the In(ter|tra)net

    * Fix the ‘C’ in ‘CSS’ to actually mean Cascade, so that nested links <a> do not need exclusive class/style declarations to work properly

    * Fix printing in IE7 to not consume/overlay  the entire browser window, make it minimizable, make it not disappear when you click print and move the mouse, and fix the printing of form elements (suffers the same weirdness that zooming a page does)… some form control "chrome" scales, others stretch, and it looks very strange

    As for the chatter on Outlook.. I’m 50/50 on this. I would like to see more capability inside the client, windowless select elements etc. but from supporting IE6 and IE7 on the Web thus far, I don’t look forward to having to handle 2 strategies for email clients!

  17. thacker noted: "The want of a browser engine for an e-Mail client simply to allow easier delivery of HTML branded content to accommodate a very small percentage of the product’s customer base [web developers, et al] is hypocritical at best. Other and more efficient technologies are available for delivery of branded content. E-mail is text [ASCII] based communication."

    I’m trying to wrap my head around the idea that html branded newsletters is a small market; the product customer base is not web developers, it’s small to large businesses that rely on HTML newsletters for marketing potential – designers are just the means. The design benefit for this has very little room for debate, so I’m unsure how you can call standardizing a method selfish if your approved method doesn’t fall under selfish as well? Nevertheless, I’d be interested in finding out what other, more efficient technologies are available to deliver branded content into an email client? Design emails will be around for quite a long time, even if people like me prefer RSS and web based publication; the goal is to have a unified appliance of a technique, not to write your own interpretation. In essence, a standard.

    In regards to word being an improved rendering engine for email, is this based on your statement of ASCII text being emails foundation?

  18. thacker says:

    Worldwide, 1.1 billion e-Mail users. 171 million e-Mails, daily.  These are, I believe, 2006 figures as compiled by The Radicati Group.

    The bulk of these transmissions, best guess is, obviously, that it is not HTML based but text based. The number of web designers/developers, worldwide, .. again, best guess is that it is substantially less than 1.1 billion. Assuming, for the sake of argument, 1 million designers, including Frey’s business argument who want delivery of HTML markup .. well, the math works out to "pretty damn small percent".

    The comment made of "small percentage of the product’s customer base" stands as stated.

    The majority of exception to Outlook 2007’s fall back onto the Word engine revolves around the problems of not being able to use standards compliant markup so that it can be rendered as designed by Outlook 2007.  The solution is the fall back to table based design and ‘dirty’ markup.

    Personal preference as a designer:  Hell, yeah for standards based e-Mail delivery. But to ask for it, let alone expect it is counter-productive to use of e-Mail and, more importantly, defeats the argument of standards benefit across the board.

    Even a 1% [10 million designers and business users] customer base wishing to decree standards for their sole use while ignoring the vast market share of all users smacks of hypocrisy in my view when compared to the general argument that use of standards should be adopted because they benefit everyone.

    The hypocrisy comment stands as stated.

    Standards do benefit everyone when applied to Web communication.

    I do not have first or even 100th hand knowledge as to what prompted Microsoft to fall onto the Word engine, but my best guess would be that business use of e-Mail, including that of the Office Suite, added considerably to their decision.  This was the primary basis for my statement that the Word engine was the proper choice.

    The better choice for delivery of branded content, RSS feeds with links to available branded RIA, Flash, PDF, WPF/E .. downloadable content or live Internet content. Opt-in/opt-out is no longer a privacy issue and the user has complete control. Standards can be applied in those methods. Arguments of additional security improvement by reduction of HTML based e-Mail may have merit.

    The comment of "more efficient technologies are available for delivery of branded content" stands as stated.  No comment was made nor implied for delivery of more efficient technologies of branded content into e-Mail clients.

    E-mail should be left for its largest market of [171 million per day] text based communication users.

