The danger of a better Internet Explorer

So the first beta of Internet Explorer 7 has been released. The reports from those who have tried it point towards the current version seems pretty much what I was anticipating. Security fixes, tabbed browsing, and not much more, i.e. little to no improvement of its support for HTML and CSS.

But according to a post by Chris Wilson at the IEBlog this is just the beginning, and the next beta will have much, much better support for HTML and CSS. In fact, nearly all of the most annoying CSS bugs and unimplemented features are listed as being fixed in beta 2. Great news! Or is it?

(By the way, I thought beta software was supposed to be feature complete – shouldn’t this then have been called IE7 alpha?).

A properly working Internet Explorer browser will make the lives of web developers a lot easier and less stressful once it’s in widespread use and IE6 can be safely ignored. How long that will take remains to be seen. I’m guessing four or five years. So that is something to look forward to.

Still, I am slightly worried about the web as a whole. If the current beta of IE7 had been what will eventually ship as IE7, it would have made it easier for other browsers to continue grabbing market share and force even the most ignorant web developers to strive towards building browser independent websites.

But with the greatly increased support for web standards that looks like it will be part of the shipping IE7, I’m worried that IE will take back some of what it has lost over the last couple of years. Not so much in the general public’s image of the browser as in the minds of developers. Many developers and geeks are early adopters and have helped push Firefox to where it is now. Some may switch back to IE now that it will hold up much better to the competition.

So what? A good, standards compliant browser is a good browser whether it is made by Microsoft or not. Oh yes, it is, and well done to Microsoft for finally getting with the program. Thank you! (I mean that.)

The threat does not come from Microsoft as much as from the many, many web developers hacks who do not believe there is a place on the web for people like me that are stupid enough to use anything but Internet Explorer on a computer running Windows.

If IE regains all of its lost market share, these people will once again be tempted to build public websites that rely on proprietary features that only work in IE, and only on Windows. I’m not talking CSS or visual effects here. Think sites that require ActiveX, VBScript or Windows-specific Java applets to work. And that would take us back to the end of last century, before the web standards movement got going.

When you look at it that way there is a chance that it would have been better for the overall health of the web if IE7 had not featured improved web standards support. Think I’m wrong? I hope I am.

Posted on August 14, 2005 in Browsers, Web Standards


  1. I agree with you, IE7 gaining back market share would be a bad thing. Things got so bad in the first place simply because IE had such a monopoly on the browsing public.

    IE7 having much more standards support is a good thing, but that’s one good thing in a sea of horrible things. I work on computers on a regular basis and the most frequent problem I help solve, by far, is spyware and malware that got on the computer through IE and its myriad security holes. Despite claims to the contrary, I see no reason to doubt that this will change in the future versions of the browser.

    I would like to see the browser market split more or less evenly between a few major players. Say roughly 33% for IE, Firefox and Safari (here’s hoping Apple gets that kind of market share too!) each, with Opera taking up the slack. That kind of diversity would do wonders for the Web as a whole and get most web developers on the right track: standards, not proprietary technologies. It would also drive feature competition and security across all the browsers; nothing but good things for both developers and end users.

    Sadly, that day seems very, very far away. But here’s hoping.

  2. Here’s a question, then - what is the difference between a developer using a proprietary IE-only effect, and a developer using a currently unsupported CSS behaviour like li:hover at the moment?

    Many designers are making use of CSS2/3 effects to enhance compliant-browser users’ visits or otherwise deliver a different experience to IE users; how is that any different to using an IE-only feature?

    Yes, one is “standards” (although how can it be considered a standard if only ~10% of browsers support it? doesn’t sound very ‘standard’ to me…) and the other is proprietary, but in terms of actual effect there is no difference between a proprietary bit of CSS and a filter like html>body.

  3. it is expected that browsers will migrate towards standards compliance - there is no roadmap for adoption of anothers propriatory systems

    so although a mainly unsupported spec feature now is marginally useful - it is future proof as its support will only increase

    so long as you dont use such features now as necessary for complete functionality your ok as the browsing public will increase in number who get to see your mastery in all its glory

    a platform or browser specific innovation will only increase in viewing public if that one browser increases its market share

  4. August 14, 2005 by Roger Johansson (Author comment)

    Matthew: The problem is when developers rely on proprietary features for the site to work at all. VBScript and ActiveX are two very common examples that exclude users of other browsers. Java applets that only work in IE/Win is another example.

    Creating visual effects or otherwise enhancing the experience for users of a certain browser is not dangerous as long as it doesn’t prevent the site from working in other browsers.

