dslreports logo
 
    All Forums Hot Topics Gallery
spc
uniqs
3975

jDyno
Premium Member
join:2001-02-20
Washington, DC

jDyno

Premium Member

Dvorak embarrasses himself on CSS instead of Macs this time

»www.pcmag.com/article2/0 ··· 1,00.asp

KoolMoe
Aw Man
Premium Member
join:2001-02-14
Annapolis, MD

KoolMoe

Premium Member

Well, it's not the mess that he claims but he does have valid points. How come IE and Mozilla work with padding/margins differently? Doesn't the standard specify how that should be dealt with? If not, why not!

But I disagree with it being a mess. I'm not an expert at this stuff but know CSS pretty well and there are very few tweaks I need to make to get a layout working in IE just like it does Mozilla...

So it definitely sounds like a rant from someone who doesn't know what they're doing (and needed another subject for his syndicated column) but he does have some good points...

Though I laughed at, "If your Internet connection happens to lose a bit of CSS data, you get a mess on your screen." He must have a pretty crappy connection!
KM

justin
..needs sleep
Mod
join:1999-05-28
2031
Billion BiPAC 7800N
Apple AirPort Extreme (2011)

justin

Mod

my blackberry regularly spits out a ton of stylesheet as text instead of a nicely rendered page (for some sites). Yes it is probably bugs with the blackberry proxy server and style sheet interpretation, but it is very annoying, and has been like this for over a year.

JAAulde
Web Developer
MVM
join:2001-05-09
Frederick, MD
ARRIS SB6141
Ubiquiti EdgeRouter Lite
Ubiquiti UniFi AP

2 edits

3 recommendations

JAAulde to jDyno

MVM

to jDyno
Dvorak is making himself look like an idiot. Now, don't get me wrong--developing cross browser compatible sites can be frustrating. I'm not dogging Dvorak for ranting about browser differences. But to take his platform and mis-represent this technology has knocked him down a few more notches as far as I am concerned.

First, you don't get mad at the technology. You might get mad at the vendors who are not implementing it properly, but not the technology.

Second, for the most part, there really aren't that many difference between the browsers. And it isn't only in CSS where you find differences. It's in (x)HTML, CSS, JS and so on. But again, this is due to the vendors interpreting things differently, or simply going against the grain hoping their version will get more popular and cause their software to gain ground (cough, MS, cough).

Third, differences that do exist are correctable if you understand the technology. For example, to quote KoolMoe See Profile:
said by KoolMoe:
How come IE and Mozilla work with padding/margins differently? Doesn't the standard specify how that should be dealt with? If not, why not!

IE and the other guys have different 'box models' from each other when IE is in 'quirks' mode. Take IE out of quirks mode and things get much better. How do you get IE out of quirks mode? Ensure your DOCTYPE is valid and is the absolute first bit of text (no white space ahead of it either) in your served page. Answering the question as to why there are differences when IE does things its own way, see the above paragraph about vendors who go against the grain.

Next, quoting Dvorak:
said by Dvorak :
The first problem is the idea of "cascading." It means what it says: falling—as in falling apart. You set a parameter for a style element, and that setting falls to the next element unless you provide it with a different element definition. This sounds like a great idea until you try to deconstruct the sheet. You need a road map. One element cascades from here, another from there. One wrong change and all hell breaks loose.

Let me paraphrase him: "I don't know CSS so when it breaks I don't know what to do. This is the fault of CSS."

Finaly, finishing that quote, this comment is absurd:
said by Dvorak :
If your Internet connection happens to lose a bit of CSS data, you get a mess on your screen.

Ummm, sure. And if your connection drops some HTML, same thing. Oh yeah, and some JS--yeah, that breaks if you drop bits to. Welcome to the Internet. If your crappy connection drops data, the stuff you did get acts weird. Reload for Heaven's sake!

Edit: for clarification

geekamongus
MVM
join:2004-07-27
Asheville, NC

geekamongus to jDyno

MVM

to jDyno
Dvorak rhymes with "Border Hack".

insomniac84
join:2002-01-03
Schererville, IN

insomniac84 to JAAulde

Member

to JAAulde
said by JAAulde:

this comment is absurd:
said by Dvorak :
If your Internet connection happens to lose a bit of CSS data, you get a mess on your screen.

