For the development of DIV+CSS, we also need to listen to another voice.

Css

Article: Andy budd/translation: Jjgod Jiang

Based vs. CSS Based Design from an objective look in Table

I have the same opinion as the author. The standard is good, but it may not be practical.

Here is the ZT article:

Over the years, many excellent articles have praised the superiority based on CSS design, lamenting the decline of tabular design. But very few people think about it, perhaps because you have to understand and use the CSS based design before you can criticize it, and once you know it, you do not want to go back to use the original old design method.

To make up for this imbalance, and because it's cool to play a big villain in this game, I decided to write an article about why, in some cases, traditional table designs are no better than CSS-or standards-based-design methods.

The form of demonizing

Before the table appeared, the WEB was a rather tedious place to use table typesetting to open a new landscape of visual page design. How much of a table's contribution to the popularity of web and web design may be controversial, but once you're out of the table, it's clear that our web designers are going to lose their jobs.

In recent years, tabular design has indeed been demonized, and Web purists will tell you that tables are meaningless to typography, so you never use it. But history has proven that many technologies have been designed to achieve a goal in the first, but have found greater use in other areas. Like the Web itself, it's not just about sharing research data at first, but the entertainment and business applications are now going hand-in-hand with information and education.

Two, just for the comfort of a little

WEB designers have been using table-typesetting pages for years, a capability most designers have mastered. Using tables so you can get the results you expect, with some simple hack, such as spacer GIFs, we almost certainly guarantee that our site looks the same on the widest Web browsers, from the lowest version of Netscape 4 to a modern browser like Safari.

Although pioneers have long advertised Web standards, most Web sites are developed using tables and incompatible standard code, so user agents have to support tabular typography for a long time. This is a fatal blow to the selling point of WEB standards: standards do not have a standard position. It is unlikely that one of the major browser vendors (well, or Microsoft) has suddenly released a browser that most Web sites cannot display.

So web designers don't feel the need to start using CSS typography and support code that supports standards.

Third, reduce the threshold of the Web

It is so successful because of its low threshold: HTML is a simple and easy language to learn, and browsers tolerate many documents that are marked with confusion. This makes it incredibly easy to publish content on the Web. Even your 12-year-old nephew can use the Frontpage included with Microsoft Office to work out a simple web site.

Based on the table design than the CSS, of course, CSS syntax is very simple, normal people will agree: you do not need to be a rocket scientist to learn CSS. Still, some of these concepts are too subtle to grasp. On the surface, for example, the Box model is simple, but I occasionally slip and fall on the border fold (margin collapsing), floating (float) and clearing (clear) are not easy to understand and difficult to use. In my experience, it takes about a 6-12-month learning curve to learn the basics of CSS to easily develop a site with CSS.

Then there is the problem that the browser support does not support. Once you've officially started working, you'll learn what browsers support, what they don't support, and some common browser bugs. Unfortunately, there are too many bugs, even "experts" are hard to gauge how much time they spend fixing bugs. It's even more frustrating for beginners, because they don't know if they've misunderstood CSS or some obscure browser bug? Maybe that's why the same question has repeatedly appeared on the mailing list, such as Css-discuss.

It will be much easier to develop a site with CSS if the browser vendors are in sync. But I still think--most people will agree that the thresholds for--CSS development are much higher than those based on tables. To put it another way, I think it also explains why CSS based design is so popular among web experts. This allows them to differentiate themselves from the amateur "front-page cowboys" and let them get back to the feeling that the Web belonged to a small group of their own. That's probably it. So many people regard web standards as untouchable "ivory towers," and advocates of Web standards look at Web design with a frenzy of superiority.

Some things are easier to do with a table

I'm sure all of us have found that we've written a rather complicated CSS for the purpose of making a table do a piece of cake. such as the appearance of the form (form), the shape of a complex and bizarre form can be easily done with a table. You can achieve similar results with CSS floating elements, but it's a lot of trouble. If you are a CSS guru, this kind of trouble is also a happy thing. But there's no doubt that if you're just a normal person, there's a guy who's going to pinch your throat and ask you how to make a small form and spend so much time with the boss, things are not so fun.

