MSDN vs MSDN2: What Links to Use


I’m Will Mason, and I manage the developer documentation team for Internet Explorer. We write the documentation for HTML, DHTML, CSS, WebBrowser Control , etc. that you can find in the MSDN Library.

You might have noticed lately that the MSDN site has changed its domain name to MSDN2. Beginning last fall, the MSDN (Microsoft Developer Network) group has been using a new publishing tools and had to keep the content published with the old tools separate from the new. As you have probably figured out, MSDN2 is the site that uses the new tools. 

You might also have noticed that the URLs have changed more than just in the domain name. To keep our file names unique, the file names are for the most part now page IDs, rather than page titles. For example, “item” is a property of many objects and there are several “item” topic pages in the Library. The IDs make sure that no two “item” topic pages can have the same URL.

Over the winter, documentation sets were slowly migrated from one domain to the other. At some point in the future, the MSDN2 domain will once again become just MSDN and will keep the new scheme for file names.

So, what does this mean to you? Well, actually not much. If you’re creating links today, you will have to use the MSDN2 URLs. Old URLS that point to MSDN (the old domain name) are redirected to MSDN2, so you don’t have to worry about them.

If you link to MSDN2 pages, redirects will be put in place when MSDN2 reverts to MSDN. These redirects will not go away. Over time, you might want to correct them, but they will always work.

For more on the reasons behind these changes, please see http://msdn.microsoft.com/msdnmag/issues/05/02/InsideMSDN/default.aspx. Keep watching the http://msdn2.microsoft.com  for more information.

– Will

Comments (37)

  1. Nate says:

    thanks, Will! I’m glad someone finally explained this. i’m glad msdn now displays properly in vista, also 🙂  images were broken for months.

  2. are they going to fix the samples? says:

    So does this mean that the samples/documentation will get fixed from:

    <HORRIBLE OUT="of date" TAGS="and" ATTRIBUTES=quoteless>This is not how HTML should look.</HORRIBLE>

    (below square brackets used instead of angled to avoid deletion)

    [HORRIBLE OUT="of date" TAGS="and" ATTRIBUTES=quoteless]This is not how HTML should look.[/HORRIBLE]

    It is the first thing on MSDN that indicates that most of the content is over half a decade out of date.

  3. I take it creating human-readable URLs for all the documents would be too big a task? Many would have to be extremely long to ensure their uniqueness, so it’s a good balance to do this for the ones which are easy.

    Great to see a real-life example of how to use redirects to prevent linkrot in an enterprise-scale web application. 🙂

  4. whang says:

    when will you autoupdate IE7 in korea,china and japan?

    we want to autoupdate ie7 in korea,china and japan because of idn(international domain names)

  5. Edwin Martin says:

    Great job! It’s really much better now. And it also works in Firefox.

    http://msdn2.microsoft.com/en-us/ie/aa740474.aspx

    One minor point: the url’s are quite ugly now. (And looking at the html-code: that’s quite ugly too. Using "<a href=’javascript:…’>" is a bit outdated, but okay, it works.)

  6. jvierra says:

    Why is MSDN2 so much slower than MSDN especially under IE7.

    Every click seems to cause teh left menu set to be reloaded in it’s entirety and re-rendered as a menu.  Why does this menu have to constantly reload?

  7. steve says:

    what about the known issues?

    Glad to see a new site for the technical information, however there is still a key piece of info missing:

    Links to the known issues.

    For example, if I am looking at:

    http://msdn2.microsoft.com/en-us/library/ms535841.aspx

    The HTML Input type=text element it gives me info on what attributes it supports etc. but is missing info on what doesn’t quite work as expected.

    If there were a "known issues" link, that listed all issues known to affect this element, it would be appreciated.

    1.) Naming the field with a name starting with a number e.g. "4FirstName" is not advised since attempting to access this field via JavaScript later will fail.  var foo = document.forms[idx].elements[‘4FirstName’];

    will return the 4th element on the form, not the field matching this name.

    2.) Saving Autocomplete data for this field will only work if the form is submitted via a Submit button.  Calling FormObject.submit(); will not save the data.

    3.) CSS background images aligned to the right of this control will flicker when the user types beyond the visible length of the text box.

