Follow Slashdot stories on Twitter


Forgot your password?

IE7 Compatibility a Developer Nightmare 416

yavori writes "Internet Explorer 7 has kicked in at last on all MS Windows OS running PCs because of the fact M$ decided to force it's users to migrate through update. In fact this has started a IE7 Web Developers Nightmare. The article actually explains that most of the small company B2C sites may just fall from grace because of IE7 incompatibility. One of the coolest thing IE7 is unable to do is actually processing form data when clicked on an INPUT field of TYPE IMG... which is pretty uncool for those using entire payment processes with such INPUT fields."
This discussion has been archived. No new comments can be posted.

IE7 Compatibility a Developer Nightmare

Comments Filter:
  • Vague FUD (Score:5, Insightful)

    by telbij ( 465356 ) on Sunday January 07, 2007 @01:25PM (#17498458)

    I'm gonna grant the author a free pass on the writing since it's obvious English is not his first language. But the rest of the article seems to be vague hand-wavy FUD and anecdotal complaints. To take two of the more cohesive statements:

    IE7 was supposed to comply more with the standards what in fact isn't true.

    The truth is that standards were not the first priority of IE7 (they are an evil mega-corporation after all), but they did do an awful lot of work on them. Just take a look at the list of CSS improvements [] over at the IEBlog. They acknowledge that there's a lot more work to do, but it's clear from this that they've solved a lot of headaches for CSS developers.

    And the MOST killer thing was the DISability of IE to submit data through "input type img" which actually was the whole sites data...

    I'm assuming the author means forms won't submit with an <input type="image"> tag. Without even testing it, I can't believe for one second this is true. This is a major backbone of HTML going back to at least HTML 2, and used in millions of websites. If this were broken it would have been fixed during beta. Microsoft may not care that much about web standards but they do care about backwards compatibility, and a lot of their decision making process has centered around not breaking things that worked in IE6.

    It's likely IE7 is going to be a headache for web developers, but this article doesn't do anything to support that argument. As a web developer IE7 hasn't really taken any of my time. So far it's been more reliable than IE6, and I look forward to the day when IE7 is the standard and IE6 is an afterthought for picky clients.

    • The developer could use a bit of javascript to hide the submit button, show the wanted image. Then an OnClick event on the image submits the form as per usual. This way it'll also degrade properly when javascript is disabled, seeing as the non-image submit is defaulted to.
      • by ari_j ( 90255 ) on Sunday January 07, 2007 @03:51PM (#17499816)
        Exactly. From what I've heard from my friends and family in the web business (including my sister, whose work recently won awards for degrading nicely), the main thing IE7 does is increases compatibility and conformance. From the sound of it (without reading the article, of course), this story is a bitchfest that probably stems from the IE workarounds that were poorly done falling apart under IE7, or something similar. In other words, it's partly the fault of the developers who are bitching and partly the fault of previous versions of IE, but probably not IE7's fault in any meaningful way.
        • Much as I despise Microsoft, fixing IE7's rendering (at least partially) was a good move. Not only did it move closer toward the standard, but it *broke* a bazillion web pages that were non-compliant but displayed as intended on IE6 or earlier.

          Previously, if you told those lazy web developricks, "Hey, your web site doesn't display properly on Firefox, which is a whopping 4% of the market!" then they would answer, "Works fine with my computer --your Firefox must be broken." Well, guess what? Now it doesn'
        • by Original Replica ( 908688 ) on Sunday January 07, 2007 @08:09PM (#17502160) Journal
          including my sister, whose work recently won awards for degrading nicely.

          Does she know you're telling people this?
        • Re: (Score:3, Informative)

          by syousef ( 465911 )
          robably stems from the IE workarounds that were poorly done falling apart under IE7, or something similar. In other words, it's partly the fault of the developers

          Just what were the developers suppose to do? *NOT* work around the bugs and non-compliance??? Or keep web sites so basic they didn't fulfil their business function???
    • Re: (Score:2, Insightful)

      by BenoitRen ( 998927 )

      I wonder if he even gave much thought to it:

      According to W3C the web browser market share of IE7 for the previous month is: 7.1% and for IE6 is: 49.9%

      Since when was the W3C site representative of the world-wide browser market share?

