Disenchanted comments on the "Web Standards Project"
May 8, 2001 9:10 PM   Subscribe

Disenchanted comments on the "Web Standards Project" and he disagrees with it. And I think what he says rings true: "There's no point in hanging a 'No Shirt, No Shoes, No Service' sign outside a web site."
posted by Steven Den Beste (71 comments total)
Regardless, my sites will have standards, even if visiting browsers don't.
posted by toastcowboy at 9:13 PM on May 8, 2001

Well written argument.

What I found woefully ironic, though, is that the page was done with CSS, and degrades rather nicely without it - just like the WaSP pages do. Funny, but if the page is supposed to look the same across all browsers - even ones that don't render CSS properly - shouldn't it be riddled with font tags?

That said, if the page is smart enough to discern the user's browser, I withdraw my complaint.

And, no one ever claimed that asking users to upgrade would be easy; no one ever claimed that sites should kick 4.x and under browsers out. I personally think it's silly to entirely kick out older browsers. But I also think it's silly to expect an old browser to render the page the same way a modern, standards-compliant browser does.

It is a mindset upgrade. Instead of making things look fine in old browsers with the potential of screwing up the display in new browsers, WaSP has flipped the switch.
posted by hijinx at 9:32 PM on May 8, 2001

I don't kick out older browsers. I just don't support them very well. Apparently, most of the users of those older browsers like it that way. Chacun a son gout.
posted by dhartung at 9:39 PM on May 8, 2001

Older browsers should never be kicked out completely, but not all websites can make them the first priority, either. People using older browsers deserve the text content, but they cannot expect to get the whole shabang every single time.
posted by jragon at 9:53 PM on May 8, 2001

CSS was developed with total accessibility in mind. The author glosses over it, but the fact is that once standards are in place across the board (and built to, of course) every web site will become more than a web site. You will be able to print it and it will look acceptable. You will be able to look at it on a television, or a projector, or a TTY device, or WHATEVER... and it will not DEGRADE, it will ADAPT.

Why do you think you don't have to think about which light bulb to buy? Standards. Or what kind of gas to buy? Standards. Or what voltage your electric shaver uses? Standards.

I am NOT going to hack my way through HTML tables anymore. I don't care a spit for NS4 users. If they are un-educated about standards, fine: let's educate them. If they don't CARE about standards: forget 'em. Let them go the way of the Amish (still around, but a silent and largely ignored minority).

(No personal offense meant to the author, NS4 users or the Amish... i know they can dig it.)

(And if my boss is reading... of course I'm going to hack my way through NS4 fixes... I know our client can't afford to alientate that crucial 9% of web users...)

(Stat via echoecho.com.)
posted by o2b at 9:55 PM on May 8, 2001

And one more thing...

...nobody really gives a damn how your site looks.

Say WHAT!?

Visual presentation is the KEY to usability on the web (not to mention branding, if I can broach such a commercial concept in such a conceptual thread). Numero UNO. If your site ain't legible, people ain't gonna use it. And if people aren't gonna use it, then what's the point?
posted by o2b at 10:01 PM on May 8, 2001

That's a very thoughtful article. I encounter mainly two camps in web design. 1) Designers who practice creative design (unique, artsy, inventive stuff). They are usually about control. 2) Designers who practice sublime design based on the idea that the visual trappings should be pretty, but innocuous (transparent) - content is the star of the page. You might call it industrial design. It's a creative process too, but in a more functional, practical way. Most web sites that adapt to the user have industrial design.
posted by fleener at 10:01 PM on May 8, 2001

On paper, I agree with the cranks -- separate style from semantic content, don't let the designers run roughshod over visual layout at the expense of that principle, etc. -- but they've declared war on the wrong threat. The problem isn't CSS or Jeffrey Zeldman, it's sites laid out exclusively in PDF, Flash, or the soupy mess of modern HTML, all of which are engineered to expect total visual control, and tend not to degrade worth a damn.

This hidden agenda [of total control] becomes obvious only when you consider that a page properly marked up with CSS - exactly the way WaSP is advocating - will work in any browser, even the "broken" ones like Netscape 4 and Internet Explorer 3. Why, then, should anyone need to upgrade but to satisfy a designer's craving?

... because when you're writing for a client, they're not going to accept that definition of "work." They're going to pick the convoluted, slow to load, impossible to edit version that looks okay in Netscape 4 over the clean, validating, stylesheet-centric designs that require support of CSS standards codified in 1996.

There are tremendous untapped benefits to predictably parsable documents, uncluttered by stylistic folderol (starting with the ever-popular "search engines that actually work"). We're never going to get there without something akin to XHTML and CSS, and with Netscape no longer posing a competitive threat to IE, there are no more market forces to drag those standards into the mainstream. So God bless Mr. Zeldman for giving it a shot.
posted by bumppo at 10:04 PM on May 8, 2001

This guy's whole argument hinges on this idea of separation of content from presentation. As a programmer, I certainly see why this is highly desirable where possible, but on the web, isn't that frequently completely impossible?

Presentation is Content! I thought that was established. Sure, a lot of data can be boiled down to text or columns of numbers, but then what does the web give us over straight text? Things like flash and such potentially allow the skillful designer to add insight to information, to communicate potentially complex things more clearly. These types of issues are what lead people like Edward Tufte to start their own printing companies to make their books, or the McSweeney's folks to send all their stuff to iceland to get printed. A tremendous part of the value of a communication is in its presentation.

Ideally, given plain data, it should be possible to view it in a variety of ways, including those chosen by the user, but this type of data is better kept in XML than HTML, and should be translated to the desired target according to rules, potentially (by default) specified by a designer. The designer should be guaranteed that most of his rules should be followed so that he can maximize the impact of his content through effective presentation. There's content in presentation, you can't separate them completely.

posted by jeb at 10:54 PM on May 8, 2001

I'm still waiting for the day when I can download a tasty pizza over the Internet. Until then as far as I'm concerned the argument of web standards is just theological debate.
posted by ZachsMind at 11:13 PM on May 8, 2001

"Well written argument."
Perhaps it would be, if it weren't as aggressive as it is.
posted by prolific at 11:40 PM on May 8, 2001

The web must cater to the user and there are definitely two diferent types of users.
If you are writing a research paper and embracing the web as a massive library of information, the style of presentation (aside from legibility) doesn't mean a damn thing.
On the other hand, when using the web as a form of entertainment, the presentation becomes extremely important.
There are so many stages in the design process, especially those considering user needs, that standards become idiotic and useless.

The argument of the Flash sites doesn't so much become one of just usability and another example of over-designing, but one of the creators ability. I've seen just as many Flash sityes that were quick and wonderful as I have seen that were slow and tedious.
posted by ttrendel at 12:03 AM on May 9, 2001

thanks for the link and for the comments here. for what it's worth, i've replied to disenchanted on the daily report.
posted by Zeldman at 12:05 AM on May 9, 2001

Yet another rant about web design (what is it about the web and its implementation that people feel they have to write in such an aggressive way about it) that's arguing in absolutes. When are people going to realise this is the web we're talking about! There are no absolutes.

The technologies, methods, design strategies and implementation of web projects must be goal driven and every web project is different. If I want to use the web as a medium of personal artistic expression I'm going to approach the development process in an entirely different way to designing an application to dynamically serve real-time laboratory data over an extranet, selling auto-parts to Eastern Europeans or allowing stockbrokers to get ticker prices on their cellphone.