  5. Is that a coincidence?

  6. (Francois: I doubt it is just a coincidence!! )

    Roger: I think your fears are well-placed on past events and experience with trends in web development, and also the fickle nature of the general public. But I don’t see the Web Standards arena vanishing overnight - for all the reasons it exists now…IE7 (and future versions) still have to meet the challenge and because of Mozilla’s opensource community - Firefox et al are able to respond far quicker to user requirements and security fixes, and this is what will keep the playing field more than a ‘one-horse race’. I think it’s a good think if a browser that will be used by approximately 80% of all web-users does approach a better level of web standards support…but I don’t think it will mean the end of web standards (and other browsers) as we know it.

  7. It’s important that there remains an ongoing development (means fixes and new features) once IE7 has been shipped. In other words, if IE7 was released and then there is standstill for another five years, things will never become better.

  8. I have similar opinion on the subject. I wish IE would just die. That, unfortunately, will not happen. As for the switchers, I think that you’re exaggerating a little bit. Yes, there will be many people who will install IE7, but there will not be such many switchers. Other browsers give you so many more features, that IE will never have.

    Plus, there is very strong ongoing movement to the standards compliant web. It’s too big now to turn back to propertiary features.

  9. August 14, 2005 by Ramon

    We all wish that IE would just die but that won’t happen.

    But actually i think IE7 will not have that much influence. It wil only run on vista and XP SP2. Not even 98, 2000 or XP SP1. If more webdevelopers start using standard compliant code, then Firefox will be just as good als IE7 . This means (al least) everyone who is not using Vista or XP SP2 wil use Firefox, because the sites are more standard compliant and IE6 wil not work anymore.

    As you see, there are a lot of sides to this and I don’t think we can predict how the browsermarketshare will be in 5 years.

  10. Here’s the difference between now and then:

    In August 2001 when IE 6 was released, they didn’t believe it was possible for an alternative browser to take any real market share from them anymore. This view continued even as late as the end of last year.

    Now, they know better than to assume their dominance can be sustained without some resemblance of quality.

  11. What I’m not looking forward to is yet another level of complexity. I use conditional includes to fix CSS problems with IE6. When IE7 is out, these “fixes” will most likely break the site layout for users who have upgraded. Now I’m forced to sniff for version X,Y and Z of a particular browser…

    Also Roger, I suspect that your estimate of 4-5 years of IE6 still being a major factor is low. Given that IE7 requires you to have SP2 installed and many, many people have been scared off from installing it because of the nightmare stories. This includes me, thus far I have declined to install it on my Windows box because I have too much important data there to risk it becoming inaccessible. But of course by then, Vista will be out and strong, right? I’m not sure about that either. I know plenty of people that are still running (cough) 98.

  12. I mostly agree with this post, but I also think that the UI features themselves will be a big problem for web developers. It used to be easy to point to tabs and the RSS icon and show Joe User that other browsers offer more than Internet Explorer does. Talking about standards has always been pretty pointless because most people simply don’t care, but they do care about practical functionality and ease of use. Now that IE7 will have some form of these UI features they were lacking, it will be harder to convince Joe User of the benefits of switching.

  13. Douglas: You can bypass those ‘IE versions later than IE 6’ hack problems in the css declaration…the [if lte IE 6] solution (which is better explained here: Solution)

    This fix works well and should continue to do so. Best of luck. :)

  14. I too worry about the implications of IE7 for web developers. Far from making things easier, I believe that many of us will now have to code for FIVE versions of IE, which is ridiculous!

    Once it has a reasonable take up, I’ll probably stop coding for IE5.0/Win, but still…

  15. That’s a good point Matt, thanks. Assuming IE7 doesn’t need any hacking of it’s own.

  16. That’s a good point Matt, thanks. Assuming IE7 doesn’t need any hacking of its own. Per some other comments here, I wish to avoid maintaining conditional includes for IE5, 5.5, 6, 7…I can live with one for <= 6.

  17. It seems unlikely to me that with so many designers/developers getting on the W3C standards bandwagon, there could ever be a return to the “bad old days” of the web. True, there are many sites out there still using layout tables, ActiveX, and VBScript; but how many of those have been created in the last year or two? Am I just being naïvely optimistic by thinking that designers and developers have largely come around to the idea of a standards-based web?

    And, speaking as someone who generally doesn’t make websites for public consumption (mostly intranet/client only sites) isn’t it time to stop worrying about IE 5? The only people who can’t install IE 6 are running Windows 95; chronologically speaking, this is the same as worrying about Netscape 4.

  18. August 15, 2005 by Felix Winkelnkemper

    I can fully understand all of your comments about the new internet explorer 7, but I think there are some points that are a little unfair.