    • by dosius ( 230542 ) <> on Sunday January 07, 2007 @01:37PM (#17498580) Journal
      I look forward to the day when links is the standard and the browser everyone targets. I use Firefox but I don't think anyone should be forced to use ANY particular browser - if it runs on Links it'll prolly run on just about anything.

    • I have a site that uses an input type=image element. I just tested in IE7, and it works fine. Perhaps there is some specific onclick method that isn't being called on some sites? I don't know, but it certainly isn't completely broken.
    • by ozbird ( 127571 ) on Sunday January 07, 2007 @03:38PM (#17499710)
      But the rest of the article seems to be vague hand-wavy FUD and anecdotal complaints.

      *waves hand* This is not the browser you're looking for.
      *waves hand* You should use Firefox for your business.
  • unprofessional (Score:5, Insightful)

    by Anonymous Coward on Sunday January 07, 2007 @01:25PM (#17498464)
    Pretty unprofessional to use the "M$" moniker in a submission. But whatever. Also it probably should not say that people were forced since they have to agree to the install and don't need to do it at all. But whatever. Typical biased press you get here from some of the folks. Many folks can be more balanced but lately we hear from a lot of "slashtards".
    • Re: (Score:3, Insightful)

      by Myopic ( 18616 )
      i don't read slashdot for unbiased news. do you? if you want unbiased news, read the new york times. wait, shit, that's not a good example. try, what, the economist maybe. no, that's not very good either. well, shit, read news from several sources then.
    • by ednopantz ( 467288 ) on Sunday January 07, 2007 @07:04PM (#17501582)
      I *love* the M$ moniker! Without reading the submission, I know immediately that the author is an idiot. It saves a bunch of time.
  • by djkitsch ( 576853 ) on Sunday January 07, 2007 @01:31PM (#17498522)
    It would be nice to see examples of failing code, for instance. I've not tried it myself yet. The author's English skills leave a lot to be desired, to the point of making his complaints a bit vague.

    The upside to this is that the same "forced" upgrade procedure MS are using to roll out IE7 can also be used to roll out bigfixes without user intervention, if IE7 was originally installed transparently anyway.

    Anyway, I can't say it a big surprise that IE7 has its own foibles. IE6 was also a developer's nightmare, with the DOM and JS environment behaving differently from Firefox, Opera etc, so it's just another workaround. I'm only disappointed that Microsoft didn't make more of an effort to bring it inline with the competition so we could avoid yet another set of JS and CSS hacks. It's not like they have limited resources!

    • I can't see that they've fixed anything very much in IE7. Our IE6 fixes used the "* html" hack, and that was fixed in IE7, but then we found we needed to include all the same fixes anyway. So now we end up having to serve two extra stylesheets.

      I ranted about that on our company blog [].


    • Re: (Score:2, Informative)

      by cdwiegand ( 2267 )
      Sure, try using a variable named status or error. In IE7, MS silently makes those reserved words. How do I know? I used them in my AJAX library at my old job. Weirdly, IE7 would break, with IE6/FF wouldn't. After doing some testing, I eventually found out that you can't have a variable named either of those two in IE7's JS. And I found NOTHING on the web anywhere where they (or anyone) said that that wasn't allowed....
      • Status didn't magically become a reserved word in IE7. The variable 'status' refers to the status bar in IE7 just like it did in IE6. IE7 doesn't let script fuck with the status bar anymore (which is good), so the status variable is now read-only. If you were storing random information by saving it in the status bar object (the status variable), then your app is going to have problems in IE7, but really you shouldn't have been doing that in the first place.
  • Bah (Score:5, Interesting)

    by F452 ( 97091 ) * on Sunday January 07, 2007 @01:31PM (#17498526) Homepage

    And the MOST killer thing was the DISability of IE to submit data through "input type img"

    Maybe I'm not understanding what the claim is, but it's easy to demonstrate that this is not true. I just tried with IE7 to submit data on a form that uses an input type of image and see that it works fine.