Page volume, maintainability, traffic, audience sophistication, content relevance, data-modelling, UI conventions, content dynamics and umpteen other considerations all impact client technology choices. You bet web standards are about control! As a developer I want massive amount of control, the more control the happier I am. I want to know that whatever solution I can come up with that best meets the needs of a project will bloody well work the way its supposed to in whatever client-applications its supposed to work in, now that's control. The kind of control that software developers have always enjoyed.

I have this idealistic fantasy where if a web site or application is crap we all know its crap because it was developed by someone who did a crap job, not because they didn't manage to keep up to date with the implementation on platform X using client Y.versionZ of what should have been a simple matter of saying something like "I want this bit of text to appear next to this other bit of text, not above or below it."

And why shouldn't producing relatively simple pages of text and graphics in a browser be as simple as using something like a word processor or dtp app? Good grief, its text and graphics for goodness sake. Why are wysiwyg web tools so crap and FrameMaker so good? Formats v. standards is why. The wysiwyg makers have barely a clue what the product of your labor is going to appear inside and how its going to mangle it, because there isn't a good enough set of standards to work within. Yet.

You'd think that a relatively young industry, waking up to its potential for global adoption, could have intelligence enough to look at every other industry in the world and realise adoption of standards benefits everyone.

But this is all moot. Standards are happening. CSS is here to stay, ECMA and the DOM are here to stay, XML is only just beginning to happen and is going to be very here to stay. And HTML is all but dead already. Not because some designers quite fancy the idea and not because Mr Zeldman says so but ultimately because it makes economic sense.

Standards mean better, faster development tools, much greater accessibility to those tools and what's produced with them, content management systems that mean a text edit can be done without needing several weeks training, seamless transition of content between media and, essentially, the people who make the web being able to concentrate on what they're good at, knowing that a team effort will result in something that works how its supposed to.
posted by normy at 12:11 AM on May 9, 2001

Two kinds of users

I agree. For the sites I use, something that kicks out my browser (NN 4.7) is usually an indication that I'm unlikely to find anything interesting (I have IE available too, and recently added K_Meleon, so I can switch if I have to).

Are there any convincing examples of this wonderful usability, that old browsers hamper, have actually useful content?

The eye-candy brigade probably do need the latest browsers, but the people that use those sites are people who have the latest browsers anyway, because presentation/design is the thing that interests them.
posted by andrew cooke at 12:29 AM on May 9, 2001

what normy said.
posted by Zeldman at 1:01 AM on May 9, 2001

Once upon a time a lowly servant girl lived in a dingy hovel next to a big palace full of handsome princes. 'Why, oh why,' she sighed, 'can't I find me a handsome prince? Why won't they look at me?'

At which a Fairy Godmother appeared in a puff of pixel dust and said, 'They won't look at you, little girl, because you are not well-Presented enough.'

'But my heart is pure!' cried the girl. 'Surely its Content is all that matters!'

'Pshaw,' said the Fairy G, 'Content is all well and good once you've bagged yourself a prince, but they won't even look at you without good Presentation. Here, let me show you.'

At which the Fairy G waved her magic wand, and swathed the servant girl in a cloak of the purest GIFs and JPEGs all bundled up in Tables. The effect was wonderful to behold.

'I'm beautiful!' cried the girl. 'I'm going to go snag me a hunk o' lovin' man right away!'

'Not so fast,' warned the Fairy G. 'This is a magic cloak. They can only see it if they're using these magic glasses.' And with that, she handed the girl a box of spectacles all labelled '3.0'.

'Thank you, oh thank you!' cried the girl, and skipped merrily down the road to the palace. Soon she was inside, handing out magic glasses to the handsome princes, who donned them and saw the most wondrous voluptuous vision of splendour they had ever beheld. 'Prithee, fair maiden,' they would cry, 'where hast thou been all mine life?'

And so the girl enjoyed the attention of many suitors, and figured it was only a matter of time before she Got Lucky. But little did she know that she had made herself an enemy. A jealous princess, who also happened to be an Evil Witch on weekends, was devising a scheme to undermine her success. 'I'll get you, my pretty,' she cackled, as jealous part-time Evil Witches are wont to do.

Soon this Evil Witch had created hundreds of Evil™-brand magic spectacles, and was spreading them around the palace. When the princes foolishly donned these, the servant girl no longer looked a thing a beauty, but a hideous hag draped in sackcloth and warts.

'Aieee! She is hideous!' they would cry, before running away screaming.

'But... but... my Content! You can still see my Content!' cried the girl, but it was no use. In tears, she ran back to her hovel, where her Fairy Godmother was waiting.

'Help me, Fairy G! They think I'm hideous!'

'What is it, my child?' asked the Fairy G.

'They're using these spectacles, and they make me look awful,' sniffed the girl.

The Fairy Godmother studied them closely. 'Aha. Evil™-brand Uncompliant spectacles. I might have known.' She turned them around in her Fairy Hands. 'Ah yes - here - the fatal flaw. This I can fix.'

'You can fix all the evil spectacles, Fairy G?' asked the girl, her eyes brightening.

'No, but I can fix your dress.' And with that she removed the girl's beautiful dress, and clothed her in another very similar one made of GIFs and JPEGs and CSS, but with an extra panel stitched into the front. 'Now when they see your dress with Evil™ spectacles, they will know what to do. And here, you'll need these.' And she handed the girl another box of magic spectacles, this time labelled 'Compliant'.

The girl raced back to the palace and paraded around in front of the princes. Sure enough, before long one or two came cautiously up to her. 'Do you perchance have any spectacles that are... Compliant?' they asked.

'Why, yes I do,' said the girl, 'Why do you ask?'

'Because you're walking around completely naked apart from this big sign down your front that says "To remove this sign, ask for my Compliant spectacles"'

And so the girl handed out her Compliant spectacles, and the princes put them on and saw her once again in a beautiful dress, and although it wasn't quite what they expected it was at least better than sackcloth and warts. And the girl made many friends, and had many opportunities to show off her Content, and discovered that most princes are superficial bastards who never stick around anyway.

Except for one, who was Good and True. And wouldn't you know it, he turned out to be wearing Uncompliant spectacles! Thanks to the Fairy Godmother's intervention he no longer saw the girl as hideously ugly, as the Evil Witch intended, but instead saw only pure Content; and yet he felt no need to ask for a Complaint pair of specs. No, he figured that naked Content with a sign up the front was good enough for him. And besides, now he knew where to get a pair if he needed them.

And so they all lived happily ever after. The End.
posted by rory at 1:03 AM on May 9, 2001

So, how long did it take you to write that Rory, or should that be Aesop? ;-)
posted by tomcosgrave at 1:38 AM on May 9, 2001

About an hour or two! So much for being first cab off the rank.
posted by rory at 2:05 AM on May 9, 2001

what normy said

normy starts off saying that absolutes are wrong and then sets out his(?) own absolute vision (which is that of one of the two groups - the eye candy group...)
posted by andrew cooke at 2:31 AM on May 9, 2001

I think it's important to distinguish 'visual elements that help usability' and 'pointless eye-candy'. The first makes for a better user interface, the second probably doesn't.

WaSP's upgrade campaign boils down to a coding style difference between: (A) XHTML (primarily DIVs) + CSS ... and (B), HTML 3.2 (tables) + HTML 4's accessibility guidelines + CSS.