If you have enough knowledge and patience, most of the things you are accustomed to doing with tables can still be implemented in CSS. Although it takes a long time, there is a limit (or a blow to give up the attempt). The point is that there are some things that no matter how hard you try or what you can't achieve, one of them is the footer bar (page footer). I often see posts from frustrated CSS authors who try to create a footer bar that sticks to the bottom of a window so that even the window does not stretch across the screen to ensure the effect. If you use the table, to make this effect is very simple, you can use the CSS alone is another thing. Why are there Web developers who don't want to use CSS? That's because once you don't use a table, simple things get complicated.

V. Exaggeration of income

There are many reasons why you can lose your form and adapt to CSS based typography, and many people exaggerate the benefits of the torrent of web standards. A large site instead of CSS typesetting can really save a lot of bandwidth. For most of the other sites, the benefit of a small Ching can be negligible.

Everyone wants the page to load faster, and standard advocates say CSS can help you do that. Most sites "design" is evenly distributed throughout the site, but the CSS based "design" is placed in one to more files. These files can quickly become complex and large, even for a small site. I recently designed a site that uses 4 stylesheets, plus 12k (though it includes blanks and annotations). Using CSS is actually loading and browsing in the first place, rather than distributing the data you want to load evenly across the entire site. In other words, the home page takes longer to download than the table layout. Only if the stylesheet has been downloaded, they will be cached without the need to download again. But after all, a site's home page is the last one you want to load so slow.

Six, attract customers

Even though web designers sometimes feel the need to tie Web standards to their customers, it's a pity that most customers don't care about the code for the site. We usually use carrots and rods, carrots are like the search engine friendliness, and the stick is the web's affinity (accessibility).

Indeed, search engines are more like the semantic markup of the page, and everyone also think that search engines like short code, through CSS and WEB standards to build sites can greatly enhance the search engine friendly site development. But there is no silver bullet. Many table-based sites are still getting high search engine rankings. A site developed with CSS may also only get a bad ranking. The key to high ranking is the content and links from elsewhere, rather than using tables or CSS for typesetting.

In addition, on the use of the "affinity" of the word to tie the WEB standards, especially CSS design, in fact, based on the design of the table does not have any innate affinity defects, the table as long as the linear, it has meaning, the content will have affinity. The current screen reader technology has been good, and most of the screen reader can be very good support for the table based site. Of course, the syntax of your site is best certified through AA affinity level, even for the more stringent AAA level, no form design is just a suggestion, not necessarily.

Another often mentioned benefit is that it allows customers to be independent of the design provider. In a world where everyone is developing in accordance with standards, it is easy for customers to change their development partners, and new developers can quickly understand the organizational structure of the site without having to wade through someone's previous markup. But it takes a lot of design providers to be proficient with WEB standards. Unfortunately, this is not the case now. While experienced CSS developers are on the rise, this is a relatively professional area, so it is risky for large companies to lock in this type of development-a lack of skilled developers. My personal experience is that if an organization is going to use CSS to develop a site, it will take a long time to maintain at least one experienced designer. So turning to WEB standards now is not a reduction in customer dependence on developers, but an increase.

Vii. Summary

There is no doubt that WEB standards and CSS based design are the way to the future. But in the process of rushing to them and advocating new technology, we will also wonder if what we preach is too exaggerated. To do something more realistically often does not reach our expectations. and dogmatic implementation of these is likely to alienate the partners we should most win.

A table based design can still exist for a long time. To attract developers, we can use examples to teach people to get started and to lower the threshold. Let's not make a new threshold. We have to be honest about the benefits and the costs. It can be difficult and time-consuming to develop a CSS site, and in some cases it is much more meaningful to use tables for typesetting than for CSS.