Yet another CSS slam
I guess linking to CSS rants is last week’s news. After all, plenty of people seemed to be linking to Dave The Dinosaur, going back to his font tags. And lots of people linked to old-schooler jwz, unwilling to give up his tables. But I just haven’t seen any links to this guy bad mouthing CSS. I mean, come on, he had so much trouble making a simple CSS layout that he, er, wrote XUL instead.
When it works right in a reasonable number of browsers, CSS is a great thing. Even when it doesn’t, it can be fun to play with, and use bugs to make fun of the other guy’s browser. But CSS is not a religion. It’s not all-or-nothing. If what you do is play with CSS and then write about it, great! Please, for my sake, bang your head against it, with a half-dozen browsers on two or three operating systems all going at once, and then tell us what you learned. But, if you want other people, who don’t do that for fun, to get into CSS as well, then you need to stop acting as though anyone who ever uses a single table or font tag is evil. If someone whose non-tech writing I love spends all their free time for three days running trying to replace their last remaining table instead of writing something beautiful, you’ve harmed us both.
Up next on the wheel-o-rants: people who fail to grasp the fact that XHTML+CSS is not a single word synonymous with valid markup.
I’m not so much bothered by Dave and JWZ’s dislike of CSS, but by their unwillingness to learn. Jamie’s little conniption fit seemed (to me, at least) to stem from an attempt to throw some half-assed CSS into his pages without really understanding what it would do or how it would work; one of the comments attached to his rant made the point pretty eloquently:
Both Jamie’s and Dave’s problems consistently seem to come from an unwillingness to do just that. So then when they whine about it, I feel I have every right to get annoyed; this is the sort of behavior that invariably provokes ”RTFM” replies when engaged in by lesser mortals, and I don’t see why they should be treated any differently.
But I’m not an all-or-nothing XHTML+CSS cult member, so I guess it doesn’t matter ;)
I agree. And besides, everyone knows font tags really are evil. There’s no reason, even in NS4, to use them. Yes, it’s a different mindset. No, it’s not that hard.
Right. I assume that my new template, with fully resizable fonts and flexible content areas that never overlap or drop to the bottom of the page, which works in IE5+ (Win and Mac), Opera 6 and 7, Mozilla and Camino and Mozillabirdbrowser, Safari, NN4, Lynx, and OmniWeb, is in the mail?
Hang about, all he said was that font elements aren’t any good. Please explain how font elements provide ”flexible content areas that never overlap or drop to the bottom of the page, ” and ”works in IE5+ (Win and Mac), Opera 6 and 7, Mozilla and Camino and Mozillabirdbrowser, Safari, NN4, Lynx, and OmniWeb”
I do believe you’re right: I read ”Jamie, Jamie, Jamie, jwz and dave, [CSS] is a different mindset”, not ”jwz, jwz, Jamie and Dave, [font tags are] a different mindset”.
Damn, does this mean my new design isn’t in the mail?
I cannot do CSS layout design, because I don’t have a Mac, and there are enough significant differences in CSS handling in Mac browsers that the only way I could do it is to throw up a WFM version and ask for Mac bug reports, then screenshots, then confirmation from other Mac users, then testing of changes, more screenies, bleah. So I depend on the kindness of strangers. Oddly enough, despite my asking them in the rudest way I can, they don’t seem to come through for me.
I just wish there was a FM to read: this one simple little problem (the heading, not the printing) generated all that noise, with the end result being that you need to have internalized the fact that if you want to center a heading, you need to put presentational markup inside your semantic markup. If that final comment, using a bit of javascript to insert a span inside headings so they’ll be styled right, was actually left with a straight face, then I’m in awe of some quality of his. Dunno quite what, but certainly something. I recently ripped a bunch of hacks out of my stylesheet, and just said too bad for buggy browsers, because it had gotten to the point where I couldn’t remember what did what to what browser, and I was afraid to make any changes in case the whole house of cards came down (in a browser I don’t use, on an OS I don’t have, no doubt). Maybe there is a FM that covers that, but I’ve sure never seen it before.