Newer browsers can understand (A) - but there is a large legacy of older browsers that were built for (B).

The main benefit to (A) is the abstraction between appearance and content. This allowed print-style sheets and user-style sheets to control the layout for (A). Most older browsers for (B) could do a large-subset of the same (don't print background) and also control fonts (override with my font/colour settings). The abstraction in (A) was the best technical step as (B) was a bloated pigdog.

This abstraction was through a change in coding style that was a problem for browsers(B) as it couldn't understand the new way of stylising a page. When receiving a page coded in (A) a browser(B) got a page without colour and fonts and font-sizes and columns. It just couldn't stylise it. A page without useful interface elements like colour and fonts and font-sizes and columns... sensible use of colour and fonts, mind you, not 'pointless eye-candy' -- the "visual elements that help usability/branding" schtick.

Browsers(B) were to like it or lump it. WaSP didn't say provide good pages for those with older browsers - that wasn't the point. It was a upgrade in coding style and providing no other option was the point.

And that's why WaSP blows.
posted by holloway at 5:54 AM on May 9, 2001

Bronze Age. Iron Age. Nuff said.
posted by holgate at 6:06 AM on May 9, 2001

ps. Love the suit.
posted by holloway at 6:15 AM on May 9, 2001

holloway, could you repeat that?

The fact that this issue has come up again actually pleases me (er, except for the repetition of topicality.) Why? I feel that if you do something that doesn't cause some controversy, it's not worth doing. Some people are going to disagree on a large scale, others are going to agree, and others will simply be ignorant. The fact that the WaSP campaign forces the same issues up time and time again shows that there's something to it.

Whether you agree with it or not, you won't forget it.
posted by hijinx at 6:25 AM on May 9, 2001

hijnx: Google pagerank you say?

> The fact that the WaSP campaign forces
> the same issues up time and time again
> shows that there's something to it.

I don't know where to begin with this one. Brilliant!
posted by holloway at 6:31 AM on May 9, 2001

I don't know where to begin with this one. Brilliant!

Your attitude oozes out of this statement and really does a disservice to your argument. I'd ask you to elaborate, but I'd have to ignore your (multiple) links, and I can't do that.

Why are people so opposed to standards? Can you explain that? People are really beating up on WaSP, and I'm not entirely certain as to why. You want to use an old browser on the net? Fine, but as a webdev, I'm using the established standards. This isn't Zeldman's (or WaSP's) whim. This is the real deal.

This Zeldman post says quite a bit and is relevant.
posted by hijinx at 6:44 AM on May 9, 2001

Providing no other option was the point. And that's why WaSP blows.

Providing no other option certainly was the point. And that's why it was a brilliant tactic. This is a political campaign. Not 'politics' as in left and right, but politics as the negotiation of change. Sure, WaSP could recommend combining the import hack with traditional style-sheet linking so that Netscape 4.x still has access to the CSS elements it can use, but that would do little good in raising public consciousness and effecting change (i.e., upgrades). Only by forcing Netscape 4.x to serve up plain-looking pages can WaSP supporters hope to send a strong message to lay-users that something is wrong with Netscape 4.x, and that message deserves to be sent.

Politics is an eternal struggle between forces for change and forces for conservatism. The browser upgrade campaign is political - we're just not used to thinking that there's a 'politics of web standards'. Because it's a political campaign, calling it 'right' or 'wrong' is not a statement of fact, even if some of the facts that have prompted the campaign (e.g., Netscape 4.x's CSS support isn't too great) are widely accepted as true. Calling the campaign right or wrong simply shows what side of web-standards-politics you're on.

So, are you a browser-conservative, or are you an agent for change? Are you happy with the status quo (or at least resigned to it), or are you going to do something about it? Because if you don't, you're looking at Four More Years of single-pixel GIFs and web-design waste: design once, code twice, or four times, or more.

Sure, some users won't like your stance, whichever side you take. That's politics. Personally, I'm happy to rally behind the orange flag.
posted by rory at 6:49 AM on May 9, 2001

Not my vision, andrew, I wish I was able enough to devise entire technologies! I just happen to work in the industry. More an observation and inference of the way web technologies are (finally!) beginning to mature.

The point I was attempting to make was that we need standards because the web and its content cannot be generalized in the way that those who argue we don't seem to wish to generalize it. Its simply not a question of eye-candy (whatever that is) = good/bad (depending on your preference). It all depends what a site is for.

Its precisely because there are no absolutes in web content source, purpose and intended audience that we need standards. Primarily because by using recognized standards the people responsible for that content can be given the freedom to concentrate on what they do best without having to worry about what they shouldn't have to.

The best of the web, however anyone may choose to define it, is made by great coders, writers, editors, data-modellers, information architects, visual designers and visionaries. How many people do you know who are great at all of these skills all at once? At the moment the web is like asking the writer of a screenplay to direct, shoot, light, record, act in and edit the final movie.
posted by normy at 6:51 AM on May 9, 2001

>> ...nobody really gives a damn how your site looks.
> Say WHAT!? Visual presentation is the KEY to usability on
> the web

The point is that in life's list of screaming priorities for the vast number of users (as contrasted to site owners and developers,) web site useability is vanishingly unimportant and web site designers are towel designers.
posted by jfuller at 7:13 AM on May 9, 2001

For those of you on the fence, read Zeldman's journey, about the redesign of ALA. He struggled with how to accomplish the political tactic of WaSP, without completely alienating non-compliant users. He isn't trying to break the community of the web; he just wants to be able to write code that performs the same way on every single browser.

He had a brilliant solution: use the import hack to fool non-compliant browsers so that they render the floating box correctly....but also create a diplomatic notice that alerts users if their browsers are not compliant.

Zeldman isn't a crazed fanatic that wants to horde control and take away web pages. We just want standards, so that we're not at the control of Microsoft and Netscape....making hacks in order to compensate for their mediocre rendering and inconsistencies. The users of the web -- you and me -- are the ones that should be in control.
posted by jennak at 7:23 AM on May 9, 2001

> Why are people so opposed to standards?
> Can you explain that?

I dislike being forced and having to deal with purposefully crippled content due to a political campaign. That hurts some users (15% of my sites) and is all about users being wrong in their choice of browser and believing that the users have the skills to understand that. Users just want to use a website.

It's not about standards. It's about browser adoption of standards, hence the timing.

And, please - if you're going to rally behind standards (at the temporary expense of interface) at least choose someone who understands them. Like naming a CSS class "blue" your page will still validate, you're an idiot, and the CSS guidelines get misused at WaSP. CSS is by definition optional - and hiding an upgrade message so CSS unable browsers will be display it is just another hack. Jennak agrees.

It wasn't a brilliant hack jennak (I reserve brilliance for original ideas), I used it a year prior before realising how dumb it was (mine was Lance's U3 with an @import) - and I got that idea from elsewhere. Now the one click patent..

I urge people to upgrade. Urging is all I believe in doing to people I disagree with.

Peace out. You've heard enough from me.
posted by holloway at 7:37 AM on May 9, 2001

Holloway --

I know Zeldman didn't make up the hack...I didn't say that. What was brilliant was how he stayed true to the principle of WaSP, but didn't alienate non-compliant users -- with tact and diplomacy.

A friendly note to upgrade isn't a hack, but a service.
posted by jennak at 7:43 AM on May 9, 2001

I don't know... This argument was all fine and well, but lets move on from this meaningful stuff and go back to the egoist linkwhores.

