Elliott C. Back: Technology FTW!

Chris Beach’s Pro-IE Blog

Posted in Computers & Technology, Spread IE by Elliott Back on July 19th, 2005.

This guy just sent me an email about his very pro-IE blog. He covers a number of things, including why IE renders the box model properly–even though it violates the w3c spec, firefox security bugs, and points out the craziness of spreadfirefox. Here’s a choice snippet:

The market dominating browser, Internet Explorer has a very resilient and flexible parser that will deal with most crap that the web can throw at it. However, several new bespoke rendering engines have become popular on the net including Gecko (Mozilla, Phoenix/Firebird/Firefox) and KHTML (Konqueror, Safari). These rendering engines, for better or worse, have stricter, less versatile parsers, which means they will fail to render certain sites as the designer intended. For example, browsing www.asda.com in Firefox currently shows no shopping aisle options and is therefore unusable. The Odeon website shows nothing but a background image in Firefox.

It’s the other end of the Firefox v.s. IE voice–and you might like it.

This entry was posted on Tuesday, July 19th, 2005 at 5:58 pm and is tagged with browser internet explorer, security bugs, box model, khtml, parsers, craziness, background image, konqueror, asda, odeon, firebird, gecko, parser, aisle, safari, crap, firefox, phoenix, email, shopping. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback.

Viewing 24 Comments

    • ^
    • v
    The IE browser is filth. Like street guns in the hands of children, the software is ubiquitous, dangerous, and irresponsibly distributed. Stop the violence.
    • ^
    • v
    I do not know about you... but I see ASDA web page "just fine" ( including shopping aisle options) on FF1.0 on RH9 LINUX.
    In other words: I could shop on ASDA site without any problem
    • ^
    • v
    Fine, fine, let's keep this marvelous parser and these non-standard pages. That way IE WILL survive, because there's no chance someone could make a better product that's compatible with these sites!
    Standards are there to ensure liberty to the users and to compete on fair grounds. If the "designer" decided to be non-standard and his site won't work "as he intended" anywhere else than in IE, then he intended something wrong.
    • ^
    • v
    Hmm, one can only prey this is a joke parody site of some kind! Obviously you've never used firefox.
    • ^
    • v
    Surely you have to draw the line somewhere with regards to being able to parse broken HTML. The example you give is www.asda.com. Glancing at the frameset source I notice that there is no DOCTYPE, no closing </HTML> tag, and the ampersands in the URLs are unescaped. Who knows what else is broken within the frame content itself! It should not come as any surprise that different browsers handle incorrect documents differently. If you insist on having DWIM functionality somewhere, put it in Frontpage, which will hopefully keep a good percentage of crappy HTML out of my browser. It's not the browser's job to correct someone else's mistakes.

    I think Chris's "standards are for losers" (not his exact words of course, but that's the vibe I was getting) attitude is the sort of thing that contributes to the incompatibility mess that spurned the browser standardisation movement in the first place.
    • ^
    • v
    I have to agree with the above poster. Standards are there for a reason.
    IE is crap and the reason its parser lets you get away with things is it probably isn't complete or microsoft wants people to fall into the IE trap.
    • ^
    • v
    My vibe is not "standards are for losers." I think that CSS1 and most of CSS2.1 are excellent. However, there are issues with the way CSS3 is being developed and by the W3C, and also how it is being adopted by Mozilla (introducing browser-specific syntax like -moz-opacity). See:

    http://www.chrisbeach.co.uk/core/scripts/entryV...
    http://www.chrisbeach.co.uk/core/scripts/entryV...
    http://www.chrisbeach.co.uk/core/scripts/entryV...
    http://www.chrisbeach.co.uk/core/scripts/entryV...
    and
    http://www.chrisbeach.co.uk/core/scripts/entryV...

    My issue with Firefox's inability to render broken HTML is fundamental. The job of the browser is to parse the web and render as much content as possible. That's what the user wants. The most powerful parser is the one that can correctly understand its input, no matter how obfuscated it is.

    Many have said that IE should not patch over other people's mistakes. By the same token do you criticise the efficiency of the fuel filter in a car, and suggest that if we get contaminated fuel, the car should choke and stop, so that the fuel producer will know there's a problem?