Ummm, sure. And if your connection drops some HTML, same thing. Oh yeah, and some JS--yeah, that breaks if you drop bits to. Welcome to the Internet. If your crappy connection drops data, the stuff you did get acts weird. Reload for Heaven's sake!

Edit: for clarification
Yea, I was kinda weirded out reading that. It's just like saying the RAR format is flawed because a glitch with my internet connection or browser corrupted the file and therefore I can't extract it.

jDyno
Premium Member
join:2001-02-20
Washington, DC

jDyno to justin

Premium Member

to justin
said by justin:

my blackberry regularly spits out a ton of stylesheet as text instead of a nicely rendered page (for some sites). Yes it is probably bugs with the blackberry proxy server and style sheet interpretation, but it is very annoying, and has been like this for over a year.
Yeah, this has nothing to do with CSS itself. If this gets on your nerves, complain about network stability and Blackberry technology, not CSS. In fact, sites using CSS for layout - under normal conditions - render excellently on Blackberry. Since they separate the content from the presentation. So, on Blackberries, you just get the text, no annoying layout or non-content images. (If it's done right.)

Same with the IE/Mozilla box models difference. This isn't a problem with CSS, it's a problem with browser makers. The spec is very clear on what the box model should be. Apparently, the browser makers weren't.
jDyno

jDyno

Premium Member

Learned response by person smarter than me.

»westciv.typepad.com/dog_ ··· css.html

JAAulde
Web Developer
MVM
join:2001-05-09
Frederick, MD
ARRIS SB6141
Ubiquiti EdgeRouter Lite
Ubiquiti UniFi AP

JAAulde

MVM

said by jDyno:

Learned response by person smarter than me.

»westciv.typepad.com/dog_ ··· css.html
An excellent response.

Mospaw
My socks don't match.

join:2001-01-08
New Braunfels, TX

3 edits

2 recommendations

Mospaw to jDyno

to jDyno
Those who know, do. Those who don't, are Dvorak.

I think I've finally figured out his recipe: Take a few blanket statements that are so vague anyone can find "truth" in them. Mix in a quick compliment to take away the sour (and any appearance of partiality). Sprinkle on a liberal dose of half-truths and combine with blustering ignorance. Make sure all theories are half-baked. Publish. Repeat.

Dvorak is a journalist, not a web designer. The technologies aren't bad. He simply doesn't know them. I'd love to see him write an article about rebuilding a car engine. "What's with all the grease? These engines all need lubrication. You'd figure after 100-plus years of development, the internal combustion engine would be friction-free by now." And so on.

I am a web designer and have been for about a decade now. (I've actually been a journalist, but that's a different rant.) Sure anyone can learn some basic Photoshop and a little HTML and throw together a web site. Many even look good. But to really pull it off elegantly, whether with pure CSS and standards-compliant or tables, font tags, and blank gifs, requires a deep understanding of the underlying standards. HTML, CSS, JavaScript, etc. It also requires understanding that transcends the technologies themselves, such as user interface design, accessibility, usability, and just being able to get that message out. If you have only partial knowledge or fail to understand the why of things, you're going to fall short. If your knowledge of the technologies is lacking, you might miss out on salient points that would help you.

It's clear that Dvorak isn't approaching this with a designer's mind so much as he probably wants his designs to coalesce in front of him automatically, delivered by some machine god that already knows what he wants. Too bad such a thing doesn't exist.

Yet, he has valid complaints under all the ignorance. Working around browser inconsistencies is a bit of a pain, but do any amount of web design and you quickly learn what these problems are and, more importantly, how to avoid them. Most important, however, is that as a web designer, it's your job to do know these workarounds and employ them. In an ideal world, it wouldn't be necessary, but name a profession where all things are ideal. If you don't the tools that are available or how web sites are designed and built you can quit the business, learn how to deal with it or make your own browser.

Better yet, Dvorak, hire a real web designer who know what he's doing and save the bitching for how Microsoft Vista will stink. Or rule. Or how Apple is finally (after 20+ years of waiting) going out of business. Or not.