With all due respect to everyone that has gotten their panties in a bunch here, lets remember that no one is right here. No one. At all.

Lets also keep in mind that we, as the developers of the web, just want control over what we do. Doesn't everyone?
posted by vandoren at 8:00 AM on May 9, 2001

A couple of things bear mentioning.

First, ttrendel wrote, "If you are writing a research paper and embracing the web as a massive library of information, the style of presentation (aside from legibility) doesn't mean a damn thing."

This is an old saw that is definitely incorrect. There is a thread on the internet that puts text-as-research-material above other types of content, and posits that serious researchers don't want or need material presented in a pleasing way.

This is demonstrably untrue. I have done surveys and focus groups with specialist physicians who clearly indicate that they prefer the same (academic) material in a beautiful site over a plain text-only context. Moreover, it is much more difficult to deliver text in volume without a decent interface - and if you are going to build a decent interface, you might as well make it beautiful, not just functional.

Design is more than making things pretty. Look at any medical textbook, for instance - generally they are very well designed. The whole field of medical/scientific illustration (which is thriving) belies the idea that text as data is all that is required or desired. The same is true in other fields I have been involved in.

In addition - I don't think that this is or should be about control - and I don't think that's what Zeldman was on about. It's about setting a floor level that is consistent enough and flexible enough to work with effectively and efficiently. Beyond that, it's open - and things like user style sheets and whatnot are perfectly fine.

My current hat is an an internal e-Business consultant with a Big Corp., and I know that web standards are extremely important to me for business reasons. It's simply more expensive to deal with the spaghetti of older table-based code.

We hire multiple vendors to do all sorts of different web projects, and web standards are extremely important to me. If I hire a company to build a site, there's no guarantee, usually, that they're going to get a maintenance contract for the site as well - who knows where the relationship is going to go? Web standards plus solid documentation are the only way that I know to ensure that we won't have a huge penalty if we do need to go with another vendor to maintain or update a site.

The problem, though, is that even if you just develop for "compliant" browsers, there are still enough quirks in their rendering to make much of the issue moot - though we have taken a step in the right direction.
posted by mikel at 8:30 AM on May 9, 2001

I support the people that want standards - as a strictly hobbyest type web person, I would love to know that X tag produces Y result, always, in every situation, because I don't know enough to know how to design (and yes, I use that work loosely!) around the problems.

That said, I sometimes wonder if designers have much, if any interaction with web users who are not so savvy about how it all works - the people who hesitate before every click, who can't find a link unless it is underlined, who have just as much right to use the web as the rest of you.

As for older browsers, those of you who stomp your feet and declare you will not design for them, and people have to upgrade - it is all very well to say, "Upgrade your browsers" but many, many people, those accessing the web in libraries, for example, and the majority of schools, cannot control the browser they use. The computers in our local elementary school are more than 6 years old - they aren't going to run IE 6, and there isn't anyone there to upgrade them, even if they did. And we are considered lucky, that our local school has computers at all.
posted by kristin at 9:33 AM on May 9, 2001

Jeb said: Presentation is Content!

Oh, jesus. Not this tired argument again. Why do people spend so much time arguing from bad starting points? The way you're arguing, it's like comparing apples to pepper or salt to oranges and proclaiming that "salt and apples" is the wrong way to go. Or something. It's pointless. Cut it out.

There are two sides to this story: the question of separation of structure from presentation (where the document structure - and content - is associated later with a ruleset that determines the eventual presentation/display of the document) and the question of separation of style from content (where the rules that determine the presentation are not embedded in the document contents itself). The two questions are orthogonal at best, marginally related at worst.

The former is all about having a workflow process from "permanent" storage format (e.g., structural markup around content, such as an SGML document instance) through transformation processes (e.g., Perl scripts, XSL, DSSSL, FOSI, et al) that result in a presentation format or "display-only" format (e.g., to Postscript or HTML or XML or RTF) that is suitable for display on a user's viewer/browser.

The latter is a question of how well abstracted the ultimate display format is - does it contain embedded processing/display/formatting instructions? Are they all embedded, or is there a mechanism where an external stylesheet is combined with embedded markup or keys? Are they all abstracted, where the naming scheme used by the markup in the final document is then used as a key to build a map between element types and stylesheet rules? And so on.

If you don't know anything about the roots of "separation of structure from presentation" in SGML and other more advanced markup language systems, or what they were trying to rid the world of (e.g., troff, TeX, Postscript, et al as storage, rather than final presentation, formats) then you may want to do some research into the topic. It's an interesting world. And it may appeal to your "programmer"'s mind to have the terminology and problem domain correctly mapped out before you begin to spout statements about them.

As for the article by "disenchanted", all I can say is that I can't be bothered to read anything that looks this bad in my browser.
posted by schampeo at 9:48 AM on May 9, 2001

That said, I sometimes wonder if designers have much, if any interaction with web users who are not so savvy about how it all works . . .

Exactly what I was wondering, kristin. I'm one of those people. As are pretty much almost all my friends, all of whom like to diddle around on the net.

I can't really speak to any of the esoteric arguments floating around this thread (most of it is complete gibberish to me), but one thing is easy enough to report from the land of the techless schmoes where I live: simply speaking, if a site is (a) incompatible with my browser (I use IE 5.5, but I know many who have old IE or NN 4.x--and why not? Upgrading is intimidating, time-consuming, and fraught with disaster, i.e. "Wha--what about my bookmarks?" "What do you mean, I have to reboot?" "What the hell are these system requirements about now?" "What if it just doesn't work? I have uninstall, reinstall, and pray?"), (b) so outlandishly ugly and broken that I can't read it, or (c) just plain hostile to the eye, I (and most people I know) will never go back. Why should we?

Now, if that's something you can live with, y'know, that's fine. It's no big deal for us to hit the "Back" button or run to our bookmarks either.

I can understand designer's frustrations with outdated technology and the ostensibly uneducated people who use it. What frustrated designers should also try to understand is the commonplace reaction upon encountering a site that tells you your browser is a piece of garbage: Oh, great. Another geek lording it over me with his superiority. Screw this. I'm sorry if that sounds harsh. But I see it every day; as I said earlier, I barely know what I'm doing, and who is the person in the office that everyone runs to when something goes hooey on their computer or they want the newest RealAudio version (yeah, I know)? Yeah. Me. There's your terrifying thought for the day.
posted by Skot at 9:53 AM on May 9, 2001

As ever, I think people who are really, truly against the Web Standards push should code all of their pages with <marquee> and <layer> tags. And no doctype specification.

Anything more is hypocrisy.
posted by anildash at 9:56 AM on May 9, 2001

hello and greetings, this is my first post here at metafilter. Bear with me if I screw up the tags part.

many, many people, those accessing the web in libraries, for example, and the majority of schools, cannot control the browser they use. The computers in our local elementary school are more than 6 years old - they aren't going to run IE 6, and there isn't anyone there to upgrade them, even if they did.

I think standards are a good thing and I do support them, but this very issue is one I have a hard time with. The newer browsers, on Mac platform anyway, require the most recent operating systems, which require bucketloads of memory, which require upgrading or purchasing a new machine altogether, which can require hundreds of not thousands of dollars. I personally can't run Netscape 6 on my machine and I don't see being able to afford to upgrade my existing system anytime soon.

I'm thankful I can run IE 5.5, but if I hadn't come into this particular machine a while back for cheap I'd still be forced to run 4.x browsers. And it makes me wonder, how many folks out there are using much older machines than mine?