    This article has almost no information and it seems the only reason it was posted here is to stir up anti-Microsoft antagonism. (Now someone will say, "You must be new here.") :-)

    In my experience, IE7 is much better at supporting standards than IE6. A huge improvement in CSS support, so that now as I design in Firefox 2.0 and occasionally verify things in IE7, I see that they are very, very close. Most of what I'm doing is working with WordPress blogs so it's very possible I'm not using things that are now broken, but if anything Microsoft should be given some credit for improving their browser.

    There's plenty of reason to not like Microsoft, but this article doesn't supply much (if any) ammunition, and it doesn't do the free software crowd much of a service to engage in our own unsupported FUD.

  • by yagu ( 721525 ) * <<moc.liamg> <ta> <ugayay>> on Sunday January 07, 2007 @01:33PM (#17498536) Journal

    As some have pointed out, this appears to be an incomprehensible rant. The "article" referenced says little and backs up that little with less.

    I also notice the "submitter" seems to be the same person as the blogger for the article. Not saying this shouldn't happen, but this usually shouldn't happen... If it's good enough to get "published", it's good enough to be published by someone other than the author.

  • by NineNine ( 235196 ) on Sunday January 07, 2007 @01:35PM (#17498572)
    This is just fucking ridiculous. This little rant is not only incoherent, but it's 100% wrong. See for yourself [].

    Jesus, do Slashdot editors actually *do* anything? Seriously. Do any of them actually *read* the articles they're posting, or is it all about pageviews and keywords?
  • by tidewaterblues ( 784797 ) on Sunday January 07, 2007 @01:39PM (#17498606)
    The editors must be desperate today, or else they don't read these things before they post them. I can't find any good reason why I should trust anything this guy is saying on his blog: maybe he has run into a major issue with IE, or maybe he just does not know how to code JavaScript correctly. If his JavaScript is anything like his grammar, error is a high probability.
  • This is so True (Score:5, Interesting)

    by Slipgrid ( 938571 ) on Sunday January 07, 2007 @01:39PM (#17498608) Homepage Journal
    It still won't parse the DOM. Stuff that is simple in Firefox, will never work in IE 7.0. I gave up trying to get some features to work.

    For instance, I have this js based terminal emulator. I don't want to edit that package, but just use js to read some fields. This is sexy in Firefox, but no chance in IE 7.0.

    if (document.Form1.tsprog.value == 'fibfm' || document.Form1.tsprog.value == 'FIBFM'){
    var pwrap = document.getElementById("pbsiwrap");
    var cells = pbsiwrap.getElementsByTagName("span");
    var item = cells[12].textContent;
    document.getElementById ("headspot").innerHTML = '';
    • You haven't been working with JavaScript for very long, have you?

      The MSIE equivalent of textContent is innerText, and it's been around since before Firefox was even thought of.
    • Re:This is so True (Score:5, Interesting)

      by mollymoo ( 202721 ) on Sunday January 07, 2007 @02:20PM (#17498964) Journal
      It still won't parse the DOM. Stuff that is simple in Firefox, will never work in IE 7.0. I gave up trying to get some features to work.

      The document.FormName.InputName.value form is not part of W3C standards. Try document.forms["FormName"].InputName.value and see if valid code works.

      I would hope that with an up-to-date doctype declaration (strict rather than quirks mode) Firefox would barf on that invalid code too.

    • Re: (Score:3, Informative)

      by NickFitz ( 5849 )

      The textContent [] property is defined in DOM Level 3. Microsoft has never claimed to support that recommendation, and neither Apple nor Opera claim support for it either. Gecko supports it, although the documentation doesn't make it clear that it's from DOM Level 3. If you write code using features supported by only one browser, then you can hardly complain if it doesn't work on other browsers. What you've done is no different from a developer who codes only for IE-specific features, and then whines when the

  • Beta testing? (Score:2, Insightful)

    by mikearthur ( 888766 )
    I'm sorry to sound like a dick, but isn't this what the massive beta testing period is for? Microsoft release beta versions well in advance so websites can be changed BEFORE everyone starts having forceful upgrades. If you wait until these upgrades to do your porting, YOU are at fault, not Microsoft.

