On Sat, 2007-04-14 at 10:35 -0400, tedd wrote: > At 9:07 PM -0400 4/13/07, Robert Cummings wrote: > >I noticed your site isn't a pixel perfect layout. Probably why you > >haven't had to tear at your face very hard with CSS :) That whole > >box-model issue becomes a great deal more elastic when you have some > >fudge room. > > Rob: > > My site (http://sperling.com) is designed to optimize the area > provided it by the browser/user. I realize that the web is not > traditional print and I am attempting to find the "best" way to > present content in this medium. > > I have done pixel perfect web pages for clients where they show me a > Photoshop image and say "That's the way we want it to look in all > browsers AND we don't want tables." That's what absolute position is > used for. There isn't a Photoshop image that anyone can show me that > I can't make a "pixel perfect" AND compliant web site out of using > css. Who said pixel perfect image? Absolute pixel perfection is easy as you've mentioned. Now make pixel perfect stretchy pages where the widths expand to fill the available horizontal and heights expand to fill the available vertical. > As for the box-model issue, I've faced that, such as in: > > http://www.symboldomains.com/ This is a simplistice layout. It doesn't have the same issues as 3 column layouts. Also, you are using CSS hacks... just as bad as tables IMHO. > While the site is a bit dated (first release in 2003), it still holds > together. The placement of the border handles IE's difficulty in > understanding the box model (just where does M$ get these browser > developers?). > > For more "typical" sites, may I present my past life: > > http://www.geophysics.com/ Again, a simplistic layout... with CSS hacks. > I've been putting up web sites for the last ten years. I went through > the table phase a few years back and I haven't had a reason to go > back. Like documentation and semantic makeup, it just makes sense. > > In short, I am a recovering tableholic, but I've been table > independent for five years, eight months, and seven days. I know how to do layouts without tables, but not without resorting to hacks, or not getting pixel perfection a stretchy design. I'm am well aware of CSS. I am not a table-aholic, I am a pragmatist. > >You've got me wrong, I don't want to stick with tables, I want to use > >the full power of CSS, but unfortunately Microsoft has seen fit to screw > >everything up as much as possible. > > Yes, there is always difficulty in the world, but realize that we > prosper from it. If web stuff was easy, then everyone would be doing > it correctly. "Correctly" is a moving target... shouldn't you be moving away from HTML4.01 strict to XHTML already? ;) > > > but if you are as well versed as I am in CSS, then you know that > >> using tables for layout is more costly in overall development, > > > maintenance, sales, and accessibility. To me and my clients, those > >> things matter. > > > >I agree with accessibility to some degree, but for the rest I think > >you're a bit too far out on a limb. > > To me, being disabled, accessibility is paramount. Granted I don't > always live up to my mouth, but I try and that's the point. To me, undisabled as I am, accessibility is important, but what's important to me, isn't always important to my paying customers. There's a middle ground between accessibility, time, and cost. > As for the rest of my claim, my experience is that css reduces cost > for over all development -- it takes *me* less time to develop an > entire site using css than using tables. > > Site maintenance is certainly easier -- no reason to dwell on the obvious. > > Sales are better for several reason: a) accessibility brings up-to > another 12 percent more customers to the table; b) sites built with > css load faster bringing more people; c) better SEO considerations > brings even more people; d) faster site alteration to cycle products > allows more exposure and makes for a more interesting sites, which > brings back even more people. Are you implied 12% of the population visiting websites are disabled in such a way as to benefit from CSS? That's seems a tad high. Maybe 1%... if that. > The whole point of making a web site is to sell something -- ether a > product, a service, an opinion, yourself, or whatever. The larger the > audience, the more successful you will be. Using css brings in more > people. Please back up that claim with real statistics. > --- > > > Concepts like separating content from presentation, graceful > >> degradation and progressive enhancement are not just phrases one > >> imagined so they could sit on a glass pedestal and look down on > >> everyone else. They are practices that work and are far more reaching > >> than WYSIWYG table layouts. > > > >Agreed, I practice all of the above, but I feel no guilt in using a > >table for a layout when it simplifies the issue. Between box model > >issues, float bugs, etc etc, CSS just can't do what needs to be done > >yet. > > On this we disagree. I have found nothing that tables can do that I > can't do with css AND do it more effectively and efficiently. Yes, I disagree, I trust that I'm far more productive to than you *grin*. > The box-model problem and float considerations can be easily dealt > with IF you truly understand the problem they present. There are > things you can use to illustrate the problems just like in php. Awww, see now you're trying to talk down to me as if I'm new to CSS. > For example, we all use pint_r() to show us what values our variables Can't say I've ever used pint_r()... what does it do? Serve up beer? >:) > hold -- it helps in debugging. The same goes for css, try using the > rule "border: 1px solid red;" the next time you're wondering about > why something isn't placed where you want it. Once you see where/how > everything fits together, it's simple to fix it. Awww, you really are treating me like an idiot now *lol*. I'm well aware of using borders to reveal where things are. Although, that's not always helpful, because in a pixel perfect layout, and the W3C box model, everyone worth their salt knows, that adding the border increases the width by 2 pixels, and that's enough to destroy your layout ;) Of course, that doesn't happen in the Microsoft box-model until you declare a doctype... and even then, only in IE 6+ is the box-model interpreted correctly for windows. > Look, you're a competent programmer and php is far more complex than > css will ever be. You're doing yourself a disservice by letting > simple stumbling blocks deter you from using css because you have a > misconception that css isn't ready yet, because it is. That's where we disagree. I use CSS plenty, but I also use tables. I create XHTML strict documents that validate. I also use tables when I feel they are appropriate. We've only touched on a few CSS things so far, I haven't even gone into height, min-height, max-height, vertical position, display (inline-block, table-xxx, ...), position (absolute/relative nesting, throw in some floats). This stuff is mostly borked. Cheers, Rob. -- .------------------------------------------------------------. | InterJinn Application Framework - http://www.interjinn.com | :------------------------------------------------------------: | An application and templating framework for PHP. Boasting | | a powerful, scalable system for accessing system services | | such as forms, properties, sessions, and caches. InterJinn | | also provides an extremely flexible architecture for | | creating re-usable components quickly and easily. | `------------------------------------------------------------' -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