On the other hand, standards have to be implemented at some point. Imagine having to purchase a Sony CD player to listen to Sony's artists, a Universal CD player to listen to Universal's artists, an AOL CD player to listen to Warner's artists, etc. There's just no way this kind of scenario can work over the long term.

Them's my thoughts.
posted by zodiac at 10:34 AM on May 9, 2001

my $.02...

Every page should work well and look good on a G4 with the cinema display and ie 5, an old win95 pc running netscape 4, an old 68k mac running icab, a terminal running Lynx, a Palm using AvantGo, a WinCE device using Pocket IE, a WebTV, an Internet phone and my toaster with an IP address. It doesn't matter how exactely you do it, and whether or not they look identical on each device (or even between ie5 and ns4) just so long as they look presentable on everything-- designing for a 160x160 (or 320x320) Palm screen is going to be rather different than designing for a 1600x1200 desktop. If I'm using a WebTV (or BeOS box, even) I don't want to not be able to see your site. But expectations have to be changed so that the same page will look different on a different device, just so long as it works, everything is in its right place and looks good.

This is why the web is going to become harder to design/build for. You have multiple targets to build for and they're all different.
posted by andrewraff at 10:38 AM on May 9, 2001

Reaching all the way up to what hijinx wrote at the beginning of the thread, Disenchanted is meant to degrade "rather nicely" without CSS. (Where "rather nicely" in this context means, to me, "no presentation but the browser's default.") We began using CSS for layout (rather than tables and hacks) back in December of 2000.

By extension, don't confuse using CSS for presentation with an endorsement of WaSP, or the use of tables, marquee, layers and so-on as the only way to disagree with them. The old hacks are dumb. Let's purge them now.

But the article isn't about TABLES vs. CSS, and the idea that one must reject all of a platform's principles in order to reject only one of those principles is wrong and reminds me of religious zealotry. "If you don't believe in God, then you must believe it's okay to kill, commit adultery, and covet thy neighbor's wife" etc. etc.

WaSP's upgrade campaign is not only about adopting standards, it's about deciding what's best for the visitor as if you even had that authority. You don't have that authority. That's the point of the article.

And as a footnote, to schampeo, I compared the use of italicized and non italicized summaries. I found I liked the later and made the change permanent. Hope nobody thinks we've gone "Ministry of Truth" on you. I don't know if that's the element which you thought was ugly, but I know I like it better, now. Maybe one day browsers will make it easier to toggle styles on and off on the fly. A nice big button next to "Stop" would do it.
posted by wenham at 10:44 AM on May 9, 2001

Maybe one day browsers will make it easier to toggle styles on and off on the fly.

I've got that in Opera 5 right now! Except it's next to the show images button.
posted by hijinx at 10:48 AM on May 9, 2001

This is why the web is going to become harder to design/build for. You have multiple targets to build for and they're all different.

I don't think so.

The designers of the browsers do the work of making standard web pages display optimally in their software or devices. That's their problem, and it's made easier to solve if you keep your pages compliant and the presentation well separated.

No firm has a budget large enough to test their site on every possible device - from handheld computers to jacket-lapel embedded screens and whatever other wacky stuff is invented. But we don't have to. It only becomes our problem to test everywhere if we're writing code that only works somewhere.
posted by wenham at 10:49 AM on May 9, 2001

Worker productivity declined for the first time in years last quarter, probably because the phenomenon of web designers doing everything three or four times to support multiple browsers finally caught up with us.

Zeldman is right on the money. While there are a few times when you cannot separate content from presentation--like Tufte's favorite example of the historical map showing Napoleon's retreat from Russia--in most cases, you can and should.

You go, Wasp.
posted by steve_high at 11:59 AM on May 9, 2001

Someone who can say that web pages ought to look good on any device, such as WAP, and then say out of the other side of their mouth that treating ancient versions of Netscape like WAP is bad, just isn't thinking things through properly.
posted by dhartung at 1:09 PM on May 9, 2001

No one's doing that, dhartung, are they? Certainly not Zeldman et al, they're doing the opposite. The whole point of standards is to enable alternate rendering engines. Well, not the whole point - but a goodly part of it.

You can't have alternate rendering engines if a) they don't conform to published standards when developed and b) websites don't develop with reference to those standard methods.

WaSP is just trying to push things along.
posted by mikel at 2:11 PM on May 9, 2001

Yay for Zeldman. I'm totally in his camp on this one.
posted by lizardboy at 2:44 PM on May 9, 2001

Sooner or later, the bulk of websites will be written in XML and their presentation will be handled via CSS or XSLT. Valid XHTML sites will keep working (browsers won't stop supporting them) and valid HTML 4 sites will keep working (browsers won't stop supporting them for a long time, if ever). XML-based sites will simply be capable of doing more than (X)HTML sites, for those who require that power.

But INVALID HTML sites - and *most* sites on the web are built with invalid HTML - will eventually stop working in browsers. WaSP thinks developers and site owners should know this and prepare for it by rethinking their construction methods. You may yell at the weatherman when he says it's going to rain, but the rain still falls.

With or without WaSP, the web will continue its transition from a browser-specific, hacked-together visual medium to one that holds true to its inventor's vision. The way we've been building sites will stop working. The rain will fall. WaSP wants to make sure you have a raincoat, i.e. a transitional strategy. How far you take it and how quickly you implement it is up to you and the needs of your audience. WaSP believes that, at minimum, new sites should validate to HTML 4.01. The rest is up to you.

Sure, we're agitating for quicker transitions where appropriate, since that will influence browser makers to hold to the course of implementing web standards, and help persuade toolmakers (Adobe, Macromedia) to support these standards instead of (or in addition to) churning out non-valid 4.0 browser markup and code. But there is a difference between persuasion and force. Persuasion speaks to the like-minded and to those on the fence. It does not force those who are opposed to do anything they don't wish to do.

But what about "the user?" Aren't we forcing "the user" to change?

Interestingly, Netscape 4 readership has gone UP since A List Apart turned off stylesheets for 4.0 browser users. These readers do not care that they are missing the site's layout. They find it EASIER to read a non-styled ALA than the old half-CSS, half-table-layout ALA that displayed so poorly in their browser.

No one has been harmed by ALA's adoption of valid HTML and the transition to a CSS layout. Inevitably, most of the current Netscape 4 users will upgrade to a more conformant browser. Those who can't or choose not to will continue to read the content, which is their primary concern. As a piece of web user persuasion, the ALA redesign failed. But it may have succeeded as a model for other sites whose creators are eager to stop using hacks but fearful of alienating readers.
posted by Zeldman at 3:38 PM on May 9, 2001

"No one has been harmed by ALA's adoption of valid HTML and the transition to a CSS layout. Inevitably, most of the current Netscape 4 users will upgrade to a more conformant browser." - Zeldman.

"Upgrades are likely to occur, but it's important that designers allow the visitor to upgrade on her own schedule and not anybody else's." - Disenchanted

It's uncanny how much WaSP and Disenchanted's views have in common.

Where they diverge seems to be with how we treat the visitors. Visitors are not pawns in our campaign to have browsers and other tool vendors support standards.