    TFA was written by a guy who only recently has started porting sites to Firefox, so it's not really surprising he's finding this to be a pain.

    Really, the only people this will bite are people who didn't care abo
  • by enharmonix ( 988983 ) <> on Sunday January 07, 2007 @01:42PM (#17498650)
    With easily 50% of the replies to this story in favor of Microsoft and their standards compliance (never thought I'd say that), I feel it's safe to announce that Hell has, in fact, frozen over.
  • by n3tcat ( 664243 ) on Sunday January 07, 2007 @01:56PM (#17498768)
    FTA: "I recently needed to rewrite a web site so it works on firefox too..."

    I'd take what this guy has to say with a large grain of salt if this is how he treats his sites.
  • by Xugumad ( 39311 ) on Sunday January 07, 2007 @01:58PM (#17498788)

    Okay, firstly, I'd be bloody amazed if the pages in question validate. The guy goesn't give any link to the site, though, so I can't tell.

    Secondly... if you're using lots of client side Javascript to make a site work, you're asking for trouble. Google can do this, because they have massive dev and QA teams. If you don't have the manpower to do enough testing (for example, in the beta period) and fix problems, maybe you should make your site simpler.

    Every single web application I work on, worked perfectly in IE 7. Even, yes, the ones that use Javascript. This is achieved by:

    • Validating all pages. Okay, they're dynamically generated, so it's possible an error will slip past testing, but this really helps.
    • Testing under multiple browsers. The dev team works with Firefox and Safari, and does a QA pass under IE after any major revisions.
    • Minimising use of Javascript. If Javascript doesn't provide a significant obvious benefit to the user interface, it doesn't get used.
    • Minimising use of Javascript. If Javascript doesn't provide a significant obvious benefit to the user interface, it doesn't get used.

      Right on. Furthering that, if it won't degrade nicely, leave it out instead of putting a "this site requires JavaScript", and many sites don't even bother with that message anymore. Nothing infuriates me more than requiring client-side code to generate somebody else's content. If I have to enable scripting to see somebody's page, I probably don't want to see it.

      Perfect example is slashdot - if I don't have scripting enabled, I can still read and even participate without ever running a script on my side

    • Amazing! Sensible web development practices.

      I wish we had done that at the dot com I once worked for.....

  • News? (Score:2, Informative)

    by eldepeche ( 854916 )
    This is just a fucking whiny blog post. Give me a break.
  • the fact M$ decided to force it's users to migrate through update

    While I personally think MS Windows is lame, I do have a work laptop with XP installed. It is simply untrue that an IE7 upgrade is forced. It may be true that most users will end up with it installed because of their upgrade policies or habits, but IIRC it was as simple as not agreeing to the EULA to avoid its installation.

  • If, despite the assertions in this thread that the article is pure FUD, developers that built their software on MSIE did indeed get their software broken by the forced upgrade to IE7, it's still no big deal. These developers built their software on a single implementation, and then that implementation changed. Well, it happens. Now they just have to update their software to work with the new implementation. There is no faulting Microsoft for this. If, instead of developing for IE, they had developed against
  • I thought Internet Explorer had been a compatibility nightmare for years...

    I realize that what is meant here is probably compatibility between versions of IE, rather than compatibility between IE and other web browsers, but still. Seeing how badly IE does in the latter, I would not be surprised to see it did badly in the former, too.
  • Seriously, I have no freakin' clue why, but if you hit [], it crashes HARD. I just had a user point this out to me last night (seldom used page), so I haven't investigated why yet.

    The page loads a .csv file with the MS-proprietary tabular data control (TDC.ocx) and uses data binding to display information on the screen. The information may then be manipulated by JavaScript which can convert foot-points to newton-meters, and apply various filters to the data.

    It w
    • Re: (Score:3, Insightful)

      by 503 ( 236565 )

      Dude, you seriously need to clean up that HTML. There are validation errors galore (object in the head, mis-nested table/form tags and an open option tag, for example. Second, the CSV file has five columns for the header but only four for the data. Third, you refer to the wrong column names in your data fields. And fourth (and this is causing the crash) your data filter is messed up. Remove the filter parameter from the data object and it no longer crashes the browser.