    PS The sample to interact with this field and display the field’s value, is broken and does not work, and the sample HTML code should be:

    <input type="text" value="" name="textbox" size="15"/>

    PPS A link to the future public bug tracking system would also be a great idea, for when developers find a new bug that isn’t documented yet.

    thanks,

    steve

  8. The DHTML is cool and the design looks good overall. Are the web pages for each division of Microsoft’s website designed by people in those divisions or are there some full time employees dedicated to that task?

  9. hAl says:

    The MSDN2 pages load very slow and look terrible when opening a new link in a small windowpane in IE6.

    Also losing the verbose links is a loss as they made it easier to navigate and interprete links.

  10. Scott says:

    MSDN seriously needs to drop all references to DHTML. It’s not an official standard and is synonymous with poorly written JavaScript.

    http://msdn2.microsoft.com/en-us/library/ms536913.aspx

    See what I mean? Poorly written JavaScript and HTML.

    <SCRIPT LANGUAGE="JScript">

    needs to be changed to

    <script type="text/javascript">

    The onclick event should be attached to the body element in the JavaScript; stop encouraging people to use event attributes.

    And as mentioned above, the HTML needs to be updated too. There are no attributes TOPMARGIN or LEFTMARGIN on the body element. Element and attribute names should be lowercase. Attribute values should be quoted.

    Another thing I notice, is that HTML element which have been deprecated in HTML 4.01 are not marked as such. For example:

    http://msdn2.microsoft.com/en-us/library/ms535189.aspx

    In the standards information section, it should say that the element is deprecated in HTML 4.01 and people should use <strong> to give text strong emphasis. Some people will still use it incorrectly, but better than using presentational HTML elements.

    And last but not least…

    http://msdn2.microsoft.com/en-us/library/ms535245.aspx

    "This element is an extension to HTML"? You’ve got to be kidding me. The embed element is a proprietary element that isn’t part of any official HTML standard and will cause pages to fail validation. Same goes for bgsound, marquee, nobr, etc.

    Chances are pretty good I could find other worrisome things if I keep looking. The source of the pages is just slightly better than the examples on them. Do we need to send Molly after you? 😉

  11. Will says:

    @Scott: What part of "Extension to HTML" implies to you that this is a part of a standard?  This text explicitly means that the element is not defined in a HTML standard.

    Regardless of whether or not a given tag is "standard" or even desirable, they all need to be documented.

  12. gabe says:

    http://seattlepi.nwsource.com/business/317739_msftpatent30.html

    assuming this news article is correct things could get interesting

  13. Mike says:

    The most annoying thing about MSDN (and MSDN2) library online to me is that if you click something in the tree on the left, it will always refresh the ENTIRE page.  C’mon guys…maybe you can AJAX-ify this thing.  It should only need to refresh the content area.

  14. William says:

    come on guys, take it easy, it is IE afterall: the headache of all developers.

    Now after spending days, nights, weeks of extra working to make my websites appear as right in IE as it appears in any (so called)real browser, I decide to find an answer, in name of all webdevelopers in the world.

    I’ve already sent a mail to Microsoft, which was never replyed. Not even with a automatic message or something.

    So maybe here I can find the answer to this never shutting question:

    WHY, BUT WHYYY INTERNET EXPLORER JUST DONT FOLLOW THE W3C STANDARDS? WHY??????

    man this is somthing you will find in EVERY FORUM, people just pulling their hair out because what should be a standar code needs some kind of weird alternative fix to work with the IE. And why? do they think they will enforce the standars to follow what micro$oft wants?

    man thats crazy. M$ just copyied everyone all life, why cant they just copy the standards so?

    please, someone clear that out for me!

    Firefox is open source, why dont you just copy the damn thing so youll finally have a real working product!

  15. Fduch says:

    Interesting notice: have you ever seen non-security IE update? I haven’t.

    So the bugs really aren’t fixed anymore =) Sad but true

  16. TMaster says:

    (bug report)

    In no add-ons mode, a web page allocates all available virtual memory. If I’d be using a pagefile, my system would have stopped responding.

    IE7/XP Home SP2 – severity unknown.

    To view the URL of this page, visit http://preview.tinyurl.com/yu2onz

    Please allow e-mail or web based bug reports without a sign-up procedure.