And much of my attitude toward the problems of CSS comes less from them, and more from some poor kid who just wants to write a weblog, but his Currency Blogger template gets weirded out every time he puts a <blockquote> in a post. I don’t think that the entry level for web publishing should be a complete understanding of blocklevel and inline elements, implied closing tags, and every limitation of Blogger’s paragraph model. I have a lot of fun playing with things like that, but the overall goal of weblogs is to make it fun and easy to publish on the web, not to separate the CSS experts from the people who give up in frustration.
I guess I don’t see what all the fuss was about; the whole heading problem was a simple little thing. Three or four people pointed out that it can be fixed in two ways: a span inside the heading, or setting the heading to
display: inline. If Jamie actually had grasped the difference between block-level and inline elements in the first place (which, let’s face it, isn’t hard), this would have been a non-issue.I agree, though, there there is a distressing lack of a manual to refer folks to; I’m a permanent amateur and what I know of CSS I’ve gleaned partly from looking at other people’s style sheets and partly from repeatedly breaking mine. The spec is only useful if you already grok CSS and just need a reference, and I know of almost no good tutorials. Simon Willison says he’s going to write some, and they can’t come too soon. And maybe, just maybe, some enterprising blogging tool will eventually put out a good HOWTO on styling. But I think what CSS really needs is a community-driven site with code samples and comments mixed into documentation, and not just the sort of ”here’s a layout, here’s how it works” galleries that are fairly common anymore — what’s needed is a comprehensive reference. Something like the online PHP manual comes to mind.
I definitely did get some helpful and informative comments after I posted that, but please note that I only posted that after having spent hours searching for comprehensible documentation and examples on the web. And I’m good at that sort of thing. So one of the big problems here is not so much that there no TFM to R, but that there are many, many, many FMs and they are all full of bad advice and outright lies.
But, to everyone who says, ”what’s the big deal, just put in a SPAN”, my response, which I made several times, was ”wasn’t the whole point of CSS to separate presentation from markup? And doesn’t the fact that I have to do that, for this simple, simple effect, mean that CSS has failed at that goal?”
And if CSS doesn’t separate presentation from markup, then what’s the point? If I still have to embed presentation in my body text, then I’m just trading one presentational language for another, and why would I want to do that? That means it’s not better, just different.
(And yes, there are things CSS can do that tables can’t, but my pages don’t need those things, so why should I care? Especially since, clearly, there are things that tables can do that CSS can’t, like work properly in browsers that are more than a year old.)
Sometimes it feels like I’m speaking some moon-man language that nobody else can ever comprehend. I don’t think the things I was saying are really very hard to understand, and so many people assign interpretations to them that are just… wacky. Do you honestly think I’d spent the time to write a rant like that without having done at least a little exploration first?
First, a disclaimer: I am an amateur and anything I say should be taken with a grain of salt.
Now, I’m not so sure you have to introduce purely ”presentational” markup to get the effect you wanted. For one thing, you could just as easily set the heading to
display: inline. For another, it certainly would have been possible to restrict the width of the heading’s box; that’s just a simplewidthdeclaration. And even if we go for thespanoption, it’s still not necessarily ”presentational”; you wanted something that behaved like an inline element, so we create one with thespan. You also want it to be structurally apart from the surrounding elements and have the semantics of a heading. So it goes inside a heading element. What’s presentational about that?And frankly, I don’t know what sort of research people do; all I do know is that I see a lot of badly-thought-out, begging-for-RTFM rants about CSS out there, and amazingly the ”problems” folks seem to be having with CSS aren’t hard to fix; in fact, they typically aren’t even problems, just a lack of adequate understanding. Yours was a case in point, and the ”CSS is BS” attitude didn’t help any; after all that ranting was done, I was left wondering whether you understood the difference between a block-level and an inline box. I still do wonder about that; it may be that you do, it may be that you don’t, but your post certainly seemed to indicate the latter.