  19. tranditional content layout using table tag. as css become popular, more and more designer using’s box modal different with other browers althought with !DOCTYPE declaration it can switch presentation modal. but always float this css rule trouble me. 100% width can only using 99%, for others browers supproted W3 standard there no this problem. CONTENT LAYOUT is a big issue. as i know i do not use css’s float, i only can use table markup. there’s no more alternates.

    dom table-related element’s property .innerHTML, can not on-runtime write this attribute.

    javascript + asp  Reponse.BinaryWrite(?) using vbscript is easy.

    ASP.NET comsume lots memory. webpage is just a flush stateless if we not think about session,application,etc. why not design a class for just-in flush contents?

    here my scripts fregement

    var sn = Snake(bcached);





    if we can let css do more work, so we need consider the importance the HTML TAG commonly used attributes. do you think it’s a good idea?

  20. caw says:


    SVG is an old technology

  21. Dave Bacher says:

    What would be really useful is if when maintaining backwards compatibility, Microsoft could use a custom processing instruction tag instead of DocType.

    Versions of Internet Explorer, ASP and other products do not necessarily coincide with versions of the standards.  An HTML 4.01 page may need to run in backwards compatibility mode, etc.

    In an XML document, Microsoft Internet Explorer looks for a ProgID Processing Instruction, and then uses that to open the document.  It would be incredibly useful for Microsoft to use this mechanism, and not the DocType, to switch on features such as Quirks.

  22. Aedrin says:

    "Fix the ‘C’ in ‘CSS’ to actually mean Cascade, so that nested links <a> do not need exclusive class/style declarations to work properly"

    For whatever purpose would you need nested links? That just sounds like poor HTML to me.

    [a href=""]One link [a href=""]and another link[/a][/a]

    What concept are you trying to imply? That link 2 is related to the other? It just doesn’t make sense…

  23. All good points thacker (I’ve emailed our Postini rep for some numbers of HTMl vs. Text email NOT spam related, for some added interest if needed), and I can agree to an extent on some of your arguements, especially that RSS and other web deliveries of content are a preferred focal point of sending notifications that are styled, which if you read my previous comments, I already pointed out was a preferred format (and my assumption was then you were hinting at other methods for stylized email, as you cleared up). However, the basis for your numbers is total email… and in the case of arguement here, I’m talking business professionals and concerning myself with Business email. Especially since MS Office is, atleast in their demographic, more of a target for businesses whether small or large. All businesses that sell a product, from small local work I do, to large companies such as send stylized emails as a point of market and alternative customer outreach. The percentage of this transmission for daily business activity I would assume is huge, although I haven’t dug deep enough to find quality data to back my opinion on this. While typical email communication is going to be text based – marketing outreach is going to be largely based on html emails.

    So my criticism maybe misdirected for my goals here: My clients overwhelmingly want to send HTML emails for customer outreach, business growth, and alterior marketing potentials. This ideal is largely similar to other businesses where I do not hold them as clients. These businesses may use Office products to design the occassional email and send them out to potential clients, but in the end will eventually use a designer or developer to automate or standardize the process. While Outlook holds a majority on desktop clients: hotmail, yahoo, aol, gmail are still serious players in the delivery of email content. Their rendering of stylized emails has it’s flaws, but consistency of some standardized design method for this market is a necessity for businesses to continue to grow. While marketing emails maybe a small percentage based on your numbers compared with average everyday emails (which is most likely the same as reading web content and target advertisements on web content ratio), it is a lucrative, and finacially driven market; both for the target demographic of Microsoft, and for corporate america. Marketing is after all how they reach a larger customer base.

    We may not like this, we may think it’s insignificant compared to other email communications… but it is a money maker, and it is a concern. If there is no standard to the design method, we have a financial burden to these companies and a failure to their marketing methods.

    From your perspective, I can see that the idea of rendering word documents or creations back and forth would make the rendering of outlook with word’s html capabilities positive. The best of both worlds would be that Office uses the same rendering engine as IE – so that there’s continued consistency across the board for these companies. Having a consistent rendering engine would eliminate the floodgates (and by consistent I mean Office, Outlook, and IE if we’re going to push this route), and pose no inconvenience to the larger base of email users you’re defending as the majority.

