Skip to: Navigation | Content | Sidebar | Footer

Weblog Entry

A Response

September 10, 2004

Jeff Croft points to an article published on ZDNet last week which implores standards advocates to drop the religion. It’s a pretty big church…

In response to the launch of the Web Standards Project’s Browse Happy campaign a few weeks back, John Carroll shared some thoughts about what alternative browsers ought to be doing to topple IE’s lead.

Boiling down John’s article reveals only a very small grain of actual suggestion: to challenge IE, other browsers must duplicate all its proprietary features. His final conclusion appears to be that standards are good, but slow, so why not use IE as the de facto standard and let the W3C play catch-up?

I’ll tell you why not.

Because the legion of developers and designers clamoring for Microsoft to improve their standards support is growing daily. No one is begging Mozilla to implement ActiveX and the rest, John’s pet issues aside.

Always remember the wisdom of Monkeyboy.

Reader Comments

September 10, 01h


I do understand that. I said the exact same thing, didn’t I?

September 10, 01h

Erm, guys — I think a few of you are getting confused. Jeff Croft didn’t write the article on ZDNet, he merely reported it.

Robert says:
September 10, 02h

I apologize, since this really offers nothing to the argument, but I felt like someone had to say it.

All I know is that playing by the rules (standards) should count for something. Is it fair to make up your own rules in the face of competition—as Microsoft did with Netscape? Isn’t there something to be said for doing things the right way? And who really believes that caving to IE’s shoddy implementations would really stir up new competition with Microsoft? Ooh, they’ll be shakin’ in their shoes!

There’s a reason why alternative browsers are getting noticed—because they do things right and they do them well. But that’s not enough, it’s up to us designers and developers to ensure that standards are upheld.

Jon says:
September 10, 04h

I think I said it all in the reply I posted on ZDNet:

“… it’s my opinion that browsers should follow standards, allowing the internet to progress to newer and better standards, rather than standards following browsers …”

Although, to be honest, it doesn’t really surprise me that people are writing articles like this - I remember defending NS 4 18 months ago (then I discovered the Zen Garden - thanks Dave) - but what does surprise me is that semi-influencial websites like ZDnet are publishing it when it’s blatantly Wrong.
Oh well…


September 10, 04h

Well said Dave! That ZDNet article was the twittiest thing I’ve read in quite a while! John Carroll has obviously never paid any attention to standards. It is much easier to follow the standards, then try to comprehend the behaviour of IE’s many extensions.

One of his arguments for why users switched from Netscape to IE is because pages built for IE weren’t rendering correctly in Netscape. Of course, that is one of the only things he wrote, that seems even close to being correct, but he has a point. I’ve been saying it for a while now, it’s about time we, as a community, dropped support for IE, and allowed our pages to degrade gracefully — that *is* how CSS was desinged. The only problem is getting the idea past the marketing people that run the businesses; however for personal sites, there’s nothing holding us back.

If users start to find that sites are just not rendering properly in IE, and are advised that it would render in a standards compliant browser such as Mozilla, Opera or Safari, then I believe they would start to switch. Of course, there are still some sites that require IE, but thanks to the Moz, Opera and Safari development teams, that number of sites has been significantly reduced.

Lach says:
September 10, 06h

Lachlan; I find it far more likely that most IE users will simply assume your site is broken and stop visiting it. I’d be surprised if more than 10% of people would be prepared to switch browsers all by themselves, and I’d be surprised if a majority of people even knew what a browser was.

PTDC says:
September 10, 08h

He finishes saying that standards are necessary, but somehow completely misses the point of having them in the first place. Now this might be because I don’t use IE anymore and know little of its ghastly proprietary features, but it seems that it’s IE that needs to catch up to the standards and not the other way around.

web says:
September 10, 08h

“HTML standards, as standards-compliant HTML/CSS sites DO render well in IE”

They do? I spend the majority of my time fixing IE’s buggy problems.

File this guy under “tool”

Dan says:
September 10, 09h

“HTML standards, as standards-compliant HTML/CSS sites DO render well in IE”

That’s only because developers spend so much time working around IE’s incompetent rendering so the site will render well to IE users.

The guy perhaps needs to make better use of the “view source” function.

You know, perhaps Moz. and the others should support IE’s worthless proprietary features. Then we could all be vulnerable to malicious ActiveX applets and we’ll get to see the fancy gradients on the Microsoft’s homepage.