Maybe we should be happy that Dvorak got a new theme about which to espouse his tremendous ignorance. Really, stick with something you're qualified to do. Something along the lines of "Would you like fries with that?" Dvorak exists to blather on about things he knows little about, get some attention and move onto the next target of his ignorance. The part about "cascading" really proves that out.

CSS is hardly perfect, but it's a damn good tool, and (for me, a professional) works better than any alternatives. It's not easy, but it's not meant to be. Until it becomes easy enough that anyone can do it by thought alone, I'll still stick with effective.
meta
join:2004-12-27
00000

meta to JAAulde

Member

to JAAulde
Those responses dont change the fact that CSS sites are a pain in the ass for users. Try selecting text on half the sites that heavily rely on css and you will end up selecting half the page, or nothing at all. CSS is a pox on the web.

Mospaw
My socks don't match.

join:2001-01-08
New Braunfels, TX

Mospaw

Got any links to back that "selection" problem up? I have seen it happen, but in what, 0.5% of sites out there. I keep hearing this is some sort of epidemic, but nobody ever posts proof, or even a single link. Text selection is almost like the "big foot' of CSS problems. Lots of people have anecdotal evidence, but nobody ever produces anything believable.

CSS might be a pox on the web, but it makes the web look a lot nicer. And it's easier to assemble than tables.

JAAulde
Web Developer
MVM
join:2001-05-09
Frederick, MD
ARRIS SB6141
Ubiquiti EdgeRouter Lite
Ubiquiti UniFi AP

1 edit

1 recommendation

JAAulde to meta

MVM

to meta
said by meta:

Those responses dont change the fact that CSS sites are a pain in the ass for users. Try selecting text on half the sites that heavily rely on css and you will end up selecting half the page, or nothing at all. CSS is a pox on the web.
Can't say that any CSS site (coded by me or anyone else) has given me any real fit. But I am pretty good at positioning the mouse where I need it...

But even if that were an issue--that text selecting problem, does that one little glitch cause the whole technology to be problematic or cause a whole site to be un-usable (or a pain in the ass as you put it)? Of course not. Show me a bug free technology...

Furthermore, if there really is a text selecting issue (which as I said I have not seen), it isn't the fault of the technology. It is the fault of the vendor who is mis-implementing the technology. Do you really think the individuals who wrote the CSS standard said, "I know, let's add in a messed up version of text selection!" Get real.

geekamongus
MVM
join:2004-07-27
Asheville, NC

1 recommendation

geekamongus to meta

MVM

to meta
pox (pks)
n.

1. A disease such as chickenpox or smallpox, characterized by purulent skin eruptions that may leave pockmarks.
2. Syphilis.
3. Archaic. Misfortune and calamity.
I fail to see how CSS can be associated with any of the possible definitions for 'pox'.

Every single argument I have ever heard against the use of CSS is always, without fail, based on a lack of understanding of the technology. People who refuse to learn it, who lash out against it, who say it is too buggy, who say it is terrible, etc etc etc, all have one thing in common: they do not know how to use it.

Not once have I ever seen anyone learn to use CSS to an intermediate degree abandon it, write it off, or continue to blather about how terrible it is.
meta
join:2004-12-27
00000

meta to JAAulde

Member

to JAAulde
The vendors each follow the same specification and all come to different conclusions. These are the same exact coders that are capable of following http, ftp, and dozens of other protocol specifications, so why the discrepancies here? Why so many of them in one place? The standard is clearly flawed, and should be scrapped in favor of a new standard properly outlined with CONTRETE EXAMPLES provided (YES that means you bums at the w3c need to start CODING what your preaching) instead of just arguing about policy.
meta

meta to geekamongus

Member

to geekamongus
There is no point arguing with CSS fanboys that are unwilling to see the negative side of their favorite toy to avoid doing work. Thats right, you use it because its fewer keystrokes and less thinking. If you want a page that behaves as people have come to expect over the years, use tables. If you want random behavior and differing appearances use CSS. How hard is that to understand. Im done posting for today, webdevs give me a headache.

JAAulde
Web Developer
MVM
join:2001-05-09
Frederick, MD
ARRIS SB6141
Ubiquiti EdgeRouter Lite
Ubiquiti UniFi AP

JAAulde

MVM

said by meta:

There is no point arguing with CSS fanboys that are unwilling to see the negative side of their favorite toy to avoid doing work. Thats right, you use it because its fewer keystrokes and less thinking. If you want a page that behaves as people have come to expect over the years, use tables. If you want random behavior and differing appearances use CSS. How hard is that to understand. Im done posting for today, webdevs give me a headache.
The pages I code all look the same in all browsers.