I don't think the DOM sniffers and hidden DIVs come across as a way to persuade visitors, so much as it feels like another twist on the "white man's burden".
posted by wenham at 4:39 PM on May 9, 2001

Wenham: That's it exactly. The WaSP plans to invade Africa and give disease-carrying blankets to Native Americans as a gift from the Great White Father in Washington.
posted by Zeldman at 4:50 PM on May 9, 2001

I think that Chris's point wasn't that web site designers shouldn't start using modern encoding on their web sites. (He himself is doing so on Disenchanted.)

I think the point was that they shouldn't be trying to coerce users into doing an upgrade of the browsers they're using.

Server side, knock yourself off. Have a great time. But client side, keep your fingers to yourself.

The problem here is that the site designers in WaSP are trying to treat the web like a PDF file, where the creator of the file has perfect control and the delivered experience is identical to all users. Why should the web follow that model? Why should web site designers have the expectation that their site will always look the same for everyone who views it -- and why should they care?

There will be a few sites where the actual point of the site is to deliver an "experience", where web design is art rather than function. However, such sites are also not going to be critical or important; they'll be showcases to demonstrate the creativity and competence (and ego) of the designer.

But for the vast majority of sites, it's the content and not the form which is of primary value to the visitor, and as long as the content is presented in a fashion which is not outright ugly, to the point of making it difficult to derive that information, then the web site has done its job. This isn't fine art, it's industrial art, and industrial art has always been first and foremost utilitarian. That doesn't mean it can't also be beautiful, just that when beauty and utility clash, utility wins.

Take font choice, for example. I happen to despise serif fonts, and one of my ongoing gripes has been web sites which hard-code the use of a serif font. A few days ago I started using a proxy called "Proxomitron" which filters HTML as it's coming in, and one of the rules I'm using now is to delete all places where specific font names are forced by the web page. Which means that I see everything in Tahoma, which is as I like it. And maybe sometime I'll decide to switch to MS Sans, or to Verdana, or to something else -- and everything I view will switch.

Why does that rumple the fur of a web site designer? Why should a web site designer be thinking "Dammit, I designed this site to use Times New Roman, and that bastard viewer is using a sans-serif font to look at it. It doesn't look right that way; it doesn't coincide with my image. Grumble." Why does this matter? The point is that I can view the information being presented, and for me at least a sans font is easier to read. I find serifs distracting.

There are a lot of reasons why people choose the obsolete browser they do: inability to upgrade (e.g. against corporate policy, or because a new version is not available for their platform, or lack of a data pipe for a 12M download), fear of upgrading, use of a minority browser for political reasons, or simple apathy. That is their decision to make. No-one is trying to say that web page designers should go out of their way to make the page look the same for every silly browser in existence -- on the contrary, the point Chris was trying to make was exactly that the idea of uniform presentation itself should be abandoned.

Zeldman, your comments here about server side miss the point, and fall into precisely the intellectual trap that Chris pointed out: "If you disagree with anything we say, then you must be disagreeing with everything we say, so we'll ignore the part you say you don't like and make our case about something else instead." Chris never said that people should not use XML, or not use CSS (he uses CSS himself on Disenchanted); he says that you should not be nagging people to upgrade their browsers. Not even "gentle reminders". It's none of your business what browser your viewer is using and you need to get over the idea that you as a page designer will have -- or need to have -- full control over how the page will appear on the screen of the viewer.

You are not an artist. You are a packaging designer. I don't buy a loaf of bread so I can admire the sack it comes in, and I don't actively pay attention to the visual beauty of most web sites I visit.
posted by Steven Den Beste at 4:54 PM on May 9, 2001

Zeldman, you make an excellent point with your sarcasm. It hasn't been an unforseen biological disease infecting Africans, but an unforseen disease of hidden costs, bugs, and broken dependencies affecting many users who've tried to upgrade.

Every upgrade costs something. Let's stop preaching (or "persuading" if you like) and trust the visitor to understand how the weather is changing and make the transition to better software when they're comfortable.
posted by wenham at 5:05 PM on May 9, 2001

I really don't know what to say on this one. I can't say i'm a fan of webstandards, but I can't say i'm a fan of CSS either.

Although I do upgrade my browsers, I won't be doing so for a long, long time. This is because my internet connection is slow and expensive. Not all areas of the world have super-fast internet connections at affordable rates. Also, my hard disk drives are relatively small. I hear my friends talking about their 70GB monsters that they have sitting inside their computers, when all I have is a rather small 1gb in mine.

To me, Internet Explorer 4 is NOT old, and whenever I get a site popup onto my screen that says "Your browser sucks, we will now redirect you to an obnoxious webpage that tells you to upgrade it" I just click the close button. Another website that doesn't get visited by me. What they SHOULD do is redirect you to a plaintext version of the site, and leave a tiny tiny footnote saying "For Shiny-Spangly Style and layout why not upgrade?" This will leave many older browser users less bitter for sure.

Fairvue Central does this, as my Internet Explorer 4 gets a plaintext version of the site (even though the box saying upgrade is large and painful to look at)

However, I have encountered many site which consider Internet Explorer 4 an antique and dump you onto the webstandards page. They lose a hit.

In my personal opinion it is better for someone to be able to see your site without the ritzy glitzy extras than not at all, and plaintext must surely be supported on every platform. So why do many sites find it difficult to create a plaintext version of their site as well as a site the way they would like the viewer to see it?

Its just beyond me.

As for my dislike of CSS... well, whenever I see a badly created HTML only page, I can still read it, theres never going to be some stupid box floating ontop of the text, even though the text may be segmented or completely abused. However whenever I see a badly designed site with CSS, text disappears, images are all over the place and the whole site is generally unreadable.

And just incase people are thinking 'heh-heh this guy has an old browser, its to be expected', think again. I also have Opera 5 (and Netscape 4.5, Opera is infact my first choice of browser due to its speed), and it is unbelieveable how many sites are unreadable in opera due to CSS code. Even Opera 5 is considered to be a browser compliant with WebStandards.

Yet, some of these slopily coded sites, look just *great* in Internet Explorer 4, because the person who created the site tested it with Internet Explorer, but never bothered with other browsers. Ironically Internet Explorer 4 is an antequated browser, according to many sites. Its interesting to note that the Older browser which should be upgraded displays the site correctly, as the designer intended, yet the newer browser displays it as a mess.

Well thats my rant on the whole subject, yes I hate CSS, yes I hate upgrading my browser. I know some people *love* CSS, and I know some people *love* having the most up to date software on the web, but I'm not like that, my opinions and views differ to that of the next person, and whenever I get forcefully moved to a webstandards page it feels like "You're too old to be allowed to view this content, but if you wanna upgrade, waste 24 hours of your life attempting to download this large installation, only to have your system riddled with new bugs".

Ahh shut up Frieza, you're rabbiting on now.
posted by Frieza at 5:05 PM on May 9, 2001

I don't know about the rest of you, but I've always designed my sites to work best in every browser and to degrade gracefully in any case. A uniform experience is of the utmost importance in everything I do. I've considered my ability to "hack" (as you put it) HTML one of my talents since I started designing sites more than six years ago, and I won't apologize for ensuring that my sites look uniform whenever I can.

I like Netscape, as it performs many functions that Internet Explore and/or Opera CAN'T and WON'T. My primary browser is still Netscape 4.7, as version 6 is a hideous mess. I would rather use a browser that works the way *I* want it to than be forced to adapt to one that renders better but performs worse.