September 10, 09h

As I said on my site, non-standards features aren’t inherently a bad thing. Let’s not forget that Mozilla has a whole boatload of “-moz” attributes in CSS. But, the Mozilla implementation clearly varies from Microsoft’s. The difference is three-fold:

* Mozilla uses the -moz prefix as a way of delineating their non-standard features from those baseline, standard features that can are supposed to be able to count on.

* The Mozilla extensions were created with the purpose of getting them into the baseline CSS standard standard at some point — not with the intention of creating diversity so that developers would choose it over IE (which is clearly Microsoft’s plan).

* And the biggest, and most important difference is this: Mozilla made sure it had it’s ducks in a row with basic standards support before worrying about bonus features. It’s obvious that the Mozilla team puts baseline CSS support first, and any “bonus” features second. And that’s the way it should be.

September 10, 09h

I am saddened by this sort of thinking, and wrote a rebuttal far too long for a comment. You can read it at

hemebond says:
September 10, 09h

I have my default charset set to utf-8. Upon visiting the article I was faced with a page dotted with invalid characters.

I didn’t bother reading the article. If they can’t get the basics right, why should I listen to anything they have to say?

September 10, 09h

“Lastly, recognize that developers will favor a technology that serves 95 percent of their target market”

Actually, developers will favor a technology that they know will be around for awhile AND serves a majority of the market, because they don’t like rewriting their site/application/system/whatever every year. This means a more popular subset of what IE supports (vis. standardized HTML). If developers used IE’s proprietary extensions, there is no guarantee that IE will be the dominant browser for the future. His whole argument hinges on that 95% rule.

But what happens when IE is no longer 95% simply due to browser dilution? After all Microsoft has clearly stated that development will pretty much stop for IE until IE7 comes out for longhorn (yes, I am familiar with the latest IE service pack). What will happen to the browsing landscape for the next 3 years? Will more people adopt alternative browsers because they need the features? Or will web users stagnate on IE 6? If “everybody getting viruses” is used as a motivating factor to switch browsers, I ask you, just look at how many people see it as SOP to reformat & reinstall windows every few months.

September 10, 10h

Jeff, you were invited to the party, we will welcome you with open arms, dont feel left out, bub!

September 10, 10h


You lost me. Huh? :)

Jeremy says:
September 10, 12h

Well said Dave. And Jeff, please understand, it’s Microsoft who needs to step up, not the rest.

Peter says:
September 11, 02h

About moz’s proprietary -moz- properties; its syntax, too, is based on a standard :)

And a lot of them are just CSS3 properties with a leading -moz- (like border-radius,, which makes sense considering it’s still a draft, and thus subject to change. At some point the -moz- part could then simply be dropped.

There’s a bit of truth in one of the last paragraphs though. Non standard things (like innerHTML and overflow-x/y) can sometimes be just what the standard missed. Standards are good, but you don’t always have to *agree* with them. Individual browsers can sometimes have some ingenious feature that should become a standard.

mangoduck says:
September 11, 02h

john carroll’s points seem to contradict themselves. he says:

“If makers of alternative browsers are truly serious about challenging IE’s dominance, then they need to ditch the standards religion and make their challenge credible. They need to support EVERY feature of Internet Explorer, at least as it pertains to the rendering of HTML pages.”

this makes no sense. in his little history lesson above that, john omits microsoft’s tactics in securing explorer’s dominance after it took the lead, which was by implementing a slew of proprietary html features, and more recently, css properties, with the intention of making both designers and users reliant on them. microsoft uses their dominance to perpetuate itself and strongarm the online community into doing whatever they feel like, official standards be damned. don’t think for a second that they want to be a good citizen and work with the w3c at the sacrifice of their marketshare.

regardless of whether any of these live features are a good idea, conforming to explorer’s pseudostandards will do anything but topple its lead. the only logical action is a boycott of all proprietary rules and markup, including the ones in gecko. like someone already said, -moz properties are basically working demonstrations and in my opinion shouldn’t be taken as anything more.

now i don’t mean to say that including working extended features as a proposal of sorts is a fundamentally bad idea, because this happens with other softwares and gives good results. for example, apple often rolls popular third party add-ons into future releases of os x. the main difference though is control. apple is one company with one development team setting one list of standards for their operating system. the web is thousands of joe schmoe developers and no enforceable quality control of any kind. if every one of them made proposals via rendering engine the industry as we know it would collapse. the world of web development is a particular case where there is either one standard, or there is no standard. no gray area allowed.


mangoduck says:
September 11, 02h

“The presence of non-standard features is not a bad thing, but is a reflection of the fact that standards cannot hope to encompass all that is possible in software. Standards are designed as a means to ensure interoperability when such interoperability is necessary, not a limiter on progress.”

standards are not about invention, rather they aspire to guarantee that existing technologies will operate and interoperate how they are intended, which is always both desireable and necessary. i submit that there can be no significant progress unless standards exist. nonstandard features are indeed a bad thing in that they undermine any progress we could hope for. falling in line with unofficial features feeds the entropy machine.

i would also like to note that john carroll is a software engineer and screenwriter, who specializes in the design and development of distributed systems using java and dot-net, as stated beneath the article. where does one see anything about web design and development? would you say this man is qualified to be asserting such outlandish rhetoric? what would you say his rant is really worth?

