Summary
I seem to visit more and more sites where the pages don't render right, to the point of being unreadable and unusable. And they're trying to sell me something! Why is this happening?
Advertisement
Things overlap or are put in the wrong place, so you might not even be able to reach the component that you need to manipulate. Often I'll have to switch from FireFox to IE in order to use the site.
I have created the CSS for my own site and I know that -- despite all the high-flying promises of CSS -- you can get radically different behavior across browsers for even relatively simple CSS. My solution is to dumb down my CSS even more because working consistently across browsers (without a Herculean effort on my part) is more important to me than being fancy.
So one contributor is probably the failure of the promise of CSS. It's definitely not write-once, run everywhere. And if you have been drawn into believing that it is, then you'll write your code for IE and assume it works across browsers without actually testing it.
However, most of the problems seem to occur when components are involved, so I'm wondering if the promises made by AJAX (Asynchronous Javascript And XML) might also be the culprit. The idea is that someone has gone to the trouble of making all these AJAX components that should work transparently across browsers, so you don't have to worry about it. But if you consume such components without testing them, maybe they don't actually work that way.
One of my designer friends has said that I have my screen resolution set too high, and that's the problem. I'm at 1600 x 1200, but that's what my monitor (an NEC AccuSync LCD 200 VX; not an unusual off-brand) insists on. Maybe it just makes these problems more obvious, but there's no way we're all going to lower our screen resolution, making pointless the better monitor, in order to compensate for stupid web design. My designer friend is about to change to a desktop Mac with a nice monitor that could easily have higher resolution than what I have, so we'll see what he says then.
What do you think is the reason for the degradation of web-page usability?
There are way too many permutations of browsers, extensions, toolbars, operation systems, etc for anyone to really test all of them. In some ways it is too bad that Web 2.0 had to be backwards compatible with Web 1.0. It would be nice to start over.
Good CSS design isn't exactly a picnic, but that's pretty harsh. I don't want to sound like a Microsoft basher here, but if you are actually starting with IE, that could be your problem. We have pretty good success getting cross-platform consistency with CSS starting with Firefox. There is inevitably some hand-wringing and gnashing of teeth when we first test it in IE, but those problems are generally solvable.
I'm certainly willing to concede that maybe the designs I've worked with are simpler than yours. It could also be that for me "consistent display" has more inherent tolerance.... I'm not the kind of person who looks for pixel-perfect designs in my Web pages across browsers... I can cope with the occasional "Strayhan gap" between elements. The Web is not print, and never was really intended to be.
Heck, I'd probably just resort to Flash and Flex if I really needed a rich branded experience with those kind of tolerances, but for serving document-style information to the widest-possible audience give me XHTML and CSS.
> There are way too many permutations of browsers, > extensions, toolbars, operation systems, etc for anyone to > really test all of them. In some ways it is too bad that > Web 2.0 had to be backwards compatible with Web 1.0. It > would be nice to start over.
I'm not even talking about a lot of permutations here. Just the latest FireFox. Seems like that should be testable, but maybe my designer friend is right, that the higher screen resolution isn't something people test on.
> > There are way too many permutations of browsers, > > extensions, toolbars, operation systems, etc for anyone > to > > really test all of them. In some ways it is too bad > that > > Web 2.0 had to be backwards compatible with Web 1.0. > It > > would be nice to start over. > > I'm not even talking about a lot of permutations here. > Just the latest FireFox. Seems like that should be > testable, but maybe my designer friend is right, that the > higher screen resolution isn't something people test on.
Screen resolution is just another set of permutations. :)
With good design (and thanks to CSS), screen resolution is not a problem.
Here in Russia there are lots of proponents of 'rubber design', when everything resizes to accommodate a bigger / smaller screen. Try resizing this: http://lenta.ru/
Oddly enough, many well-established Western sites have design set in stone; try resizing this: http://www.bbc.com/
As for resolution, not only 20+ inches monitors start to emerge everywhere; 3-5 inches monitors do so, too. All these net-connected mobile phones and PDAs are getting real browsers (like Mobile Opera). Making a site that looks good at 1600x1200 and is usable at 320x240 takes a good web designer :)
I've noticed this problem for a very long time. In my case, I run the native resolution of my large monitors. However, I also increase the font size whether I'm on Linux or Windows. (I LOVE Gnome's font display!) I also have to increase the minimum font size in Firefox, and this seems to be where the problems really start appearing for me. I don't think very many site designers consider users outside of their own demographics, which is "usually" the younger crowd using high resolution with very small fonts. Those of us a little older or with failing eyesight can't use those smaller fonts.
I love having a large monitor with high resolution for my programming work. I do have to put up with inconsiderate site designers who don't account for people with different eyesight capabilities. And, yes, this applies even to some popular sites that are trying to sell me something. However, I've never had a problem on Amazon.com, any Google applications, or other very high traffic sites. So it is possible to get the proper display for most users if you "want" to go to the effort. It's not possible to get perfect display under all circumstances. But with my own designs I test very widely and able to get very reasonable display no matter what size fonts, resolutions, etc. a user wants to throw my way.
> I have created the CSS for my own site and I know that > -- despite all the high-flying promises of CSS -- you can > get radically different behavior across browsers for even > relatively simple CSS. My solution is to dumb down my CSS > even more because working consistently across browsers > (without a Herculean effort on my part) is more important > to me than being fancy.</p>
Sorry, I found this ironic. mindview.net has a white square on most of the pages which I believe is supposed to be the background color for on screen text. The problem is that the text always overflows the right edge of the square in the latest versions of Firefox and Konqueror (both on Linux). While it is consistent and does not interfere with site navigation, it is very ugly.