    I still find no qualms with wanting unity on how any of their applications render stylized, html content – it causes no harm to text email, and fixes the failure of multiple rendering engines.

  24. Chris Beall says:

    "I met with Chris Wilson and members of his team to address developer concerns regarding feedback and bug reporting mechanisms. This is understood to be a priority concern for both developers and for Microsoft, who really do want to provide decent bug reporting so as to improve developer input on IE and other products, too."

    I found this statement interesting, as I’ve just spent about 3 hours on the phone over the course of two days, just trying to report a rendering bug in IE 7.  Everyone I spoke with was patient, polite, and ignorant.  They did not have IE 7 installed on their systems; they had to put me on hold and go to another system to reproduce the problem.  They pointed me to information on VML, which was not in use on the failing page.  I was passed from one technician to another, about 8 in all.  Each such pass started with a repeat of requests for my phone number, email address, and "Now what problem are you having with IE?" although all of this data was already entered into the case record.   Every word that was spoken to me made the assumption that I was doing something wrong and that the technician would work hard to help me correct my error, rather than even acknowledging the possibility that the error was really in MS code.  I was asked to do things that were obviously unrelated to the issue, but appeared on the technician’s diagnostic charts.  Each pass from one technician to another involved an impressive name, such as ‘Professional Support’ or ‘Research Team’, but the knowledge level of the people did not approach those titles.  My issue is now with ‘Development support’ and the last person I spoke with gave me some confidence that my report may actually reach the development team, but the time and emotional cost of reporting this one rendering bug has discouraged me from ever trying it again.

    So, how WOULD I want the system to work?

    1. A public bug-tracking system, where reported bugs are never deleted from the system, but remain as a historical record, with a symptom-based search engine.  This would let me see if something I perceive as a bug is already known to Microsoft and being worked on or perhaps has a workaround available.  The current KnowledgeBase shows primarily problems that are ‘resolved’ via customer workaround.

    2. An on-line bug submission tool available to anyone.  It can have whatever data-requirement constraints are needed to prevent frivolous or incomplete reports from being filed.  It should strive to collect enough data so that the problem can be reproduced, at which point it becomes Microsoft’s problem and not mine (I can track it using the public tracking system).  It could even incorporate the diagnostic scenarios now used by those technicians, but with an option at each step to say "This question is not relevant to the problem I am reporting."  This would let me report bugs on my own schedule and reduce my exposure to music-on-hold.  It would also use computers, rather than people, to do rote work.

    3. A policy that only the person originating a bug report can close that report, i.e. the submitter must agree with the resolution, unless the submitter declines to provide documentation necessary to reproduce the problem.

    Here’s what ‘support’ looks like from out here in the real world.  Start at The Introduction says "You can submit issues that are bugs or suggestions." (rising hope).  So you click on ‘Available Connections’ (where no registration is required, that’s good) and find ‘Internet Explorer Feedback’ (euphoria).  Clicking on that brings you to "Thank you for visiting the IE Feedback Site. The site is temporarily closed. It will re-open in the future." (depression).  But there’s a link to the "Internet Explorer 7 Support page" (cautious hope).  Skipping the basic FAQs on that page, you click on the TAB for "Internet Explorer Support" where you find… a phone number, the same one I called two days ago (loop to top of this post) and a pointer to microsoft.public.internetexplorer.general, where I had posted my bug report on November 27th and which has never received a response (thoughts of violence to self and others, not necessarily in that order).

    Chris Beall, who does not translate "This is understood to be a priority concern" into "We’re going to fix it".

  25. html born with browsers. if we do not use such as css,dom,script etc, we also can do necessary works, inserting images,text,linking to another addresses, layouting contents, etc. I think these functions are enough for common bussiness situation. Why we now using css,script,dom??? yeah! apparently, facing some sophiscated business areas, web designers design a page not simply fill contents for HTML document, they need consider codes of structure & architecture of server pages more. they need easy maintainbility and extensibility. for example, css for HTML, is departing some functions such formatting, layouting from HTML, thus HTML can do what it shoud do. ie versions are changed so quickly, but it do not concerns the important&basic functions  should be fixed and implemented "first", on the contrary, it concerns not very useful new features more!

    conclusion: list the most important & basic functions still bugged in IE which should be fixed and implemented in time. even it can be solved using a tricky.