    I do not want to talk about any kind of bundling the Internet Explorer with Windows. I also do not want to talk about security-issues which are of course a serious issue.

    Let’s talk about CSS and standards-compliance. When I read your comments, I get this picture: Before Internet Explorer, everything was fine, every browser was standard-compliant. Than Internet Explorer came and everything was gone. It had no improvements but IE-only-features. Amen!

    But is this true? Remember! Netscape 3 did not understand CSS at all, Internet Explorer 3 did a first try. The CSS-Interpretation of Netscape 4 was awfull, IE’s Interpretation was quite okay and was improved until Internet Explorer 6. When Internet Explorer 6 was published (and bundled with Windows XP), Netscape still published versions of Netscape 4 (Netscape 6 was published in April 2000). As you know, there was no CSS-improvement at all. So, if we want to be fair, we cannot blame Microsoft for everything. If there had been no Internet Explorer, it is very likely, that a lot of us would not have known anything about CSS at all, we would still code the NS4-way.

    Now, things have changed, now IE is the old browser, and Mozilla (Firefox, Netscape) is the shining star. If IE gets improved, Firefox and Mozilla have to think about new improvements to stay on top - if they do not, they repeat Netscapes failure of the 90s.

    Enough “if” and “when” for now. We all will be part of it - could be an interesting time.

  19. I disagree that an increase in IE’s marketshare will mean a return to using IE-only features, because in IE7 they are implementing only existing web standards.

    Remember, Netscape too introduced features that only worked in its browser.

    The problem is, as others point out, that IE6 will remain around for years because people don’t wish to upgrade to XP.

  20. August 15, 2005 by Roger Johansson (Author comment)

    IE-only features don’t necessarily have to be new to cause problems.

    What worries me is that an increase in IE/Win’s marketshare may tempt some developers to increase their reliance on existing IE-only technologies. And I am not talking about W3C technologies, but proprietary Microsoft technologies that other browser vendors simply cannot support.

    Some of you mention that so many web professionals “get” web standards now that going back to the old ways is unlikely. I hope you are right, but the majority of web developers I run into are more or less unaware. Most do not read web standards blogs.

    New sites are still being built that only work properly in IE/Win. I see it all the time. Public sector, privately held companies, small, big, everywhere.

    My estimate of 4-5 years for IE7 to be the dominant IE version is probably too optimistic. I wasn’t aware of it only being available for Microsoft’s newer operating systems. Being a Mac user I pay very little attention to what Microsoft is up to in other areas than IE.

  21. If Apple can build on the success of the ipod and encourage these customers to buy Mac hardware - hopefully there may be a time when developers have no choice but to cater for Mac users (and as a result non-IE/Win users).

    Unless Microsoft bring out a new IE for mac (noooo!).

  22. I’m a bit afraid that the MS jump on the ‘webstandards’ bandwagon amounts more to PR than anything else. Of course they’ll support more W3C specs (HTML 4.01, CSS 2.1), and fix some bugs. Would be about freaking time, too. The WASP will be happy. Maybe it, along with some improvements in the UI, will help preventing some people switching to some other browser. Stats will probably stabilise at 65-70% IE.

    But I’m not convinced that will amount to much, in terms of standard compliant browser. As long as that browser is affected by ‘hasLayout’, it will be a disaster.

    The one thing that MS does not talk about much is the integration between IE and XAML in their next OS, and all the frameworks that come with it. And this worries me a lot more, in terms of ‘sites build for IE only’. It is akin to ‘you can use those roads, as long as you use my car’.

    I don’t trust MS, even if I have respect for what individual devs try to do. I’m really afraid that will say more of real dirty tricks in the next five years, to lock people in the use of some particular software packages.

  23. I think you missed the point. I don’t care if IE7 is a market leader or not. Nor any other browser. What I care is does it properly supports standards. If it does then it’s fine. Really fine. I can continue using Firefox and if average joe wants to use IE7 then I’ll let him.

    IE7 being market leader = not bad. IE7 not being continually updated and fixed (as was the case with IE6) = very bad.

  24. Regarding standards support: Yes, it’s important to remember that IE was the first major browser to implement CSS even slightly correctly. They dropped the ball with 6.0, but IE 5.0 was a godsend to a lot of developers, and IE 5/Mac was, for it’s time, the best available browser, as far as standards support went.

    Regarding proprietary features: They may make life difficult for developers, but they also push the web forward. Without IE’s ActiveX XML support, we would not have XMLHTTPRequest, which is still not a “standard”.