As to the "web is getting worse" argument, it is the same as it has always been. Most regularly maintained sites look the same in most browsers, less regularly maintained sites look more horrible as time goes on, simple sites always look the same, which is, well, simple.
AJAX hasn't added anything to the equation. Ever since IE 4 people have been using DHTML to attach all kinds of bells and whistles to their page. Unless designed well, these bells and whistles reduce functionality and make pages render incorrectly on many browsers.
> There are way too many permutations of browsers, > extensions, toolbars, operation systems, etc for anyone to > really test all of them. In some ways it is too bad that > Web 2.0 had to be backwards compatible with Web 1.0. It > would be nice to start over.
Yeah, maybe Display Postscript. You laugh but it was a real language. Worked pretty good too.
I don't know which kind of sites are giving you trouble, but I guess it's simply because the Web is getting more popular. At first, only coders wrote webpages; and so we had all those blinking tags everywhere. After a short time, great designers had show us how beautiful (and standards-compliant) a Zen Garden [1] could be. But it was a secret, only the great knew how to do it. Unfortunately, now we have script kiddies cutting and pasting Ajax code to put together half-baked pages everywhere. Unfortunately, this does not affect only amateur websites - even established companies can get it wrong. Some people don't even know how to evaluate professional work anymore - if the site works for him and it's cheap, it's good enough.
I see some parallels with the rise of word processing and desktop publishing tools. A couple decades ago, in order to write a book one had to hire an editor and a typesetter. It was easy to separate amateur work from professional work. Now everyone can write, and the quality of the printed books dropped off dramatically. This can be seen even in the quality of something as simple as company letters. Now that everyone can write, no one knows how to write anymore. That's the same for the Web.
That's not the fault of CSS, 'cause it did right what it could possibly do: gave a method to control the page layout in much more detail than one could possibly do before. The bad thing about CSS doesn't really depend on CSS - just no browser but Konqueror and Amaya do really render CSS instructions as they ought to be rendered. MSIE6 ignores most part of CSS, MSIE is much better and so is Firefox, but all three support neither version of CSS completely, so the designer has to fail due to lack of support for features he relays on. As a matter of fact, there's no way to create a correct XHTML+CSS document for each of Firefox and MSIE with some special layout that should be available. And AJAX is even much worse - most of AJAX-generated pages violate XHTML and CSS, and therefore *SHOULD* be rendered differently in different HTML engines.
I don't know, CSS certainly isn't free of the blame, the problem is that while CSS lets someone create almost limitless page layout, the majority of the layouts will fail in one way or another given a major change in font sizing/resolution. So the web is getting "worse" for those that like bigger fonts/resolutions as web designers that know/use CSS become more common and update or create more of the new fancy web pages.
Given a nice layout, and supporting major font size changes, it seems web designers are forced to make a change, and since supporting font size changes is not always easy, they seem to be favoring the better layout. (And yes, fonts will overflow wrong on even well designed pages such as found on www.csszengarden.com)
> before. The bad thing about CSS doesn't really depend on > CSS - just no browser but Konqueror and Amaya do really > render CSS instructions as they ought to be rendered. > MSIE6 ignores most part of CSS, MSIE is much better and so > is Firefox, but all three support neither version of CSS > completely, so the designer has to fail due to lack of
Do remember that the CSS "standard" hadn't been finalised when MSIE6 was finalised. Microsoft put in what they could as well as what they had proposed that hadn't been included yet officially into the spec. Mozilla/Netscape did the same, causing conflicts. In the end some things were adopted that were matched by neither, causing rendering problems.
MSIE7 and Firefox had more time, so have more complete and correct (according to the CSS 1.1 spec) rendering, but the same story repeats in regard to CSS 2 (as well as some backwards compatibility problems).
> support for features he relays on. As a matter of fact, > there's no way to create a correct XHTML+CSS document for > each of Firefox and MSIE with some special layout that > should be available. > And AJAX is even much worse - most of AJAX-generated pages > violate XHTML and CSS, and therefore *SHOULD* be rendered > differently in different HTML engines.
And that's why so many sites have problems these days, why it seems worse than in the past. The same sloppy html code (likely generated by some wysywyg tool) that a few years ago would go unnoticed because the browsers silently ignored the errors or tried to compensate now make things break because of the far greater complexibility of the sites themselves with the tons of Javascript, XML injection, etc. etc.
The people writing that html still write the same quality code as before, but with the complexity of that code now reaching the point where browsers can no longer compensate for it those flaws now become visible to the end user.
> Do remember that the CSS "standard" hadn't been finalised > when MSIE6 was finalised. > Microsoft put in what they could as well as what they had > proposed that hadn't been included yet officially into the > spec.
They had loads of updates after the standard was released. They had a lot of time to fix it, but they didn't simply care.
> MSIE7 and Firefox had more time, so have more complete and > correct (according to the CSS 1.1 spec) rendering, but the > same story repeats in regard to CSS 2 (as well as some > backwards compatibility problems).
GECKO supports some tag-value combinations that shouldn't exist, while ignoring some other that a really needed, but have some corresponding Netscape tags. It's not the lack of time, it's the lack of will and standard's meaning understanding.
> The same sloppy html code (likely generated by some > wysywyg tool) that a few years ago would go unnoticed > because the browsers silently ignored the errors or tried > to compensate now make things break because of the far > greater complexibility of the sites themselves with the > tons of Javascript, XML injection, etc. etc.
If you are using firefox under linux, there are many pages that don't render correctly. Many of these same pages render with firefox on the mac and windows, and many of them render with Konqueror on linux.
Flat View: This topic has 15 replies
on 2 pages
[
12
|
»
]
Previous Topic