  26. steve_web says:

    @Chris Beall: 2 words…. WELL SAID!

    You’ve hit the same nerve we developers have been hitting for days, weeks, months on end now.

    Its not that the Public bug tracking sucks,

    Its not that it is flawed, and you can’t mark bug b, as a dupe of bug a.

    Its not that it requires a login

    Its not that it isn’t searchable

    Its not that the response time is horrible

    Its not that the issues don’t seem to get resolved quick enough

    Its not that the bugs found aren’t acknowledged


    I T S   T H E   F A C T  T H A T   I T   J U S        T   D O E S N T   E X I S T !!!

    Please, please, please, will someone take the initiative to get a public bug tracking system online… the old IE Feedback was at least usable, whereas now, we have nothing.

  27. ash says:


    Right underneath the option ‘disable script debugging’ you’ll find an option named ‘display notification about every script error’.  Uncheck this and IE will no longer notify you about script errors.  This is disabled by default.


    They probably have good reasons for not placing it back online.  Perhaps IE8 will be written with a different engine.  Or perhaps they found the old one quickly filled up with a lot of repeated and redundant information.  Given the hate out the for MS and IE, there is a fair chance it would be filled with rubbish.

    If you want it so much, why not open one yourself?  Then let everyone know.  I’m sure there are free options around.

  28. Adam Simpson says:

    Thank you for your hard work, Molly.  I’m a Web developer with a small software company and I look forward to better standards support in Microsoft products.

  29. Anonymous Coward says:

    Speaking of standars…. Any chances you can get them to fix the <hr /> tag? The IE implementation of that is terrible and it should have been fixed in IE7.  Try styling it once.  Or do a google search on how all everyone else has to hack it to work (they end up resorting to <div>’s).  Try putting a height on your <hr />, giving it a background, and getting rid of the stupid border IE gives it.  Doesn’t work does it?  Yeah looks like you guys have some work to do.

  30. Chris Beall says:


    You replied to steve-web, regarding an open bug-tracking system "If you want it so much, why not open one yourself?  Then let everyone know.  I’m sure there are free options around."

    Interesting idea.  Let’s test it with a thought experiment:

    We put up a tracking system.  Anyone can report IE bugs on it.  We define criteria for a bug report, in terms of what supporting documentation is required, to weed out the "IE is broke" reports.  We moderate, to weed out the "MS sucks’ reports.  We include a search engine, so you can hunt for duplicates.  We let you tag a bug as being of interest to you, so we can judge pervasiveness.  We add  information on workarounds as it becomes known to us.


    – If you encounter a bug for which there is a workaround, you can use the workaround.

    – If you encounter a bug, you can find out if it is known to the group, thus saving the time required to document it.

    – If a bug is pervasive, we can focus group resources on finding a workaround.


    – Somebody has to set it up.  Needs a server.

    – Somebody has to moderate.

    – Bugs don’t ever get FIXED, because there is no process to report them to MS (pretty much as today).

    – If we DO find a way to push the bugs over to MS Development, there’s no way to tell if or when they are likely to be fixed, thus avoiding the need for a workaround.

    – Nothing we would do is official, so if a workaround breaks with a new release of IE, we can appeal only to our own resources.

    From my perspective, this does not sound attractive.  I want to create standards-compliant websites that work anywhere.  I don’t do browser sniffing, which eliminates many workarounds.  My objective in reporting bugs is to get them fixed, so that neither I nor anyone else has to think about them again.  Without at least one-way communication with MS, I don’t see how that happens.

    Chris Beall

  31. thacker says:

    Anonymous Coward–

    As of yet, I haven’t experienced any issues with the horizontal rule and styling it with CSS for IE 7. It has a caveat that if you want a colored horizontal rule, it requires the same color applied to both background color and color. This will produce a warning in CSS validation.


    hr{background-color:#ee7511;border:1px solid #333;color:#ee7511;height:5px}

    The above style produces an orange horizontal rule, 5 pixels in height, with a 1 pixel black solid border.