    Microsoft has pushed the web forward by providing non-standard features which are undeniably useful to web developers. The W3C, being a more academic institution, often fails to see what’s really going on in everyday web development, and hands us things like the css float model.

    As a web applications developer, I’ve noticed that most of MS’s proprietary features are geared towards people using the browser as an app environment, usually one that is behind-the-firewall, with pre-determined browser requirements. I think this breeds a lot of animosity from the standards advocates, who are for the most part developing sites for the public web, which are in general less complex and very often not much more than “brochureware”. They have no use for, and often are completely unaware of, some of IE’s powerful proprietary extensions.

    Microsoft’s proprietary features like XML data islands, CSS javascript expressions, CSS behaviors, and fine-grained event handling (onreadystatechange, etc) are not much use to content centered sites such as blogs and corporate websites, as opposed to data-centric, transactional websites. What’s really odd is that MS has so many tools revolving around data views and grids, but never implemented “overflow:scroll” for TBODY.

    I still use Firefox for everyday browsing, and think Mozilla’s baseline architecture and extensibility (both in the UI shell and it’s ECMAscript implementation) are far superior to IE, mind you.

  25. You certainly make a good point. But to say MS is in any way faultless in the matter would be wrong…

    Let’s face it; fair play isn’t really Microsoft’s trademark. Microsoft was built on unfair business practises. MS wouldn’t have been where it is today, if it hadn’t been for the fact that they stole the source code from CP/M in 1980.

    And of course, we all remember what happened to the Firefox-team, about five years ago, back when it was called Netscape Navigator.

    And to Dave’s comment:

    If Apple can build on the success of the ipod and encourage these customers to buy Mac hardware - hopefully there may be a time when developers have no choice but to cater for Mac users (and as a result non-IE/Win users).

    There is another side of that coin as well.. Firstly, Apple struggles with the same thing as Firefox; the consumer’s fright of everything new.

    Secondly, if Apple were to gain the bigger piece of the computer cake; I don’t think that wouldn’t be that good of a good thing at all, really… We’d have the same situation as on the Windows-platform… Viruses, 3rd-party hardware that’s not fully compatable with the OS or other software etc…

    Actually; I think today’s situation is as good as it gets, so I’m jus going to enjoy it as long as I can…

  26. On a bit of a tangent here, with the release of OS X 10.4 Tiger, there isn’t a default install of IE for the Mac. So far I haven’t really seen anything from Microsoft about developing a version of IE7 for OS X, and the last comment got me thinking….

    If Microsoft doesn’t develop an OS X IE7, they will have a lot harder time getting the majority of the browser market share once again.

    Pardon my somewhat obvious realization, but this is something that just came to mind sadly.

  27. IE7 has been re-branded as ‘Windows Internet Explorer 7’ in the release of it’s new logo. Notice that it’s Windows, instead of Microsoft.

    In my opinion, this is a clear indication that Microsoft does not intend on developing Internet Explorer on other Operating Systems. This was foreshadowed when Microsoft dumped its Macintosh Internet Explorer team.

    So in fact… with the popularity of Linux and Macs rising in recent years, more people will be using KHTML (Safari) based, or Mozilla based products (Ephiphany or Firefox).

    In any case, IE7 seems to be focused on fixing the blatantly wrong bugs with IE.. which is a good thing. With the addition of full PNG alpha channel support, and CSS bug fixes, IE7 doesn’t look all THAT bad when it comes to supporting standards. Sure, it doesn’t support them all, but it’ll be enough to give decent support of CSS1, I hope. That should be enough for developers. The features of CSS2 and CSS3 can be left to to those who are smart enough to download another browser. :)

  28. Is david gee the only one who gets it? Developers use the proprietary features of IE because it enables them to create application-like features (showModalDialog, mshtml, etc) that no other browser offers.

    Developers interested in using this functionality aren’t as concerned about market share and will continue to use them despite IE’s market share at any given time.

    Honestly Roger, your argument applies to every single browser on the market. If any of them gain the ubiquitous majority, we’re in for a lot of proprietary code. Each new browser version calls for new browser hacks and code detection. Why is IE different?

    I for one welcome any browser that improves on what it did before. No browser is perfect and I don’t suspect I’ll find one that is in the next 10 years. :)

  29. August 15, 2005 by Roger Johansson (Author comment)

    First of all, let me restate that I’m not really arguing that better standards support in IE7 is a bad thing. I think it’s great. I’m just putting forward some thoughts and asking what you think. And please keep the discussion going :-).


    As a web applications developer, I’ve noticed that most of MS’s proprietary features are geared towards people using the browser as an app environment, usually one that is behind-the-firewall, with pre-determined browser requirements.