And what’s more, rants like that are often accompanied by tired old complaints about ”CSS people” that just don’t fit with my experience. Maybe I don’t have enough experience — that’s certainly possible — but I’ve been lurking in a lot of places for a while and have yet to see the sort of thing you complained about.
Now, is the CSS visual display model confusing at times and badly in need of work? Heck yeah. I spent twenty minutes the other day trying to figure out why a background was only being applied to a narrow strip of screen until I realized that the CSS I’d specified put a lot of elements out of the normal flow and left
bodywith a computed height of about 30 pixels. That sort of stuff is counter-intuitive and needs to be fixed. Supposedly CSS3 is going to be an improvement, but I can’t make heads or tails of a W3C document so I don’t really know. Ditto browser implementations; right now most of them still suck in a lot of ways. But there was a time when tables didn’t work in every browser, either . . .wasn’t the whole point of CSS to separate presentation from markup? And doesn’t the fact that I have to do that, for this simple, simple effect, mean that CSS has failed at that goal?
Well, in my mind, it means that CSS has not yet succeeded. But success is not a binary option — it’s a spectrum. CSS doesn’t have to be perfect to be useful; it doesn’t even have to be perfect to be a good step towards separating presentation from markup.
[I only use CSS for styling fonts and colors, not for layout…]
I understand your point that to be perfect, CSS ’should’ make it possible to do what you want with just one tag, but I still count it as a win that it only takes two tags to do what you used to use a whole table structure for.
More compact, easier to read, easier to change something site-wide. This is why I use CSS for some things.
Right this heading thing is really easy to fix, without presentational mark-up, or spans or anything in the HTML
Just use CSS2 child selectors, of course they don’t work in IE but then that browser hasn’t been upgraded in an age.
The thing is that there were just as many headaches in the early days of getting tables to display your layout probably. So you might ask why go through the headaches again? well for one thing the resultant html is much more transportable, easier to work with a backend, and much easier to read and update.
Understanding CSS seems torequire the same dedication that learning HTML did 6 years ago… you have to go looking for it! but there are many very good resources, and of course the good old view source!
While I appreciate your comments, and understand your position, when you see people who are pushing back at CSS zealots referring to ”arrogance” and ”unreality” and ”complete frickin’ morons”? They are talking about people who can say ”sure it doesn’t work in IE but so what?”
There are two realities on the web: those of us who get pissed off when mozilla doesn’t release a nightly for a couple of days (or, those of you who can’t believe I don’t just build it myself), and a rather larger group who either think that the Internet is that big blue ”e”, or know or sell to people who think that. So, while ”just use CSS 2 selectors” works fine as a political statement, it’s the sort of political statement that will get you fired instantly in much of the actual, physical world.
+1.
Note, however, that JWZ seems to occupy neither of these realities. He tried to go from the most bloated form of Tag Soup to the purest form of semantic markup + CSS, and he spooged all over his blog because he had to use an extra span tag to accomodate IE 5. And he continues to spooge over here on your blog with the same crap. Jesus, get a grip. The born-again zealots are always the most annoying.
The thing that gets me is that tonnes of people (perhaps including myself) want to take up ”pure” XHTML and CSS without realising what that actually entails. They output XHTML, complain about CSS giving them fluid layouts (well, duh) and all they really want to do is create a normal HTML 4.01 page. So why not build something in 4.01 and output it as such? It’s still valid. It still works. You can still use minimal CSS if you want.
Of course, some people don’t even care about validity at all (further than ”it works in IE”). This is what used to irk me about Dave Winer, when he used (?) to run Userland. He made content management systems and didn’t seem to give a hoot about users on browsers other than IE or those struggling on teeny modems. He had a chance to change things, things that the big CMSs still can’t get right… and, like them, he didn’t bother.