    To eliminate the error use the following:

    hr{background-color:#ee7511;border:1px solid #333;color:inherit;height:5px}

    Then add the following to a conditional IE style sheet:


    If you are experiencing specific issues, possibly post them with the CSS code and what your objectives are?

    Chris Beall–

    Molly Holzschlag did not make such a comment. It was another poster. IE Beta forum was taking a lot of heat and abuse — to the point that it became a damn zoo. My guess might be the development team is waiting for things to calm down for a bit, letting people discover that some issues are not true bugs but poor code in the content — before resuming a much needed bug tracker/workaround/resource forum?

    I am not kissing anyone’s butt .. but it took some courage to introduce IE 7, deliver it as a critical Windows update and abandon years of FrontPage silliness. That includes a lot of corporate Intranets that were built around IE 6, built without standards or forward thinking.

    I can appreciate the frustration. Been there, done that, said to hell with the tattoo but seriously considered the MacIntosh t-shirt.

  32. This is very exciting stuff and I will be glued to the blog to see how this all plays out. Thanks for the update Molly, thanks for all the hard work you are doing to make this happen.

    If, or should I say when, Microsoft starts doing things right it will be a huge step forward.

  33. steve_web says:

    @Chris Beall,

    The 2-way correspondence is the critical issue.

    There were dozens and dozens of bugs filled in the old IE Feedback site, that were originally reviewed by MS staff, and received comments to the effect of "Hmm, we didn’t realize that this was broken" to issues that the development community had known about for years.

    Several hundred thousand developers and web enthusiasts will find bugs in IE faster than any developers/testers/QA team at Microsoft, simply by numbers, and variety of applications.

    Mozilla Firefox gets 1000’s of issues raised in their bug tracking software, on a similar basis, however because of the 2-way communication and community participation, the bugs get narrowed down to simple reproducable test cases, get assigned to a developer/team to fix, and well, get fixed!

    Without this 2-way communication, IE’s quality and standards/spec compliance can never be as good as any other browser out there, whether Safari, Konq, Firefox, Camino, Opera, or whatever.

    Oddly enough, even with all the anti-MS sentiment out there, most developers just want the issues confirmed, addressed, and fixed.

    Long story short, if IE is fixed up, and on par with the other modern Web Browsers, then there won’t be anything to complain about.

    There’s an old saying that says if you do something right, observers might tell 1 person about the good thing, but if you do something wrong, the observer will tell 9 friends how bad it was.

    If you bought a Starbucks coffee, and you got an extra nickel back, you’d hardly think it worthy of telling anyone..

    But if you handed over a $20, and you only got a quarter back (and there was nothing you could do about it*), you’d be livid, telling everyone you met that day/week/month.

    *This is what IE bugs are like.. you suffer with an issue, but have no control vote or even say as to getting it fixed.

    Microsoft, please re-open a public bug tracking tool for Internet Explorer, this isn’t a want, or wishful desire.. its a need.

  34. Chris Beall says:


    I realize that the comment I quoted was not Molly’s, but rather her paraphrase of a statement made to her by Chris Wilson.  As he does not seem to be present here, Molly was the only one I could respond to.

    My response to Chris Wilson would be, "Show me".  It is easy to say that something is a priority, but much more meaningful to actually take effective action.

    My response to Molly would be that I have great respect for anyone who can negotiate with MS for more than an hour without banging their head against the wall in frustration.  Extracting a promise for action is a significant milestone.  Nailing down the details of what that action will be and then gently but assertively pressing for it until it actually occurs is, however, the real measure of success.

    Chris Beall

  35. Chris Beall says:


    re: 2-way communication, I agree completely.  Firefox and the other second-tier (in terms of usage) browsers get good developer support because developers can see the results of their own actions.

    How many times do you ring a doorbell before you decide nobody’s home and just walk away?