    Sure, and that may be almost acceptable in such environments (I say almost because I detest the attitude held by companies that force their employees to use a specific browser).

    The major problems occur when these technologies are used on public websites, making it difficult or even impossible for users of other browsers to use the site.

    Jonathan: Yes, it applies to all browsers. IE is different because it is the dominating browser right now and because of the attitude held by many Microsoft-loving developers. Maybe it’s just a Swedish thing, but there are legions of such developers here.

  30. But what is the minimum percentage of market share that will make developers feel comfortable using features that exclude some web viewers? My feeling is, the developers who are truly devoted to building sites that break if you aren’t using IE are going to feel justified (in their minds) continuing to do so as long as IE enjoys a comfortable lead. (What is it now, high 80s? How much lower can we expect it to drop, really, considering most web users aren’t invested in this war, and just click the “Internet” link on their desktops?)

    Reliable standards across the top browsers has to be the more important goal. If IE7 is going to join that group, that has to be seen as a win. Fringe developers who don’t care about excluding viewers won’t be swayed by a few percentage points either way.

  31. August 16, 2005 by Niklas Blomdalen

    Nice article =)

  32. I don’t think we’ll have to worry about IE7 dominating the web in the same manner IE4+ did. The dialogue of open-source software as well as standards compliancy is really beginning to take place between developers and regular web surfers. It will only get better.

    When IE7 finally adopts CSS2, every other browser (led by Safari, I’m willing to bet) will be embracing CSS3, SVG and so on. Microsoft simply aren’t able to move as quickly on browser development as the mozilla organisation. That has become glaringly apparent in the recent two years.

  33. An awful lot of education has occured in the past few years regarding standards etc. However, in my experience, the most likely people to create more sites with Visual Basic, Applets, Active X etc. are your more typical application developers, as opposed to web developers/designers - mostly out of habit and inexperience on the web.

    I think the future for this does look good for standards though. Most of the bad code I’ve seen written in the past few years has come from Visual Studio and ASP.NET. I’ve seen some shocking, horrible code come out of there. The programmers using loved it from the coding end, but for me on the front end side, it was a nightmare. The good news is that ASP.NET Whidbey (the next version of ASP.NET) will be XHTML compliant. As long as its outputting XHTML (or HTML 4 - as long as its valid), things should be much better in the future. At any rate this should negate any IE-specific code, although if ASP.NET continues to serve different HTML to different browsers - then some of the IE7-related fears voiced here could still cause us problems.

  34. James, ASP.NET serving different HTML isn’t necessarily a bad thing. We serve different CSS. We serve different JavaScript. It’s all in an attempt to make our content accessible to the largest audience. And ASP.NET is developed in such a way that a developer doesn’t need to know HTML. See, as much as some of you like to hate Microsoft, they try and make life easier for developers. Look at how much effort they put into documentation (I still use MSDN over the gecko DOM reference, blech). The largest side effect is platform lock-in. That’s not a bad trade-off, is it? :)


  36. I think the above comments have missed one vital part of the firefox maket share issue which is that ‘normal’ (non early-adopter) users will be put off by the number of large company websites where firefox ‘breaks’ - actually the site is badly coded, but average punter does not know that and naturally blames firefox.

    If ie has better standards support then hopefully more sites will be built to standards ergo firefox is a more viable alternative.

    I accept the ‘why would they switch if ie works’ argument, but I see it as mirrored by a ‘how can they switch when sites don’t work’ argument, so I am not sure firefox take-up is helped by continued poor standards support in ie?

  37. As someone quite new to the full potential of CSS and standards this is interesting reading. From what I know and have read Roger might well turn out to be correct and IE7 could holt FireFox’s - until now - growing user base, which would not be good.

    IE’s lack of proper CSS implementation is indeed frustrating, and the sooner we have a major competing browser to keep Microsoft well and truly on their toes the better.

    I dream of the day when a website designed exactly to W3C specifications works perfectly across browsers.

    One day - maybe!

  38. Jonathon: I probably should’ve elaborated on what I meant by different HTML. Controls would be outputted as divs for IE, and tables for anything else - including Firefox - which is kind of self-defeating in context of advances in CSS. Other than that I don’t necessarily have a problem in serving different HTML, depending of course on how its done.

  39. August 17, 2005 by Felix Winkelnkemper

    What do you mean by “Firefox [..] is kind of self-defeating in context of advances in CSS”?

  40. Felix - I meant what’s the point having CSS layouts (i.e., non-tabular layouts) for a site when it’s serving tables up to non-IE browsers (which then gives you partial tabular layouts).