Evan says:
September 11, 12h

The reality is that the MOS browsers have to walk a fine line.

It’s easy to forget that for every one of us who is a devoted reader of Mezzoblue, there are ten designers who simply test in IE and deploy if it works.

Yes, these people suck. Yes, I personally would be thrilled if these people would make some effort to edumacate themselves. But unfortunately, these people not only exist, but they produce the vast majority of the sites out on the web. Any modern browser that doesn’t make *some* attempt to recognize this fact is toast.

So you pick and choose your battles. Mozilla and Safari do not mimic IE’s craptastic CSS behavior, thank goodness. But Mozilla and Safari *do* implement XmlHttpRequest. Is that evil? Or pragmatic? You tell me.

Carmelyne Thompson says:
September 12, 03h

“Imagine that you bought yourself a new car and it runs just fine and then you start getting berated by some motor engineer trying to tell you that your driving experience would be 2% better if you installed a new third party suspension all by yourself. Would you bother? Neither would most other people.”

Standards aside, a user makes it his/her business when faced with a hijacked home page, annoying pop ups and pop unders, as well as being redirected to unsolicited sites. Thanks to IE activeX and scripting for vulnerabilities that plague any user equiped with IE. Don’t you consider spamware softwares as third party to augment an already existing anti virus program in your desktop?

What reaction would you get from a regular user when you say.. “Hey you know… you wouldn’t need spamware so and so if you used Firefox. You’ll save atleast $30!”

I wonder which reaction it would be. Would it be googly eyed or the dropped jaw?

September 12, 03h

I love this bit, right at the top of the article: “Avoiding a browser used by 95 percent of Web surfers will make you less likely to get hit by viruses which target it.”

Very logical. Explains why there are so many more exploits targeting Apache than IIS. I always wondered why that was…

Golightly says:
September 12, 06h

IE is bundled as part of the Windows operating system. That’s why it’s the dominant browser, because most people don’t know how to change it, probably couldn’t be arsed to change it even if they know how, and may well just end up getting errors even if they did.

So long as it’s integrated into the OS it’s here to stay with a huge share of the market. I don’t like it anymore than the next person, but it’s time to deal with that fact, get over it, and move on.

Imagine that you bought yourself a new car and it runs just fine and then you start getting berated by some motor engineer trying to tell you that your driving experience would be 2% better if you installed a new third party suspension all by yourself. Would you bother? Neither would most other people.

It is not that hard to design something that works across the browsers and you all know it.

Get over it, move on.

mangoduck says:
September 12, 11h

i think you’re taking this too lightly, golightly. a two percent better experience may be accurate for all i know, but it’s only that low because developers put elbow grease into things so users don’t have to. oh wait, sometimes they don’t!

also, i agree that designing for all browsers isn’t hard, but i think it’s harder than you’re thinking it is. an anecdote:

i’ve been struggling recently with a design where several overlapping background images are pinned to the bottom and bottom right. any time you’re dealing with 100% heights in css, things can get difficult.

first, firefox does whatever you tell it. good boy. test in ie win, doesn’t work. find a workaround for ie win, it breaks firefox. rework the layout and finally manage to get both to agree, you’re thinking it’s all downhill now, then opera win mangles it. ie mac and safari don’t like min-height, and every workaround you try has no effect so there’s not much you can do. frustration. consider using tables, or making a new design from scratch.

this being said, i think the real debate here is adherance to standards, not design practices.


this brings up another thing i meant to mention about john carroll’s article. he refers to old netscape and ie as if they’re separate software platforms or something. designing specifically for one or the other, and even giving up on netscape in the end because ie was making bigger advances and rendered pages correctly that were meant for netscape.

it’s all one web, john. design with standards, then make sure it works in everything under the sun. if browsers of old were competing with each other instead of running towards the goal line, then they had lost sight of the grand scheme. since then moz has woken up, but microsoft has not.