I am webdev, see me code.

Mospaw
My socks don't match.

join:2001-01-08
New Braunfels, TX

1 recommendation

Mospaw to meta

to meta
I sense just a bit of anger here. Bums? Why so vitriolic?

FTP is a relatively simple protocol. You can read and understand it in a day or two. The same with HTTP and likely "dozens of other protocols". There is no "fudge factor". They are protocols for allowing machines to speak to each other in narrowly defined circumstances.

CSS is an entirely different animal. By its very nature, it's more complex. It covers a vast array of possible formats, designs, tastes and even, no pun intended, styles. Comparing FTP to CSS (standardized or not) is like comparing a door handle to a jet engine.

I'm not saying that the CSS standard is perfect. Written into it is room for interpretation. And as you have so elegantly pointed out, different vendors interpret things differently. I still don't see how this is a problem with the technology. I call it flexibility.

Rather than attacking things on a wholesale level due to a few minor deficiencies, your energies would be better spent actually trying to improve that's wrong or coming up with a better solution.

So, since CSS is so bad, so worthless, so utterly beyond repair, what's better for the same task? Or what are you doing to improve it?

geekamongus
MVM
join:2004-07-27
Asheville, NC

2 recommendations

geekamongus to meta

MVM

to meta
said by meta:

There is no point arguing with CSS fanboys that are unwilling to see the negative side of their favorite toy to avoid doing work. Thats right, you use it because its fewer keystrokes and less thinking. If you want a page that behaves as people have come to expect over the years, use tables. If you want random behavior and differing appearances use CSS. How hard is that to understand. Im done posting for today, webdevs give me a headache.
This sums up well my point from above. A deeper knowledge (heck, a basic knowledge) of CSS would reveal that it is very easy to create entire sites that have no 'random behavior' or 'differing appearances'.

Seen MSN lately? How about Yahoo? Nary a layout table in site!

rjackson

join:2002-04-02
Ringgold, GA
Netgear R6400
Switches Trash Bin
Apple AirPort Extreme (2011)

3 recommendations

rjackson to meta

to meta
said by meta:

Those responses dont change the fact that CSS sites are a pain in the ass for users. Try selecting text on half the sites that heavily rely on css and you will end up selecting half the page, or nothing at all. CSS is a pox on the web.
I think I asked you in the other CSS thread for some examples of this, because I personally haven't encountered a site you couldn't select text on that was caused by a CSS layout. You didn't respond there, and I suspect you won't here either.

»Re: CSS vs. Table design - do you agree or disagree?

Discussing the pros and cons of a technology is one thing. Making blanket statements about something without backing it up with facts is just empty rhetoric. This is your chance to put up or shut up.
meta
join:2004-12-27
00000

1 edit

meta to geekamongus

Member

to geekamongus
Click for full size
Click for full size
»torrentfreak.com/ ~~~ try selecting the text "The world leaders..." in the latest version of IE. there is one of MANY example sites people come across every day with hosed layouts. See attachments for examples of other pages not loading right.
meta

meta

Member

Click for full size
Firefox fails too.

Mospaw
My socks don't match.

join:2001-01-08
New Braunfels, TX

Mospaw to jDyno

to jDyno
The entire argument is a bit silly to me, upon further reflection.

Dvorak is attacking a standard and cites as evidence of its problems issues of implementation. The analogies as to how this is defective thinking are too many to post and mostly obvious anyway.

CSS isn't perfect. I don't think anyone here would even try to argue that. But it's the best we've got for what it does. Standards have made the web orders of magnitude easier to deal with from everyone's standpoint: developers, users, and so on.

Remember back in the old Netscape Navigator 4.0 vs. Internet Explorer 4.0 where the "standards" had very little overlap, and you had to effectively make a different web site for each browser? I do. It was painful, and using CSS back then even more so. But it wasn't CSS's fault! It was Netscape's fault, Microsoft's fault, and the fault of everyone else who made browsers that didn't adhere to the standard. No, "one site for all" isn't here yet. It may never be. But standards have gotten it a lot closer.