  41. At least it’s a start, like you said in the post that there will be increasingly better CSS support but this is not all Internet Explorer Lacks, I recommend people to email and suggest the CSS comments rather then saying how bad IE is, contact the developers and tell them you never know you might get a nicer browser, for everybody to code to.

  42. In the “IE’s best” / “no! Firefox is best!” type discussion going on above, I think a couple of salient points have been missed. I’ve only skimmed some of the comments so don’t burn me if i’ve missed something!

    One, it’s not just about which browser is best, but the fact that websites are designed that work only in IE. Even today, corporate websites exist that have messages on them saying “You are not using IE5+, please upgrade your browser”.

    This isn’t IE’s fault, per se, but a result of developers a.) not following standards, and b.) using proprietary IE elements that break the site in other browsers.

    On the other hand, if a developer uses standards it works pretty well across most browsers, but can break horribly in IE. This isn’t because Firefox/Opera/Safari etc are using proprietary methods, but because they use standards! How can it be a bad thing to standardise rendering across browsers?

    MSIE could cope with standards and they could still design their own clever, ground-breaking proprietary add-ons. Why must it be an either/or argument? Fact is, IE is shot so full of bugs it’s become a laughing stock for very good reason.

    The other point is accessibility. Designing standards compliant sites that work in IE is tricky, as you have a LOT of IE CSS and rendering bugs. If they fix them, great, but at the moment they ain’t fixed. This causes accessibility problems, which in many countries is now a legal problem, not just a web developers problem.

  43. August 17, 2005 by Felix Winkelnkemper

    I still do not understand, why you should serve IE with divs and others with tables? I could understand the other way round. But I think both is a misunderstanding. In a perfect world, you would not have to think about browsers at all - all you’d have to care about is valid code.

    Lee is perfectly right. Almost any kind of discussion about browsers is useless and gets more useless, when IE improves.

    Lee talked about accessibility and mentioned, that it is not possible to craeate accessible websites in IE. I do know what you mean, but have a look at screenreaders and “talking browsers” like the IBM homepagereader. They all depent on MSIE. Up today there is no working technology based on Firefox nor on Opera.

  44. August 17, 2005 by bodaniel

    As long as the massive amount of Microsoft-loving developers used to one-click-away-from-a-great-web-application tools stay faithful, Microsoft could do whatever they want with future versions of IE.

    As long as the developers can get yet another Microsoft diploma (a crash course in dot-this and dot-that) instead of learning HTML, CSS etc and their boss knows that the latest web app works in IE (the only web browser?), they stay happy.

    The only thing you can do is hope that you never have to be part of a development team full of these Microsoft-loving developers.

  45. August 17, 2005 by ProfSnape

    Roger, have you tested IE7?

    If not, why not get a copy to test before you write a article about “rumors” or something that other have written?

    And why not discuss what standard is, is it something that a group have writen down but few follows or is it what the majority of usera ar using?

  46. August 17, 2005 by Roger Johansson (Author comment)

    ProfSnape: No, I haven’t personally tested IE7. What Chris Wilson wrote at the IEBlog (see the link in my post) is good enough background for writing this. And I wouldn’t call that rumours.

    A standard in this context is obviously a recommendation published by the W3C, not something created by a single company.

  47. August 17, 2005 by ProfSnape

    I wrote “rumors” in “” because a rumors is just something you heer or read but don’t have tryed och seen yourself

    You can also define a standard that most user is using, often called “defacto standard”.

    So if you want your page to be usefull to most people use the “(defacto) standard” that these user use.

    The whole “Internet” is build on “defacto standard”.

  48. August 17, 2005 by Roger Johansson (Author comment)

    Yes, you can define “standard” that way. However, that is not how web standards are defined. Web standards are not about “most” people, they are about all people.

  49. Ok, here is the problem from your side of the fence to my side of the fence.

    Jørgen Arnor Gårdsø Lom writes above

    MS wouldn’t have been where it is today, if it hadn’t been for the fact that they stole the source code from CP/M in 1980

    Which you link to an article, that clearly states that they bought the OS for 50,000 in 1980 (which would be the equiv of $126,982.94 today). I wish people would pay me that much to rip me off.

    Its statemenst like that, that are all to common from your side of the fence, that makes people on my side of the fence want to have nothing to do with anything that you stand for.

  50. Sorry about the triple post - I got no indication anything happened, the browser did not appear to submit - just appeared to do nothing.

  51. August 17, 2005 by Roger Johansson (Author comment)

    Zach: Sorry about the commenting trouble. I’ve noticed that posting comments can be very slow sometimes. The sysadmin is aware of it and is trying to find out what is causing it.