As for jmz, I think a few of his complaints were a little unfair. Just like those that ”drink the CSS kool-aid” you can’t expect CSS to be able to do everything. Just because you _can_ do a certain things using elements in HTML, doesn’t mean that CSS _should_ be able to replicate this. Anyway, luckily his problem was pretty easy to solve (at least in browser(s) that actually support CSS). He got his answer after 5 hours of asking his readership (which is pretty quick, if you ask me) and it went something like this:
h1 {
background: #040;
color: #0F0;
display: table;
font-size: large;
margin: 0 auto;
padding: 6px;
border: 1px solid #0E0;
text-align: center;
}
…well, that’s my version of it anyway, but that wee ”display: table” was the important bit. As hinted, it doesn’t work in IE but then – and this is also important – that’s their bug, not CSS’s. Until the day of the one, completely valid uber-rendering engine, you either write the newest, simple valid code that only works in the smartest browsers; or write in tonnes of hacks and crap that takes account of every other browser; or code simply in an old HTML format that almost every old (or stupid) browser will understand perfectly.
Funny you should mention going back to HTML 4.01 (or, as Hixie suggests for some reason, 3.2) – I was just looking at the latest version of XHTML 2 (which you are supposed to avoid like the plague unless you have some reason to actually need the XML of it), and noticed that the new model for the scripting module rocks: <script> can include <script> and <noscript>, so you call an external script in one language, and if the UA doesn’t understand that language or can’t load the external resource, it will interpret the script that’s the content of the outer script element. It’s a really sweet looking model, and you can only get it by going to XHTML 2, because HTML is a dead language.
Ah. I hadn’t picked up on that before. That _does_ sound neat. It certainly would’ve been helpful if it had’ve been around in existing specs for remote commenting systems to work off of.
I wonder if they’re still gonna have that weirdness about having to have a block in a NOSCRIPT. The spec seems to say that it will, but I’m not the best at deciphering them.
I’ve still been toying with the idea of switching forward to XHTML 2.0 myself, not because I need to in any way, but just for something new to play with. I suppose any visitors to my site aren’t exactly going to thank me for it but, well, almost all of them are already probably wondering why everything looks so wonky…
It did look to me like noscript requires block content, but then before I tracked down whether script still has the odd block|inline definition it had before, a crash took my twenty open tabs away, and I just cried for a while and went to bed. So maybe they solved it by making script block-level, too, but I’ll have to look again to find out.
? I thought it all looked just fine, it just didn’t work ;)
Ayuh, still very weird. Both script and noscript are themselves both block and inline (), but noscript’s content is block (heading, list, block, actually, though I’m not sure why heading is separate, since it’s also in the Block Module). It’s a little less weird when you accept the Block Module’s statement about ”block” being semantic, not presentational, because a stylesheet can turn any element’s presentation to inline (or table, or whatever), but it’s still a bit twisted, that an anchor may not contain a paragraph, but if it contains a noscript, it must.
Something else to consider — I never decided to become an expert in HTML. The zealots seem to assume everyone wants to be one. I don’t. I have enough things I am an expert in. I wanted to write for the Web, and I learned how to do it. Then they told me to throw out everything I know, and I don’t want to do that. Why would I? No one has ever explained that to my satisfaction. Why can’t a tool for blind people skip over the elements of a table? Maybe the table even *helps* them figure out how things relate? Why wouldn’t a table also help a semantic web parser, if such a thing ever exists? It’s troubling when people use emotional arguments and hand-waving so much. I tend to believe they don’t have answers when they do that. Anyway thanks for the pointer to Hyatt’s piece. I’ll read it fully next time I care about CSS. Warning, that may never happen.
I’m an alpha geek, and i strongly advocate the use of CSS layouts, and not table layouts to my friends to help people who might someday use an alternative browser to view the site.
I even go to the extent of getting 2-3 hours of sleep a night while they are doing it to be able to offer my geek skills to them in converting over.
Yet even in all those, the thought springs to mind that ’tables are not depreciated’. Yes, tables are a bad thing, for layout purposese, because they break the viewing of the site by alternative readers. And this is all due entirely to the fact that such readers view a table as just that, a table with each cell read left to right, and then down a row.