Eliminating standards isn't the solution. Only chaos can result from that. Embracing them may not be for you, but that's irrelevant. A true pro knows what to do. We're not totally "there' yet where standards are truly standard. But we're closer than we ever have been and even stubborn companies are realizing that the wild west days of web innovation are over, and everyone is going to have to learn to play together, using the same toys.

geekamongus
MVM
join:2004-07-27
Asheville, NC

geekamongus to meta

MVM

to meta
That looks like poor coding by the torrentfreak people. 186 validation errors on that page alone, and many are unclosed tags and other things that would trigger exactly such behavior in IE.

Again, not the fault of CSS; the fault of the designer/coder in this case.

Mospaw
My socks don't match.

join:2001-01-08
New Braunfels, TX

Mospaw to meta

to meta
said by meta:

»torrentfreak.com/ ~~~ try selecting the text "The world leaders..." in the latest version of IE.
said by meta:

Firefox fails too.
Actually, IE and FF didn't fail. They're rendering the content.

The web developer failed. This is the fault of CSS how?

If I'm ignorant of the law, that doesn't mean I'm allowed to break it.
meta
join:2004-12-27
00000

meta to geekamongus

Member

to geekamongus
And what of the random behavior from the discovery channel website? Clearly they arent halfassed and they STILL SUFFER a similar mis-rendered fate

geekamongus
MVM
join:2004-07-27
Asheville, NC

geekamongus

MVM

said by meta:

And what of the random behavior from the discovery channel website? Clearly they arent halfassed and they STILL SUFFER a similar mis-rendered fate
It actually looks fine in Firefox on my PC. BUT, again, the site is poorly coded and validation errors abound. That would be Discovery's fault, not CSS.

rjackson

join:2002-04-02
Ringgold, GA
Netgear R6400
Switches Trash Bin
Apple AirPort Extreme (2011)

1 edit

rjackson to meta

to meta
said by meta:

»torrentfreak.com/ ~~~ try selecting the text "The world leaders..." in the latest version of IE. there is one of MANY example sites people come across every day with hosed layouts. See attachments for examples of other pages not loading right.
Torrentfreak is poorly coded, plain and simple. For what it's worth though, I only experienced the selecting thing in IE6 on Windows. All other browsers/platforms had no problem at all selecting text.

Your other example, Discovery News, is a Flash problem. You can't overlay divs (the survey thing) over objects embedded on the page. Again, has nothing to do with CSS, and is the result of just a bad job done by the developers.

Please show us a site that uses valid markup and proper CSS that exhibits the problems you describe, or make an example yourself.

All you've shown so far is some examples of people not using their tools correctly. If you hold a hammer backwards, can you drive a nail properly?

Mospaw
My socks don't match.

join:2001-01-08
New Braunfels, TX

Mospaw to meta

to meta
Being half-assed or not is irrelevant. I've seen some major companies have sites that didn't work well in some, if not all, browsers. I've seen lots of half-assed sites that were absolutely brilliant in their execution.

Being big, having a lot of money, or somehow being a well-known name doesn't excuse poor implementation, inability to adhere to standards, or a lack of validation. it also doesn't guarantee that those things will happen.
meta
join:2004-12-27
00000

meta to geekamongus

Member

to geekamongus
Adhereing to standards? what standard! nobody can agree on that at all. The documents about CSS fail to specify concrete source code for an application to CORRECTLY render the formatting!
Maybe the requirements of the web development community are changing. Perhaps we should be working on new standards that arent quite so hard to figure out? Maybe we should call it SSS (stupid style sheets) so all the idiots told to use them instead of tables have a shot in hell of getting it right. Nobody seems to be able to agree on anything about css, and the designers arent exactly stepping in saying who is right and who is wrong, so where does that leave everyboedy in the middle of all this? Screwed. And im sick of dealing with sites like this. The pro-standard people blame the individuals and the implimentations, while the con-standard blame the standards creators and the standard itself with complexity and lack of clear example to work from.
Good game all around. I cant wait for a day when you get stuck at a form because the submit button is obscured by some misrendered hovering DIV. Then you will understand the frustration.