  17. TMaster says:

    (bug report)

    In no add-ons mode, a web page allocates all available virtual memory. If I’d be using a pagefile, my system would have stopped responding.

    IE7/XP Home SP2 – severity unknown.

    To view the URL of this page, visit http://preview.tinyurl.com/yu2onz

    Please allow e-mail or web based bug reports without a sign-up procedure.

  18. EricLaw [MSFT] says:

    @fduch: The bi-monthly IE patch rollups contain security fixes, as well as reliability and other fixes.  

  19. EricLaw [MSFT] says:

    @TMaster: The page in question is using Chunked Encoding to deliver at least 40+ megabytes.  

    How large is the page in question?

  20. EricLaw [MSFT] says:

    @TMaster: Given sufficient memory and CPU time, that page will render.  

    The bulk of the page (39.7 out of 39.8mb) is made up of 2,647,632 empty table cells.

  21. Fduch says:

    @EricLaw

    It’s good to haer something is being fixed. Hope some people out there are becoming less frustrated.

    I found the page where the bug that randomly happens on neowin.net since IE7 beta is 100% reproducible for me:

    When I navigate to this page:

    http://wakaba.c3.cx/sup/kareha.pl/1130723862/

    IE7 loads it and then shows error windows "Couldnt’t open …; Operation aborted". After I click OK the page is gone and all I see is "Cannot display tha webpage".

    The most amusing fact is that saving that page to disk doesn’t fix the problem which means it’s not web server issue.

    That page has very simple HTML layout.

  22. harry_richter says:

    @ EricLaw

    Eric, thanks for standing up in this forum, where you have to work in a sometimes very hostile environment.

    In your reply to Fduch you quote a KB article. If I follow that link, I get the German translation of this article (I am using a German 2K3 SVR and German IE7). I would rather prefer to get to the original, as the translations often loose precition in the process of translating. Is there a way to get the original EN-US version by default?

    Thanks!

    Harry

  23. Mr Mason,

    1- When will your MSDN2 team upgrade the coding practices in MSDN2 webpages, in the provided examples, etc?

    2- MSDN2 webpages have thousands and thousands of validation markup errors, CSS parsing errors, deprecated coding practices, incorrect/wrong coding practices, etc.. When is all that going to be addressed, tackled, roadmapped and fixed?

    3- The old MSDN used to have interactive examples. When is that going to come back?

    4- All of MSDN2 webpages use unsemantic markup code. Nested tables and deeply nested tables is frequently seen. Over-constrained tables (<table width="100%" …> and <td width="100%">) is a frequent (but wrong) practice at MSDN2. Tables used for layout is very frequent at MSDN2. When is all that going to change?

    5- Even the content of many MSDN2 webpages has errors or is confusing, misleading. Even diagrams, schemas which are supposed to help understanding have mistakes and serious errors. When is this going to be addressed?

    Gérard Talbot

  24. Fduch says:

    EricLaw [MSFT]:

    I looked at your link. At first I thout taht’s not the case because all other pages worked, but then I foun someone made javascript insertion on that particular page and it really did document.body.appendChild.

    It’s good that MS acknowledges the bug though I wonder when that article was first written…

  25. EricLaw [MSFT] says:

    @harry_richter: I’m not sure if there’s a way to do this by default, but if you add "/en" to the end, it appears to work….

    http://support.microsoft.com/kb/927917/en

    @fduch: The article in question was written sometime around IE7B3/RC1

  26. Fduch says:

    *ashamed by his spelling errors*

  27. Point Ignored says:

    "Friday, June 01, 2007 12:57 PM by EricLaw [MSFT]

    @fduch: The bi-monthly IE patch rollups contain security fixes, as well as reliability and other fixes.  

    "

    The issue is not that there is no non-security updates, in the sense of "performance improvements", but rather that there is no fix for:

    any bugs in the DOM, JS, CSS, and HTML rendering and such.

    course we all know these are being fixed in IE8, on the roadmap site:

    http://www.roadmap.internetexplorer.com/

    and in the bug tracking site, we can watch the progress:

    http://www.bugtracking.internetexplorer.com/

    oh, yeah, no not quite.