Yet many blind people do not use the web, and for much the same reason that people don’t use CSS styling. They can’t understand the web. It’s so ridiculously styled and laid out that they might as well read a news paper. Now i can’t change the rest of the web, but i can do what i can to make sure that the whole of the web i do have access to works for them. THen again, as i said at the start of this, I’m an alpha, and i know all the hacks.
Most people aren’t. Even many alpha geeks get the XHTML wrong. CSS is a mess, and needs proper implimentation, but at the end of the day, it is two steps forward, one step back. Easy to write, a bitch to debug, if it ain’t your code.
I use CSS for positioning, and would advie anyone else to do the same. Then again, your mileage may vary.
Re: ”I never decided to become an expert in HTML. The zealots seem to assume everyone wants to be one. I don’t.” Well, for many years, you used to be the CEO of a company that made *web publishing software*. Is it too much to ask that you have a better-than-average understanding of the output medium?
Re: ”Why can’t a tool for blind people skip over the elements of a table?” Screen readers have special modes for navigating through tables, but it’s still very hard. Simple, well-marked-up data tables (that actually contain proper table headers and table data) are difficult enough. Tables which have multiple layers of purely presentational table cells (last time I checked, scripting.com had nested tables 4 levels deep, have you changed your layout since last summer?) and where many of those cells have spacer images without ALT attributes… well, it quickly becomes impossible to find any actual content.
No hand-waving here. I’ve used all the current generation of tools, and I can’t make heads or tails of your nested tables in any of them. Maybe some future generation of tools will be smarter, but right now they’re not. Maybe some future generation of visual browsers will all render CSS perfectly, but right now they don’t. There are parallel arguments here. A few small sacrifices to the ”purity” of markup (an extra span tag here and there, in JWZ’s example) greatly increase your chances of cross-browser CSS compatibility with the current generation of browsers. A few small changes to your templates can also greatly increase your accessibility with the current generation of tools.
I wrote about all this last summer: http://diveintoaccessibility.org/ You linked to it generously, but you never implemented even the simplest of techniques on your own site. You complain that nobody makes this stuff easy, puts it in a language you can understand. You were my target audience for that book. You, and only you. Obviously I failed somehow. Was I not clear enough? Were the explanations not compelling enough? Were the examples not detailed enough? I’ve always wondered.
All things being equal, I would have preferred silent implementation to high-profile empty gestures. I didn’t care about the linkage; I set out to change the world a little. But I guess the world will have to change without you.
Not every CEO is an expert on all aspects of his or her business. It’s impossible. Especially in a complicated area like Web content management. I relied on advice from people more knowledgable than I am on CSS and HTML.
Now Mark, a few other people are now on the record saying you’re wrong. Why don’t you assail their reputations instead of mine.
To everyone else, this is just what I’m talking about. Just being truthful gets you challenged at an integrity level. These CSS zealots are ridiculously self-important. There *are* other pressing issues in the world today (sorry for the sarcasm).
Wow, you make it sound like UserLand was some huge company with multiple layers between the CEO and the low-level designer grunts. When in fact it has always been a small independent company, where you wrote much of the company’s code yourself.
Whatever. I think my point is clear to everyone else, and I’ve learned to treat that as a victory when debating with you.
Oy. UserLand was and is a small company. Not much more to say about that. Believe it or not I try to tell the truth all the time Mark. I know that’s hard for you to believe, but you’re just going to have to get over it.
All you’re doing is lowering my Winer Number.
Moving on…
Mark would you mind telling me what point you’re trying to make? Are you saying I’m lying? If so, please have the guts to come out and say it. You haven’t really made an accusation. First you use the term CEO, then you say I’m being grandiose for using the term in my response. What is your complaint? Please be specific. I’m listening.
”I wrote about all this last summer: http://diveintoaccessibility.org/ You linked to it generously, but you never implemented even the simplest of techniques on your own site.”