As for sites that have starting touting the WaSP "This site would look better..." bullshit, they've slowly starting dropping off my list of daily visits. I can do without the rhetoric being shoved down my throat, thanks.
posted by Danelope at 5:09 PM on May 9, 2001

But INVALID HTML sites - and *most* sites on the web are built with invalid HTML - will eventually stop working in browsers.

Say what?

Most sites on the web are build with invalid HTML. But browser makers are going to make browsers that don't work with them? Starting when?

Sorry, I just don't see this one. The basic rule in HTML is be strict in what you create and forgiving in what you accept. I just don't see the browser makers getting rid of the ability to parse bad code so long as so many sites out there are made of it. There's nothing in it for them; a new browser that breaks the majority of sites on the web is going to have a lousy pick-up rate. IE 5.0 Mac and 6.0 Win have that subtle little switch (is there a DOCTYPE? Does it have a URL?) in them to determine whether or not they parse the site with the strict engine or the bug-compliant engine. I don't see them dropping this approach at any time when so many sites use non-compliant code. It just doesn't make sense.
posted by geneablogy at 6:31 PM on May 9, 2001

Web designers ... shouldn't be trying to coerce users into doing an upgrade of the browsers they're using.

And politicians shouldn't be trying to coerce voters into improving tax systems, improving social security, and providing universal health care. Or dismantling social security and privatising public hospitals. And so on.

Server side, knock yourself off. Have a great time. But client side, keep your fingers to yourself.

Keep yer hands of my tax dollars, bub! Don't tread on me!

The site designers in WaSP are trying to treat the web like a PDF file, where the creator of the file has perfect control and the delivered experience is identical to all users.

Bilge. The aim is not that the delivered experience be identical for all users, but that it be adaptable and predictable. Visit my weblog in a 5.x browser and you will see full-screen graphics and tailored layout; visit it in Netscape 3 or 4 and you'll see only the header graphic, and a white page full of Times text. Visit it on a WAP phone and you'll see only the words on a tiny display. Fine by me. If that's the way you want to read it, I have no problem with that.

What I do object to is that when you visit the same page in Netscape 4.x without the benefit of the import hack (as I've done when testing it out) you'll see the text trailing down the left in a one-inch wide column because of bugs in their CSS engine. That's not how I want it to be seen under any circumstances, because it makes me look clueless, and it's not how the user wants to read it, either. I would much rather that my pages suffer graceful degradation than ungraceful detonation. The former is easy to achieve with CSS. The latter is difficult to avoid without it.

You should not be nagging people to upgrade their browsers. Not even 'gentle reminders'. It's none of your business what browser your viewer is using...

Which misses the point. The browser upgrade campaign is not primarily targetted at some amorphous mass of Joe Users. It's aimed at web-designers, as JZ himself points out above. Who reads ALA - your Great Aunt Mabel? No: web-designers do. (Unless your Great Aunt Mabel is also a web designer.) The campaign is about paving the way for a future where the web has no place for the old hacks, and that time is coming fast - but the longer that web-designers keep ignoring CSS and using the old hacks, the less time they will have to adapt to the new world when it arrives. And when Joe User sees a mess of webpages that don't work in the copy of IE8 that comes bundled with Windows XP3, who's he going to blame?
posted by rory at 6:45 PM on May 9, 2001

I just don't see the browser makers getting rid of the ability to parse bad code so long as so many sites out there are made of it. There's nothing in it for them; a new browser that breaks the majority of sites on the web is going to have a lousy pick-up rate.

I just don't see hi-fi manufacturers getting rid of the ability to play records so long as so many people out there own so many of them. There's nothing in it for them; a new all-in-one stereo that won't play the majority of recorded music in existence is going to have a lousy pick-up rate.

(Bought any cheap all-in-one stereos with a record player lately?)
posted by rory at 6:49 PM on May 9, 2001

What they SHOULD do is redirect you to a plaintext version of the site, and leave a tiny tiny footnote saying "For Shiny-Spangly Style and layout why not upgrade?" This will leave many older browser users less bitter for sure. Fairvue Central does this, as my Internet Explorer 4 gets a plaintext version of the site (even though the box saying upgrade is large and painful to look at).

This is, for what it's worth, the approach recommended in the Browser Upgrade Campaign, and it sounds like Fairvue is following it the letter, as are so many of us. If a designer uses CSS for all aspects of display and layout, and your browser has CSS switched off (either through the import hack or by your own choice), you will get a plain-text version of the page. Which, as you say, is hardly something to feel bitter about. (You might even prefer it, if you're the type who switches graphics off anyway. It'll sure look better than a table layout full of broken-download icons where 1-pixel GIFs were supposed to be.)
posted by rory at 7:04 PM on May 9, 2001

Tiny Tiny Footnote is a different thing from huge massive text box at the top of the page. But thats a real trivial thing. The fact of the matter is that there are quite a few websites that redirect you straight to webstandards, and don't bother with the text only version. Its evil!! Evil I say!!
posted by Frieza at 9:17 PM on May 9, 2001

Frieza: "The fact of the matter is that there are quite a few websites that redirect you straight to webstandards, and don't bother with the text only version."

And is the above phenomenon what the WaSP aimed at? or is it simply some thoughtless people's doing?

As far as I remember, I've seen quite a few warning lights published after the launch of the BUI saying things like "be responsible if you're going to use the DOM sniff and redirect old browsers" or "better than send people to the WaSP upgrade page, build your own". I never heard "hey all you people out there, shut out of your sites people with old browsers, and send them to us."

This looks to me as a general misinterpretation of the BUI. A regrettable one too.
posted by Tara at 12:29 AM on May 10, 2001

There is no point in arguing religion. But ... to those who believe that browsers will never stop parsing broken sites, or never stop supporting proprietary, legacy coding practices, open Mozilla or Netscape 6 and try visiting a DHTML site designed between 1998 and 2000. Unless the site has been recoded to support the W3C DOM, it will fail.

That's because it was designed to support either the IE4 DOM or Netscape 4's LAYERS. The site does a quick sniff, decides you're using Netscape, and sends you Netscape 4 proprietary DHTML, which Netscape no longer supports.

Do you really think this won't ultimately happen to all the legacy code? Do you really think browser makers will worry about supporting sites that haven't been updated since the 1990s?

Eventually, in order to parse XML and fully support the DOM (and do so in a browser that takes under an hour to download) their rendering engines will run out of room for proprietary code and broken markup, and badly authored sites will stop working.

Let's look at it this way. I'm willing to bet that browsers will continue to support sites authored to W3C recommendations. You're willing to bet that browsers will continue to support sites authored to the quirks of old browsers (or simply ineptly). Think like a client. One developer authors to W3C recommendations. The other tells you it doesn't really matter, because browsers will always parse "any" code. Who would you trust with your money and your brand?
posted by Zeldman at 12:59 AM on May 10, 2001

Again, you're missing the point. No-one is saying that when you code a site that you should use old coding practices.

USE MODERN CODING PRACTICES! DO IT! We all agree with that! (Happy now?)

But don't nag people to upgrade their browsers, and don't lock them out if they don't.
posted by Steven Den Beste at 5:47 AM on May 10, 2001

Locking out users was never really an option. Some careless individuals have used it, but I doubt that they are listening to this discussion.

As for "nagging" people, you say "nag"; I say "educate." The problem with your perspective is that people will never know that they should upgrade their browsers to experience a better Web. Such knowledge does not demand action, but without it, how can anybody make the choice to upgrade or not if they are not aware of the benefits?