    About different sides of the fence: This post is not intended as Microsoft-bashing! I am no fan of Microsoft’s software or the way they do business, but I’d rather not be put in the Microsoft-hater camp.

  52. August 17, 2005 by ProfSnape

    “Web standards are not about “most” people, they are about all people.”

    In that case we can only use standars that all browsers can use, so if only one or two off all big browser can’t use something then we can’t use that in our webpages?

    If one browser can’t show it then all user can’t see it the it is not a standard.

    Simple logic. :)

  53. August 17, 2005 by Roger Johansson (Author comment)

    ProfSnape: Unless you provide a fallback, yes. But providing fallbacks and using graceful degradation is what sensible web development is about.

    It obviously also depends on what kind of feature we’re talking about. If it’s a visual effect - like the CSS property text-shadow - that is only supported by a subset of browsers, it’s safe to use. If the feature is related to functionality you should provide a fallback for browsers that do not support it. JavaScript (ECMAScript actually, since we’re discussing web standards) is a good example of that.

  54. Felix - I don’t understand it either! That’s my point, it’s silly, but ASP .NET controls will automatically serve different browsers different HTML at the moment. It’s Microsoft’s ideas, not mine. I totally agree with you re: valid code.

  55. This is a bit offtopic, but regarding ASP.Net serving different HTML to different browsers: This is easily configurable by editing your machine.config file. A simple google search (here’s one result) will show you how. ASP.Net 1.0 breaks down into two camps - “uplevel” and “downlevel” browsers. As it was released (AFAIR) before Firefox, Safari, et al, it considers pretty much everything except IE5.0+ a “downlevel” browser, unless you tell it otherwise. From what I’ve seen of .Net 2.0, it considers Firefox an “uplevel” browser by default.

    Roger, I agree with what you have to say about businesses forcing their employees to use a certain browser. However, there’s a difference between forcing your employees to use a certain browser and forcing your employees to use standard equipment to access work-related functions. That being said, I create all of my prototypes to work in IE and Gecko-based browsers. Unfortunately, when my prototypes get handed over to the.Net development teams, who are very often outside of the country, it gets more difficult to ensure compatibility.

    And agreed, .Net developers are by and large the sloppiest, most Web-unaware coders I’ve ever come across. 9 times out of 10 they don’t know a DIV from a SPAN, and this is very much the fault of Microsoft and .Net, which abstracts HTML tags completely away from the developer (for example, in .Net, a SPAN is an asp:label, and a DIV is an asp:block).

    Don’t get me wrong, I’m a huge fan of standards-compliant browsers, I just find it quite funny that all the “standards advocates” are wetting their pants over “AJAX”, which would not exist if Microsoft didn’t implement proprietary technologies in their browser. XMLHttpRequest was built into Mozilla as an answer to Microsoft’s ActiveX XML implementation, and was later added to other browsers such as Safari and Opera. To this day, it is still not a “standard”.

  56. Erm, release candidates (RC) are supposed to be feature complete, not betas (and certainly not alphas, of course).


  57. August 18, 2005 by Roger Johansson (Author comment)

    david gee: Good point about AJAX (I’m one of the standards advocates not wetting their pants over it). Of course adding new features is OK. Developers not using those features in a responsible way is not.

    Slapo: Depends on who you ask ;-). Wikipedia thinks beta software should be feature complete: Development stage.

  58. August 18, 2005 by Felix Winkelnkemper

    Have you ever thought about, why the W3C does NOT talk about standards but about recommendations?

    If “A standard is, what all people use” was right, there would be no use in standards at all because it is almost impossible that everyone uses them!! I would say, that a W3C recommendation is something every webdesigner SHOULD CARE ABOUT, no more, no less.

    Just two additional remarks:

    1. Do we have fences here? I do not see any!

    2. Stealing is mostly the wrong term. In a German Forum, I just read, that Firefox has stolen tabbed-browsing from Opera. The same people that wanted to have tabbed-browsing in Internet Explorer for years now say, that Microsoft has stolen the idea; of course Microsoft has not stolen CP/M. They have bought an Operating System called QDOS, a system nobody of us would have ever heard of.

  59. August 18, 2005 by Roger Johansson (Author comment)

    I think the discussion has drifted a bit off-topic here.

    Let’s get back to whether better support for web standards (as defined by the Web Standards Project) in IE7 will increase its market share, and if that will tempt more developers to use technologies that only work in Internet Explorer for Windows without providing any fallback for other browsers.