    Chris Beall

  36. Chris Beall says:


    Back on April 4 I posted my experience trying to report an IE 7 rendering bug to MS.  Here’s a followup:

    I was contacted via email (which I generally prefer over phone calls) by a Support Engineer with the Internet Explorer Development team.  He had reviewed the data already entered into the case record (a first!) and correctly restated the issue.  He then stated that "This is a known bug." and provided me with a workaround which should function across all browsers, i.e. it conforms to standards.  I will need to implement this on all impacted pages as the workaround alters the HTML.  In my case that is trivial, so my issue is resolved.

    But I did raise two issues with the Support Engineer and his manager:

     1. Since this was a known bug with a workaround possible, why was there no KB article on it?

     2. Why isn’t the bug tracking database public, which would have saved us both the cost of a duplicate bug submission?

    The answer to the first question was that KB articles are created only for severe or pervasive bugs (mine is neither, depending on your definition of ‘severe’).  The answer to the second question was that, yes, there was no public access to the bug database.

    Finally, the engineer suggested "You can always post your question or problem in the msdn forums as they are monitored and answered by our development teams."

    After perusing the msdn forums, I have responded to that suggestion.  It is not yet apparent whether any further action will result.

    Chris Beall

  37. PeteLe says:

    I just posted a post about the IE Connect database in response to thacker, steve_web and Chris Beall.  I’d love to hear some feedback on it.

  38. dek says:

    come here.. the biggest free ebook colelction..

  39. dek says:

    come here.. the biggest free ebook colelction..

  40. Brian Sexton says:

    It is sometimes impressive to see gears turning, pistons pumping, and steam blowing, but as both a Web developer and a Web user, none of that makes any difference for me unless something of reasonable quality actually rolls out of the Internet Explorer machine and into the hands of Web users.  Yes, the IE 7 rendering engine is better than that of IE 6, but it still not as good as those of several other Web browsers; the lack of proper XHTML support may not seem like a big deal to many, but the still-horrid CSS support in IE certainly is.  Every employee in every group at Microsoft could jump up and down waving the standards of Web standards and interoperability, but until it seems that all of the supposed enthusiasm is actually going to result in products with better standards support and interoperability in the foreseeable future, the rest of the public seems unlikely to share it.

    Now that the honeymoon with Internet Explorer 7 is over, the excitement has died down and we are left with the reality that we are in the same situation we were in before: Internet Explorer is inferior to other browsers and Internet Explorer is holding back the Web.

  41. James says:

    What is Microsoft’s position on the WHATWG being used as starting point for HTML 5?

    Or, because aspects of the specification overlap with proprietary technologies being developed by Microsoft, oh, I don’t know, lets say XAML, will it be shunned, basically screwing over everyone who doesn’t dance to your tune?

    What’s the spin this time?

  42. Chandra says:

    Great Work Molly.

    To all the great engineers at MicroSoft involved in IE, Can you please provide us with a simple way to have IE6 and IE7 installed side by side for testing our sites? I am sure it would not be that difficult if you really decide to do that.

  43. netvance says:

    Thanks for the great input Molly! As independent software developers we appreciate your work very much! Greetings from Vienna 🙂

  44. thacker says:


    The official Microsoft solution for running other versions of IE is to run them on a virtual server.

    However that is not a practical solution for many. This application will allow you to install IE3 through IE6 on a non-VISTA system with IE7 installed.

    For those who wish to test linear structure and accessibility usability on the Lynx text only browser with an easy install for Windows [I have not tested the install on VISTA]:

  45. Dating says:

    Hey everybody! Molly Holzschlag here. As some folks might be aware, I’ve been visiting Microsoft, refining our work goals in relation to standards, and meeting some really great people in the process. One thing that’s really got me excited is how man

  46. Weddings says:

    Hey everybody! Molly Holzschlag here. As some folks might be aware, I’ve been visiting Microsoft, refining our work goals in relation to standards, and meeting some really great people in the process. One thing that’s really got me excited is how man

Comments are closed.

Skip to main content