      That javascript could use a good going

  • Apparently TFA is just a bogus rant, but I can imagine many of the brain-dead monkeys that design for IE-only would be up in arms about this new version of IE. IE 6 was released all the way back in 2001, many of these folks probably "grew up" on IE 6 and never knew or cared about anything else. Now they are finally being reminded that the web is an ever changing place.

    And to me the funniest part is that this not only affects actual web content, but also locally installed HTML, help files, and apps that stu
  • Forced Upgrade? (Score:5, Interesting)

    by rudy_wayne ( 414635 ) on Sunday January 07, 2007 @02:18PM (#17498942)
    "Internet Explorer 7 has kicked in at last on all MS Windows OS running PCs because of the fact M$ decided to force it's users to migrate through update."

    Oh really? I still have IE6, although I never use it except when forced to. Who are these people who have been "forced" to upgrade?

  • by nick_davison ( 217681 ) on Sunday January 07, 2007 @02:20PM (#17498956)
    Our brick and mortar business went through exactly the same thing. Henry bloody Ford and his evil empire released their latest and greatest product on consumers and it really screwed us businesses that had hitching posts and stables.

    One of the worst things about "Model T" was that it belched out carbon monoxide. Seriously! Compare this to a horse where the worst you have to worry about is methane! For brick and mortar store owners who didn't want to pay once again to upgrade from barns for their customers, this caused all kinds of ventilation issues. People could actually die from this stuff!

    Some people say that keeping up with the times is part of the cost of doing business. But where will it end?!
  • I found out that if you call appendChild to the body tag in IE7 it will not render the page and then the browser will become unusable.

    So do something like: var x = document.createElement("div"); document.getElementsByTagName("body")[0].appendCh i ld(x);

    Do this right after the body tag and viola, IE7 starts having problems.

    It does not render the page ( unless they have fixed this ). Instead it tells you that you have a possible network error. Then, IE7 thinking that the network is down, wont access any

  • by cgenman ( 325138 ) on Sunday January 07, 2007 @02:40PM (#17499168) Homepage
    For those of you who don't like babel fish:

    Because M$ decided on a forced upgrade migration path, Internet Explorer 7 has finally kicked in on all of their MS Windows OS running PCs. In theory, there isn't much wrong with this, but IE7 has failed to comply with many fo the standards it was supposed to.

    IE7 is a nightmare of crossbrower compatibility. I recently needed to rewrite a web site so that it works on firefox... and surprisingly enough, when testing the new and the old site on IE7 I discovered that many elements do not function as expected. And "not function as expected" doesn't really cover it, it was more a question of not working at all.

    I found some pretty large incompatibilities in their JavaScript submission systems. While I can understand the behavior migration, some functions did not pass parameters correctly etc.. The worst thing was that they disabled submitting data through "input type img" (which in this case was the entire sites data)... I don't blame the programmers for setting things up this way, but I do blame IE for not keeping with standards again.

    So with IE7 we come to a new era of "web developers nightmare." It will cause many smaller websites to need to be rewritten, and therefore will crush some of the associated small companies' business.

    And a final tip from me! Try to avoid writing JavaScript without testing it 100% on all of the major web browsers: FireFox, IE7, Opera and Safari.

    Just a clue:
    According to W3C, the web browser market share of IE7 for the previous month is: 7.1%
    and for IE6 is: 49.9%

    So with a bit of analyzing as well as some math we can say that IE7 will show a market share of about 60%... which isn't the best picture for us, the developers.

    And for those that do like babelfish... English via Italian.