  60. I’m just starting out as a developer, and standards are one of the things I want to be able to pride myself on knowing and using, and so the only thing I have to add to this discussion is that the hardest part for me has been learning hacks and techniques to make what should be simple CSS as compatible as possible, and one of the main reasons for my having to do that is in fact IE, and so despite the consequences (or the delay) i’m happy that IE has made the leap (should it ever have really been a leap?).

  61. Roger: I agree with what you said above about responsible development practices. I do still wonder how the web will evolve to embrace new technologies if using non-standard technologies is taboo. I guess in a lot of ways I just don’t trust pie-in-the-sky academic institutions to come up with new, powerful real world solutions.

    As for the future of IE7, I’m not sure if Microsoft is really vesting too much on it. From all of their actions, it seems to me that they’re doing what they can to put an “acceptable” browser out there, that conforms to most standards, so that in the future a default windows installation will be able to browse the internet. If they were really striving to make IE7 the “king of browsers”, I think they’d be doing more to ensure backwards compatibility with older versions of Windows, and reviving IE/Mac. I think Microsoft actually views the days of the browser as coming to an end, and is hoping that XAML will take off and become the de-facto standard for coding internet-enabled applications. This makes sense, in the sense that XAML is designed from the ground up to do this, whereas HTML and all its related technologies have always been a hack.

    So, in 5 years time, we’ll probably be talking about how XAML doesn’t conform to the W3C XUL standards :)

  62. If IE regains all of its lost market share, these people will once again be tempted to build public websites that rely on proprietary features that only work in IE, and only on Windows.

    I hate to tell you this but most of these people have never left or changed their tactics. They are still raking in tons of money.

  63. Zach: So you would describe MS’ business moral as good? Take a look at the ongoing conflict with google, for instance…

    When you have near-monopoly on a market, you can’t simply stay on top by ‘killing off your competition’… I’m not claiming to be objective in the matter - I’m just saying that Microsoft wouldn’t have been where they are today if they’d stuck to the rules - their products simply aren’t that good…

    I’m not quite sure which fence you referr to - if it’s the mac-fence, the validation fence or the Firefox fence - I would say I’ve just crossed all of those fences, and I’m fully capable of remembering two year back, when I was a proud Windows-user, who designed pages specifically for IE - and I will say the grass IS greener on this side!

    Example: I’ve added a feel of safari to the searchbar on… Now - find ONE piece of IE that would be worth exporting to other platforms (and please don’t give me any “IE invented the web”-shit)…

  64. Isn’t it likely that the IE program managers are completely aware of this? They just need to be good enough to prevent people from switching from the default and a high enough market share to not make it worthwhile spending effort on other browsers.

  65. James: there’s a fix for which makes it render sensible HTML for Firefox, Opera and Safari. It doesn’t tackle every browser that ought to have up-to-date HTML, but it certainly improves the situation.

  66. November 11, 2005 by Micheal Hodges

    I may be a little late,but I wanted to add a few things primarily concerning why all this work and research is done.It’s for the end result, which also means the end user.If you (you meaning everybody) don’t like the standard or proprietary, then innovate!!As a developer, you should be trying to stay ahead of the game, not waiting for someone to do it for you.I work with the end users everyday (internet, intranet, corporate networks), and they don’t notice the differences between one browser or another.What they do notice is whether or not the content is easy to use, easily accessible, easily changed, etc. While this doesn’t answer the questions laid out in this dicussion, it should make all of you choosing sides realize that you’re trying to make the end user conform to what you think is right.As far as the bashing of corporate entities, claiming this company is evil, or that they are making you conform to their “standards”, that’s a bunch of hooey.So what if MS has a monopoly.What they also have is a way to keep the computer/developer/internet/end user business constantly flowing.For every person who chooses not to use Windows( and make bashes at MS)there are 10 people who have to use it, because they are not computer literate.MS has provided a means for everyone to enjoy the things they can do with a computer.Catering to such a broad spectrum of users means making things that EVERYONE can use.Not everyone is a developer, and I always thought the developers were trying to help the end user experience.As far as the market, stop worrying about power battles, and YOU be the one to innovate YOUR design so that YOU can stay ahead of the game, and help all the little people regardless of what browsaer they choose to use. I guess my point is stop whining and start working.Someone out there has the smarts to make that code work with this , and this code work with that.People as a whole need to change their thinking before throwing around comments about who is best and who is worst.The blame game answers less questions than this string of nonsense I’m typing…….By the way, this has been a great discussion!!

  67. November 11, 2005 by Micheal Hodges

    Left info out on the last post.What I mean by constantly flowing is this: Making sure you still have a job at the end of the day.

Comments are disabled for this post (read why), but if you have spotted an error or have additional info that you think should be in this post, feel free to contact me.