I got very very sick last summer, and you know it Mark. Man this makes me angry. Do you have *any* humanity or are you just cold, angry and selfish? You need to take a look at your values, my former friend.
Take it easy, both of you, my current friends.
Mark doesn’t think of last summer as the time when you nearly died, none of us does at the top of our heads, because we didn’t see it. You’ve told us about it, but all we saw was that you took a break from blogging and then came back slowly. If you remind us, we remember, in varying degrees based on how close we’ve been to someone in the same situation, but it’s not the defining characteristic of last summer to us the way it was to you and people who actually saw you in person.
And I don’t think that Dave thinks of his HTML as a part of what he says on his site, the way we do, so much as just something mechanically generated by the code, and he cares more about the email from someone stuck using NN4 saying they can’t read his site than the (hypothetical) email from a markup fan or a blind person or a dialup user (that one’s not hypothetical, I’d send it myself) thanking him for making his site cleaner and quicker. And even though getting mad at him is one of the most popular sports in blogdom, I haven’t noticed it actually getting the person anything beyond higher blood pressure very often.
So, why don’t we just start here, now, and small? Is the most inaccessible part of scripting.com that would be the easiest to change that brutally long (except for me) list of links along the left side, that a screen reader has to read without skipping, every time? Dave: can you stick an <a name=”content”></a> (or some similar name), at the top of the center column, and a <a href=”#content” style=”display:none”>skip links</a> at the top of the left column? Those two little changes will let a blind person or someone using Lynx jump ahead to reading the content as soon as they get to the top of the left column, without having to hear your entire blogroll every time, and the only weirdness that should result is that in browsers that don’t support CSS there will be a link that doesn’t seem to do much of anything.
Phil, there’s a bigger lesson here. Yeah you probably don’t know all that’s going on in a person’s life that you know through their weblog — so don’t get personal. You’ll always get it wrong. I’ve been saying this for years. This is a great medium when people use it to communicate, collaborate, coordinate, a really shitty medium for developing personal relationships.
I agree with the other stuff in your post, and suggest that you bookmark http://avantgo.scripting.com/ or read Scripting News through an aggregator. Or better yet, make feature requests at a time when my back is not to the wall defending my integrity.
Please understand that Mark always puts me on the defensive. This is a bullshit technique, and it is not going to work.
What about people who come in from search engines? What about people who come in from external links? Creating a separate text-only edition does not address either of these usage patterns. I have never seen anyone link to the text-only edition of your site, nor have I seen the text-only edition come up in search results. (These are related; people link to the regular version, and search engines assume that’s the version that’s important. If they even find the text-only edition, they’ll filter it out as a duplicate, or at least give it much less weight because nobody is linking to it.)
RSS is fine for people who want to keep up with the latest news. That is a specific, and popular, usage pattern. Text-only ”mobile editions” are a hack for mobile users who want to download all the latest news, but don’t support RSS (Plucker, AvantGo). The only reason I have a separate text-only mobile edition is because these programs insist on spidering an entire subsection of a site, so I made a text-only ”ghetto” where they can spider away. This is a specific usage pattern too; I don’t like it, but it predates the popularity of RSS, and I support it grudgingly.
But creating a separate text-only edition of an entire site and wiping your hands of the accessibility ”problem” doesn’t solve anything at all.
If you want to make an accusation, come out with it. Make your case. What are you saying? I am dishonest? I don’t care about disabled people? What’s your point? I’m tired of you beating around the bush. What are you saying?
I guess we’ll never know why Mark mixes up his beliefs about technology and other peoples’ personalities.
He says on Dive into Mark: ”I am desperately trying to stop arguing the merits of web standards, CSS, accessibility, and open source. It is quite obvious to me that these are the future of the Internet and of the computing industry in general, and if you don’t see that by now, I can’t help you.”
Hey Mark, no one asked for your help.
The problem with specific text-only versions of sites – like yours – is that they’re usually stripped of almost everything.