    I believe the browser should render all broken markup that is _unambiguous_. Of course, choking on ambiguous mistakes is fair enough. The responsibility for finding flaws in code lies with the HTML/CSS validator.

    By bloating the HTML/CSS standards (with the unmemorable DOCTYPE definition etc) and making the parsers stricter, we are discouraging beginner developers from writing the code by hand, and encouraging them to use tools like Frontpage. This is in noone's best interests. Everyone should take pride in their code, and one should code for standards. Whose standards one chooses is a whole different debate.
    • ^
    • v
    The purpose of a standard is to ease substitution of products. This allows for competition since it does not tie the user to a specific implementation. By supporting whatever garbage gets thrown at it, IE has effectively created its own standard and tied its users to its parser. Moreover, by supporting gargabe HTML, IE encourages obfuscated code that is a maintenance nightmare. The best way to ensure correct usage of anything is to disallow incorrect usage wherever possible.

    Firefox, on the other hand, adheres to a standard that demonstrates its willingness to compete on a level playing field with any other browser. It also refuses to cater to the poetic license taken by bad HTML. HTML, the language, is not poetry; it is a computer language with a defined semantic.
    • ^
    • v
    The -moz attributes are specifically named as such so developers will understand perfectly that they are Mozilla specific, and NOT OFFICIAL. If these attributes (such as opacity and rounded corners) are adopted as official W3C standards, Mozilla will support the standard and drop the -moz ones.

    "The most powerful parser is the one that can correctly understand its input, no matter how obfuscated it is."

    OK you are obviously painfully ignorant of IE's many BUGS. Things that completely break its rendering engine (esp. CSS2), and web developers have to bend over backwards to accomodate. I am a professional web developer, and have developed sites for the likes of Shell, BMW, Honda, United Way, and ATCO. An enormous amount of time is spent accounting for IE's rendering bugs. Google for "peekaboo bug" or "3 pixel gap". Some of the techniques for getting around these IE bugs include using negative margins, or having to specify dimensions when you don't really want to.

    Here's a page highlighting some of IE's many horrendous bugs:
    http://www.positioniseverything.net/explorer.html

    This is your browser that "can understand any input, no matter what"? It completely mangles 100% standards compliant pages on a regular basis.

    The only reason so much of the web renders nicely on IE is because 95%+ of the web was built for it, because of its de facto dominance due to being bundled with Windows. This is a vicious cycle we are finally starting to break out of.
    • ^
    • v
    UM, I supposed you don't like the moniker "code is poetry," then...
    • ^
    • v
    Very much the contrary, Elliott. The phrase, "code is poetry," is used to convey the importance of clear and concise code--it should not be taken to mean that slop is acceptable in the name of art.

    My earlier post spoke directly to the importance of maintainability of code. That is the objective of code as poetry. I agree with the idea of the moniker, but I also believe there is a limit to the metaphor.
    • ^
    • v
    The moz- prefix is exactly the sort of standards-compliant behavior the W3C recommends to allow browsers to support their own style implementations. Compare that to Internet Explorer, which introduces its custom extensions in the root namespace.

    I don't advise using browser specific extensions unless necessary to adjust for bugs or non-standard behavior, but Mozilla and KHTML have done the right thing here.
    • ^
    • v
    I love IE!

    I am tired of people making witty websites that don't display properly in IE. "My site is important enough that you'll probably install some crappy browser to read what I have to say."

    I am tired of people putting firefox ads all over their sites. It feels like getting teabagged every day.

    It is not more secure, it loads slower, and it doesn't come standard with my OS. I don't ever get crashes or spyware, but I spend over 10 hours in front of this computer each day. YOU are using your computer wrong. Stop polluting my screen with your stupid, stupid browser ads, I'd rather "punch the monkey to win $10,000" than install firefox.

    Standards? Yes, they are important. Replacing all tables with divs and spans because it's 'trendy'? Give me a break! It is very easy to make pages that look just the same in IE and firefox or any browser, it just takes a little bit of brains. 70% of your visitors will be using IE, do you want to cripple 70% of your visitors? OBVIOUSLY NOT!

    Making everyone switch to the same browser is not the solution, making your pages compatible with all browsers is, regardless of whether you are following standards or not. We make websites for visitors, not to satisfy some kind of coding standard, so the least we can do is not impose browsers on people. This is what taking back the web is -- making people not worry about what browser they'll use, because we webmasters can guarantee that pages will look the same on every computer.