  28. I am on Internet Explorer 7 with XP pro.  frontpage 2000 won’t allow overwriting edited version of my web site past the first publishing.  error message is "Ran Out of memory while trying to allocate 0 bytes."  Cutting back on the file size to publish at a time did not help.  Google search shows Microsoft wrote and "checked out" Frontpage 2000 software without trying an overwrite publishing of the the created site.  That seriously erodes my confidence in Microsoft.  Microsoft Article ID 264932 says, "FP2000: Users Cannot Publish Changes After Installing FrontPage".  It appears the solution is to get Internet Explorer 7 to "Allow authors to upload executables".  I cannot find out how to do this on Internet Explorer 7.  The versions of IE’s change.  I never used FrontPage 2000 until now, so I never had your 3 phone call help allowance.  In fact I only called you once since 1981.  I spent a lot of time studying books and learning FrontPage 2000.  It is really bad that you don’t allow republishing web site corrections and changes.  Please help or give this to someone who can help.  You will notice from my web site verification that I have contributed significantly over the year to solving social problems.  This spans from the USSR demise in 1984 – 1989 to Preventing suicide terrorist bombings on April 22, 2007.

    Franklin Lacy northernexp@centurytel.net

    360 378-6918

  29. @EricLaw

    >The bi-monthly IE patch rollups contain security fixes, as well as reliability and other fixes.  

    That is good to hear, but the general impression is that non-security bugs get much less attention and are generally not fixed. Any chances that the disappearing caret UI bug gets fixed?

    http://forums.microsoft.com/MSDN/Search/Search.aspx?words=Viktor78&searchKey=&lcid=1033&searchscope=forumscope&siteid=1&ForumID=924&ForumGroupID=-1

    We all know that IE desperately needs an official bug tracking system like the one on Connect that was unfortunately closed after RTM. But this is not necessarily the fault of the IE team but seems more like an inappropriate and outdated policy at MSFT that only encourages feedback and bug reporting for Beta products.

  30. cdude says:

    IE doesn’t need fixing, it’s the management.

  31. Stephen says:

    Please don’t listen to the calls for dropping the "obsolete" msdn ASP/DHTML/HTML/js/IIS pages.  These are some of the BEST documents available.  They can be very difficult to find sometimes.  It would be great if the current docs included examples.  They used to make sample apps and examples.  Now, just sample apps.  Who wants to spend 2 hrs digging for 3 lines of code?

    And some employers think a web app written 5 yrs ago only needs maintenance and not a rewrite to .Net.  Go figure.

  32. Fduch says:

    I’m getting this error sometimes:

    SysFader: iexplore.exe – Application Error: The instruction at "0x3019fd88" referenced memory at "0x0bbcf000". The memory could not be "written".

    and if kills IE7.

    How to fix it? I couldn’t find any useful info though I see I’m not the only one getting this error.

  33. ST says:

    Last month:

    GET /response.asp?MT=test&srch=3&prov=gogl&utf8 HTTP/1.1

    Host: auto.search.msn.com

    response

    HTTP/1.1 302 Moved Temporarily

    Location: http://www.google.com/search?q=test

    Today:

    GET /response.asp?MT=foo&srch=3&prov=gogl&utf8 HTTP/1.1

    Host: auto.search.msn.com

    response

    HTTP/1.1 302 Moved Temporarily

    Location: http://search.live.com/results.aspx?srch=104&FORM=AS4&q=test

    http://blogs.msdn.com/ie/archive/2004/10/08/240062.aspx and

    http://www.google.com/google.reg does not work.

    note: shdoclc.dll include "auto.search.msn.com/response.asp".

  34. Mike says:

    looks like Safari 3 (beta) installs and runs just fine on Windows 2000.  too bad you IE devs can’t get a version of 7 on win2k.  Oh well, i guess the apple (mozilla/opera) devs are just better.  

  35. d. braun says:

    A brilliant idea for a new feature in the next version of Internet Explorer –

    a built- in advertisement blocker that will hide any recognizable commercial banner or text ad INCLUDING ADSENSE.

    Sure it will harm google, hard (:D).

    but its a legitimate feature that will not come on default (although will be easly enabled by those who wish it.)

  36. @ d.braun

    I have developed a replacement for the Google Toolbar that includes a feature to block Google AdSense as well 😉 http://www.quero.at/