If I come to your site via WAP* and choose the Avantgo version because I don’t want to manually pass 18 or so screens of off-site links (not to mention the loadtime wait, and the cost at 10p/40p per minute to do so) I find that I can only access the last day’s posts. Nothing else. That kinda sucks.
Like, at the moment, there’s only two posts. :/
Anyway, in the end it’s no big deal. It only would be if it was necessary to visit your site on the move, or some Manila site over a bad connection, or a complete horlicks of a site through a screenreader or text browser.
In reality, whether some academic on the net takes up a nice, light accessible design using CSS or what-have-you on their personal site holds only slightly more importance than if any other Joe Bloggs (ooh, nice anon. pun), teacher, trainer, geek, manager decides to use a big image-mapped graphic or some mess similar to broken HTML 4 for their own sites.
It’s really only public-facing government and corporate websites (and the people/tools used to make them) that are worth evangelising and agonising over… everything else is just a distraction.
—
(* via Phil’s blogroll – sorry, it’s always handiest, and close enough to my own – via Google’s WAP filter)
”””I have enough things I am an expert in.”””
Name one.
Jonas: thanks for commenting, but you get one flaming troll warning, and this is it. Never again in my space.
You get several XHTML warnings, though: it’s enabled, just <blockquote>, no need to ”””
The basic fact: HTML is used by millions because it’s child-simple to code. CSS should have been the same. But we all know it isn’t, because those who devised it fell in love with complexity. Simple is often a lot harder to do.
Your basic point is absolutely spot on, but I think there are some subtleties around the edges. It’s probably less that they fell in love with complexity than that there weren’t enough implementers and simple users involved.
Nobody would want a spec that I designed, because it would do very little, and have glaringly stupid aspects to it. That’s why we have brilliant people working at the W3C.
But… you know how we all talk about how awful IE6 is, and how stupid they are to get so many things wrong in their CSS support? Have you ever met a ’softie? They hire a whole lot of scary-smart people, and still can’t get it right.
When I see brilliant people like Hixie and Tantek saying that there are parts of CSS specs that are completely undefined, and parts that might be unimplementable, I really wonder if we are talking about something that ordinary shlubs like me have a chance of understanding. Of course, part of it is that they don’t really intend me to: none of the recent specs sound like something they want me handcoding, so much as something that will be autogenerated, which would be fine if I thought that they were something that could be understood well enough to even build a generator, but if it’s going to produce something that I’ve got no chance of understanding, then it damn well better be something that the producer completely, absolutely understood. And it won’t be.
Another zealot problem is the claim that CSS enhances the accessibility of web pages. This is false. Screen readers rely on html elements for the information they present to users – you can do more damage using CSS to make elements look like other elements. Screen readers know how to handle layout tables; they’ve been dealing with them for years and years.
Where CSS does come in handy wrt accessibility is once you’ve constructed a good html document, using heading levels, lists and the like, you can tweak the look and feel with CSS fonts and colors. But this is using CSS for visual tweaking of already-accessible HTML, so it’s not at all important for the accessibility of the html itself.
Certainly, using CSS poorly can do more harm than good. So use it well!
Some old but relevant articles of mine on CSS and accessibility:
Everytime I read something about the ”evil” of tables and how designers should strive to eliminate them through CSS, it makes me wonder if I am doing something wrong by using them as much as I do and in the way that I do.
I’m not trying to shamelessly plug my site here or anything like that, but I would appreciate it if someone with a little more authority on the matter would take a look and let me know if I should be pursuing alternate methods of achieving my designs.
Just ignore my ramblings if you don’t agree with them. What I am looking for here is input on the table-based design methodologies.
Thanks in advance, and sorry for using your comments section like a siphon.
The Web and the UI
I can picture a few people at Macromedia laughing out loud after reading Dave Hyatt’s views on CSS (and rolling…
Yet another CSS slam
I guess linking to CSS rants is last week’s news.