Clearly, you, everybody here, and maybe, most Web designers know about the problems. Some savvy Web users might also know. However, what percentage of the Web population is that...1%...maybe, 5%? It certainly is not everyone. How do you expect the rest of the Web population to find out the benefits of this new Web that we are building with "modern coding practices"? Solve that problem constructively, and I will leave you alone.
posted by draca at 7:36 AM on May 10, 2001

draca, SDB doesn't have a solution. If he did, I'm sure he would share it with us in yet another multiparagraph novella designed to so completely overwhelm us with his monstrous intellect that we would never even think of disagreeing with him again. In fact, I've been so overwhelmed! I completely agree! If you're trying to create any kind of aesthetic visual design in web pages, YOU ARE WRONG, because the only thing that matters is the raw information dump! And we're all engineers, so we can all read pages and pages of black-on-gray sans-serif text without going blind! We shouldn't educate those web users about their computers, or better options available, because then they might get ideas above their station! The only solution for the web is total ignorance and anarchy!
posted by darukaru at 8:57 AM on May 10, 2001

First off, I would like to apologize for that post, since I was just reading WinerLog and it got to me.

Secondly, I don't have a solution either. But I'm not telling the people who *are* trying to work out solutions to fuck off. And ignoring the problem (that people are coding hackish HTML that barely works in today's browsers and won't work at all in tomorrow's), or just banging a 'MODERN CODING PRACTICES' drum isn't going to make it go away.

Web users need to be educated about how the Web works and what needs to be done to keep it functioning smoothly for everyone, and the BUI is one component of that education.
posted by darukaru at 9:05 AM on May 10, 2001

As for "nagging" people, you say "nag"; I say "educate."

Haw. You know what you call "educating" people on a topic that they didn't ask to know about, don't particularly care about, and don't understand anyway?


I'm not telling people who are working for solutions to this to fuck off either. What I am trying to say is, the average user doesn't have the blindest idea of what you're talking about, and being told that your previously-just-fine browser is now unacceptable is regarded with roughly the same appreciation as being told that your car is being recalled. Is it really so tough to understand that users are looking to engineers and designers to work something out that doesn't require them to know "how the web works"? I don't advocate this position, mind you--I myself like to know how things work--but a lot of people do. These are the same people who drive cars and know only where the gas and oil go in. You may not like it, but don't be surprised with the resistance you get when you try to "educate" them.

On the other hand, I use Microsoft Word all the time. I have no idea how it works, and don't particularly care how it does. But like most people, I sure do hate that irritating paperclip "buddy." (So I killed it.) If it jumped up at me one day out of the blue and said, "You know what? This program is so awful, it can't possibly write you a letter any more. Sorry." I think I'd go ape. So maybe I am one of those people.
posted by Skot at 9:28 AM on May 10, 2001

How do you expect the rest of the Web population to find out the benefits of this new Web that we are building with "modern coding practices"? Solve that problem constructively, and I will leave you alone.

You're trying to find a solution when there was no problem.

Using standards does not cause any problem, and fearing that "the world won't know they can have a better browsing experience" is also not an issue.

You shouldn't assume that you have to educate these poor, helpless viewers who obviously don't know what's best for themselves.
posted by wenham at 9:50 AM on May 10, 2001

Just a thought... since when was education distributed solely to those who ask for it?

Look at schools, press campaigns to draw attention to ecological problems, AIDS awareness operations... should we really wait for people to ask for education before bringing it to them?

Education always serves a cause - strong meaning of the word. There is always some idealistic thinking behind it. It gives education an evangelistic tinge, but doesn't turn it into nagging for that matter.
posted by Tara at 10:12 AM on May 10, 2001

I agree that seamless web design, based on transparent degradation, is the Gold Standard. I agree that it is somewhat rude to interrupt the seamless transmission of data with a message that breaks the fourth wall and reminds readers that they are using a browser. I agree that it is UNUSUAL to interrupt the reader's experience this way, and to ask the reader to think about technology. I agree that most readers will not want to do that.

You see, I agree. I am quite aware that the method WaSP has advocated is unusual. We feel that circumstances warrant the gentle interruption. You are free to believe otherwise, and I can respect that. I will not insult you for your beliefs.

Until February of this year, I took pride in the fact that my sites worked in any browser. In fact, with one exception (my business site), my sites still work in any browser. They just don't look the same in every browser.

But the problems with 4.0 and older graphical browsers go far beyond issues of visual display. The old browsers have accessibility problems and are limited in the functions they can perform. They expect proprietary code. They can't handle standard code because they weren't built for it. Increasingly, websites are being authored to code these old browsers can't parse. Sometimes that means the DOM. Sometimes, unfortunately, it means newer browser-specific code (e.g., code that works only in IE5/Windows). WaSP hopes developers will stick to standards and not code to the proprietary quirks of old OR NEW browsers.

Developers will not keep building 4.0-browser-friendly sites forever. Clients look at their referrer logs and see that 85% of their visitors are using IE5. Clients want advanced functionality. Developers get paid to deliver that functionality.Sooner or later, broken, buggy 4.0 browsers will get written out of the equation, and people who use those browsers will be left high and dry. In my mind, it is a public service to let folks know they have choices. Again, you are free to disagree.

You may argue that no one will be left high and dry; they'll simply visit someone else's website. Possibly. But what happens when the site that doesn't work in their browser is MP3.com, or CNET, or Microsoft.com, or TVGuide.com, or Sony.com?

One more thing. If you're not convinced that 4.0 browsers (and Netscape 4 in particular) are broken with regard to web standards -- if you prefer to believe that I am simply a weirdly fussy visual control freak -- I ask you to use Netscape 4 to view the PREVIOUS ISSUES department of A List Apart. The page consists of an HTML table (because it contains tabular data). The table cells are styled via CSS. The CSS validates. The HTML validates. The JavaScript that runs the category switcher is bug-free and should work in any JavaScript-capable browser.

In some versions of Netscape 4, THE PAGE DOES NOT DISPLAY AT ALL. This varies by incremental upgrade number (4.73 is completely different from 4.74) and by platform. In no case is the display accurate or aesthetically pleasing, but I don't care about that. What I'm talking about is that in many versions of Netscape 4, this perfectly vaild page cannot be viewed. That is not by my choice. I can't fix my markup because it's not broken. What's broken is the browser. I feel an ethical obligation to let readers know that their browser doesn't work. I understand that many of you feel otherwise. Let us disagree amicably. Life is short.
posted by Zeldman at 11:00 AM on May 10, 2001

schampeo: As for the article by "disenchanted", all I can say is that I can't be bothered to read anything that looks this bad in my browser.

So, first off, you're arguing without reading the article in question. Maybe if you had actually read it before posting your unnecessarily caustic reply to my post, you would realize that the article isn't talking about separations of separable content (such as text) from the rules that describe its presentation. That makes good design sense. I acknowledged that.

In the article, though, the author criticizes the very goal of being able to achieve consistent presentation across browsers, saying that it shouldn't matter if content is properly abstracted from presentation. What I'm saying is that valuable information is lost in this case, information that relates to the way information is presented. Take away that power, and you weaken the medium.
posted by jeb at 9:27 AM on May 11, 2001

That power was never taken away, it just hasn't been granted by everyone - usually by their conscious choice.
posted by wenham at 12:34 PM on May 11, 2001

« Older "We’re not behaving, sometimes we’re pretty wild."   |   Newer »

This thread has been archived and is closed to new comments