    I design and code dynamic pages for a living, and I am tired of these amateurs complaining about IE. Do you think your opinion counts? Why not try petitiononline.com to get IE to respect standards, maybe it'll work better than those "spread firefox" buttons. What a damn joke.

    In other words, unless you're working at microsoft or coding firefox, you probably shouldn't be worrying about insipid crap like web standards.

    Three thumbs down to brainwashed firefox zombies.
    • ^
    • v
    Chris, who are you to say the job of the browser is to view all sites? And say thats what all users want. Clearly not, as 10% now use Firefox. This browser amongst others obides by web standards, which are there for the good of the web. They are simply guidelines for coding so users can view all sites correctly. IE should have obided by these for the past 5+ years, but have chose not to. The result is web designers whom could normally simply write code following web standards in an hour, have to then hack the code, to work with IE. That is clearly a problem with IE, not Firefox or any web standards browser.

    A browsers job is not to view any piece of crap that could harm the user and means people are free to write any code causing absolute cayous.

    Websites are increasingly now following web standards, and removing improper coding for there websites. This was there and IE's problem, most sites realise this and have no problems. IE 7 beta is also obiding by more webstandards, showing clearly this was IE 6 and befores problem, affecting the entire web and how people code. This will cause as much if not more disruption, because those who built sites only following IE's invalid coding, now have to change that to fit web standards more.

    If a company or supplier breaks the rules over years and SOME people become accustom to it (though it very damaging, harmful and time consuming) it does not mean a competitor should come along, and do exactly the same thing.
    • ^
    • v
    Aw... Apparently I have to go &lt; and &gt; to put some < and >s into comments :(

    Anyway, I meant to write "You're not > 'abiding' < by the rules of proper spelling, why should your browser abide by correct standards :)"

    I personally can't wait for IE7, it will truly be a step forward.
    • ^
    • v
    Imagine what a compiler would be like if it allowed you to blatantly ignore standards. Things like ANSI and POSIX exist in the programming world for good reason, and there's honestly no reason why a browser shouldn't at the very least conform to a minimal set of standards.

    Remember back when people needed two versions of their site: one for IE, the other for Netscape? That was sheer idiocy (although understandable at the time), and that's why we need to pay attention to standards.
    • ^
    • v
    Imagine two students trying to learn a foreign language. You have two teachers - one of whom marks his student's homework as correct, no matter how blatantly wrong it is. The other teacher will not mark his student's homework as "correct" unless it conforms to the actual grammatical structure of the foreign language which you want to learn.

    Question 1 - which student learns better
    Question 2 - when visiting the country in question, which student fares better?

    Grammatical structures are there for a reason - whether one refers to a foreign language or to the source code of a website. Whereas it may seem to be much easier to fail to conform to the strict grammatical structure of html etc, this only works if enough people use the browser which displays this non-compliant code. Thus, if this is not addressed, web site developers will ´find it increasingly difficult to code for the Microsoft offering, as it will be the exception to the rule - simply watch the Firefox space.
    • ^
    • v
    Quote: "Imagine two students trying to learn a foreign language. You have two teachers - one of whom marks his student's homework as correct, no matter how blatantly wrong it is." ..

    You're another who confuses the concept of a web _browser_ with a CSS/HTML _validator_.

    The job of a browser is not to mark anything "correct." The browser should render the Web, simply that.
    • ^
    • v
    Please help - why do blogs in blogspot (for example) look different on ie and mozilla. Are there css tags to place in the blog template to make ie and mozilla blogs look the same?

    Mozilla looks fine, but IE is messed up.
    • ^
    • v
    The reason that Asda's website doesn't work properly in Firefox is because of Javascript references to variables using .value when the variable isn't an Object. Plus a few other things. Anyway, I found a way to fix it. Take a peek at my web page (linked from my name above) to see how.
    • ^
    • v
    Hello, nice site look this:
    Glad to read your site,
    What the hell is going here?!
    Wait for me,
    Did you read this?
    Can you help me?
    I'm back in USSR!
    You have a beautiful website!

    End :) See you
    What do you think?
    Good day, I'll be back :)
    • ^
    • v
    Jugshemash!
    I just want to say "hello" from Brunei internet community!

    Bye, I like ;)