    The Internet Explorer 7 has given of soccer to within in end on all the OS of MS Windows that ago to work the PC because of determined the M$ fact in order to force it is customers to migrare through the modernization. The sink really is not much evil with the this here but IE7 has been supposed to more join the champions who what in effects is not to align. IE7 is a cavalla of night for the greater part of the sviluppatori that try to join to the champions for crossbrowsing. Recently I have had to rewrite a Web site so as to it works to firefox equally... and the surprise element was that when examines the new and old place on IE7 I have uncovered that many things does not work as previewed and "the function poichè previewed" it is not the just word for it, it was more one issue than operation at all. I have found that submiting with the Javascript it has some graceful things of uncool that incolpo of the IE but some functions really did not work when no-go gage all the parameters and therefore via. And the GREATER PART of the thing of the assassin was the incapacity of the IE in order to introduce the data through "the type img of the input" that really it was the entire data of the places... Really not incolpo of the programmatori that have made it therefore but of the IE not to still maintain with the champions. Therefore really with IE7 we come to one new was "of the incubus of the sviluppatori of fotoricettore" poichè will demand the EVIL and perhaps DEFECTIVE places therefore of fotoricettore will not be rewritten and this with crushing some of the commerce of the small enterprises. And a final TIP from me! Tests to avoid the Javascript of writing without to verify it 100% on all the browsers used main of fotoricettore like FireFox, IE7, the work and the safari. As soon as for having an indication: According to W3C the percentage of the market of web browser of IE7 for the previous month is: 7.1% and for IE6 are: 49.9% Therefore with a little analyzing to the situation and the facts therefore as to make a sure one for the mathematics we can say that IE7 will show a percentage of the market of approximately 60% according to W3C... which is not the better image for we the devs.
  • Damn the modpoints...

    IE has a problem with: <input type="img" name="button" value="hitme">

    I don't understand how this is a problem NOW, as IE has had this problem for a long time. What the above snippet gets you is not "button=hitme", but instead "button.x=15, button.y=10" or something similar. IE returns you the coordinates where you clicked on the image, but not the value!!! Right. There is no value. Who would need that anyway? Doh.

    Don't remember exactly when I ran into this bug, a few years ago at
  • by nhavar ( 115351 ) on Sunday January 07, 2007 @03:40PM (#17499722) Homepage
    We have a webapp that was built for our call centers. It started out in IE 4 and then later we did some updates for IE5. IE6 came out and there were a few JavaScript things to fix-up (nothing huge) but by then we had ditched almost all of the non-standard code and moved from table layout to pure CSS layout. When IE7 came out we didn't need to make a single change to the JavaScript, CSS, or HTML. All 200+ pages just work.

    Everyone was worried that because of our choices and how heavily we rely on CSS and JavaScript that we would be in for a nightmare. Luckily, not so. However, two of the products we purchased for time tracking and for defect reporting both DIE because they have user agent checks built into their javascript. When they run into the unknown IE7 string they just chirp back a warning message and don't even try to load up. How many times do web developers need to be told "browser detection is bad use object/feature detection" before they'll get it.

    I don't think that I've had any problem with any of the normal sites that I use when using IE7 over the last couple of months (banking, news, mail, social). I get the feeling that the author of this article just doesn't know what he's talking about.

  • Bad math.. Not 60% (Score:3, Insightful)

    by CircularHowler ( 1045760 ) on Sunday January 07, 2007 @04:49PM (#17500348)
    This is flat out wrong. IE7 only runs on winXP and newer. So all the people running win2000, NT, 98 and 95 are stuck with IE6. Likewise pre-XP users are stuck with the 'old' media player version 9.0. Later the article implies that the current 49% of IE6 users will migrate to IE7. This shares the same flaw mentioned above. Anyway, people who don't want IE7 can either decline the install, or set a registry value to opt out forever. Or just don't use IE7; any product they give away for free has some ulterior motive.
  • by adamjaskie ( 310474 ) on Sunday January 07, 2007 @05:03PM (#17500502) Homepage
    As an anti-Microsoft, pro-Linux, Mac-using, pro-Firefox web developer that works for a University and thus has to cater to IE users, this writer was pleasantly surprised when he experienced very few issues even during his first IE7 compatibility tests. As none of the issues he did encounter were show-stoppers, he has been forced to conclude that IE7 is a huge leap forward, and the rapid replacement of IE6 with just about anything, Microsoft or otherwise, can only mean good things for the web as a whole.

Competence, like truth, beauty, and contact lenses, is in the eye of the beholder. -- Dr. Laurence J. Peter