Just out of curiosity, what would you change in XHTML/CSS?

Ye' old general discussion board. Basically, for everything that isn't covered elsewhere. Come here to shoot the breeze, shoot your mouth off, or whatever suits your fancy.
This forum is not for asking programming related questions.

Moderator: General Moderators

Gambler
Forum Contributor
Posts: 246
Joined: Thu Dec 08, 2005 7:10 pm

Just out of curiosity, what would you change in XHTML/CSS?

Post by Gambler »

If you had power to define a new web design language without worrying about compatibility, what would you do?
Roja
Tutorials Group
Posts: 2692
Joined: Sun Jan 04, 2004 10:30 pm

Re: Just out of curiosity, what would you change in XHTML/CS

Post by Roja »

Gambler wrote:If you had power to define a new web design language without worrying about compatibility, what would you do?
Thats what xhtml2 did, for the most part.

The problem isn't web language at all. Its *one* browser manufacturer (who happens to have 80%+ marketshare) not following those standards, and not listening to customer feedback for over 5 years now.
redmonkey
Forum Regular
Posts: 836
Joined: Thu Dec 18, 2003 3:58 pm

Post by redmonkey »

Personally, I dont agree. As far as I'm aware, the W3C produce 'a' standard not 'the' standard. I'm sure their publications used to be branded 'recomendations' not standards, not sure if this is still the case?

I see no reason why any browser developer should be expected to adhere to any one set of recomandations over another. And, if the developers don't agree with a particular implementation of a recomendation, why should they adopt/implement it? I think it's good to develop to a standard, but in the absence of any true standard it seems perfectly acceptable to me for the developers to develop to their own standards.

I also think many of the problems with cross-browser rendering can be attributed to the developers interpretation of whatever spec they are following. I've read quite a few of the specs and in many areas I feel it is open to interpretation although admittedly, I can't think of any examples off-hand.

On that note, a couple of things off the top of my head that I would change....

I'd include visual representation of intended/expected behaviour within many areas of the actual spec to accompany the already existant descriptions in a bid to cut down on the amount of interpretation implemented currently.

As you can define padding and margins etc.. like.. "padding: 10 20 30 40;" I would like to see this type of thing extended to background images i.e... "background-image: url(images/someimage.gif) url(images/someimage2.gif) url(images/someimage3.gif) url(images/someimage4.gif);" where the images would be rendered and placed top left, top right, bottom right, bottom left respectively. I think that would safe a lot of designers a lot of time and seems (on the surface) to make sense to me.

I'd like browsers and online markup validators to parse my own DTDs so I can decide which attributes should and should not be supported/are and are not valid.
User avatar
feyd
Neighborhood Spidermoddy
Posts: 31559
Joined: Mon Mar 29, 2004 3:24 pm
Location: Bothell, Washington, USA

Post by feyd »

although it takes a bit of engineering, you can do the corners background fairly easily with 4 or 5 nested elements for the 4 corners and a normal background respectively.
redmonkey
Forum Regular
Posts: 836
Joined: Thu Dec 18, 2003 3:58 pm

Post by redmonkey »

Yep, nod and acknowledged but, is that technically correct?

Although it's a widely adopted practice and one which I've used myself, you can end up with nested elements serving no other purpose than styling another element. My understanding (and perhaps wrongly so) is that styling attributes of one element should (technically speaking) apply to that element and not affect others surrounding it (unless of course by inheritance). Hmmm, not sure I've explained that very well?

CSS is afterall a layout and styling language, to me the styling of nested elements approach seems to be more of a workaround rather than an intended implementation.
Roja
Tutorials Group
Posts: 2692
Joined: Sun Jan 04, 2004 10:30 pm

Post by Roja »

redmonkey wrote:Personally, I dont agree. As far as I'm aware, the W3C produce 'a' standard not 'the' standard. I'm sure their publications used to be branded 'recomendations' not standards, not sure if this is still the case?
Correct, it is a standard. However, as to it not being the standard, what alternatives compete?

The answer is none. The browser manufacturers produce applications that render html according to the spec. Whether you think the spec is flawed (or not) doesn't change one fact: The browser manufacturers claim to support the standard.

With all the browser manufacturers supporting a single set of standards (the w3c's standards), that makes them the standards.
redmonkey wrote:I see no reason why any browser developer should be expected to adhere to any one set of recomandations over another.
Just like manufacturers of vehicles gain an advantage by having a consistent interface for gas/brake, steering wheel, and more, browser manufacturers gain an advantage by having a consistent interface for rendering html. Thats the benefit of standards.

However, several browser manufacturers have begun to differentiate themselves by embracing ideas beyond (but not in conflict with) the w3c standards. The WHATWG has embraced new technologies like Canvas and SVG, bringing new ideas to browsers and helping to get them included in future standards.
redmonkey wrote:And, if the developers don't agree with a particular implementation of a recomendation, why should they adopt/implement it?
In some cases they don't. There are more than a few examples for each of the major browsers. However, long-term, not implementing those features means they are less competitive than another browser. Thankfully, the browser wars have once again become furiously competitive, and users are benefiting more than ever. A good example is the Acid2 test. Many of the issues it tests for are obscure to be sure.

Yet, browsers like Safari and Opera are setting themselves apart from the pack by aggressively persuing compliance with the standards and fixing the issues present in the Acid2 test.

Doing so empowers webdevelopers by giving them a much more powerful set of tools for development. Thats why the browser makers do it - not just for academic compliance, but to help webdevelopers do more, and more easily.
redmonkey wrote:I think it's good to develop to a standard, but in the absence of any true standard it seems perfectly acceptable to me for the developers to develop to their own standards.
And both are happening. The majority of browser work is done to the w3c standards, and where the w3c cannot come to a consensus, the WhatWG are pushing ahead to develop new consistent implementations of ideas that can later become approved standards by the w3c. It is important to note that many members of the WhatWG are also members of the w3c, so they are far from working against them. :)
redmonkey wrote:I also think many of the problems with cross-browser rendering can be attributed to the developers interpretation of whatever spec they are following. I've read quite a few of the specs and in many areas I feel it is open to interpretation although admittedly, I can't think of any examples off-hand.
I can think of at least one easily: The box model. Many different browsers interpreted the box model concept differently, and the various interpretations resulted in a literal nightmare for anyone trying to do complicated pure-css pixel-precise layout. Thankfully, things have gotten substantially better in the last year, and IE7 looks to lurch us ahead several miles.

But that shows just how important well-written standards are - though everyone mostly followed the standard, it still ended up being a hindrance to developers. If the box model had been consistent across all browsers, we would have had far more exciting CSS layouts in the last few years.
redmonkey wrote:I'd like browsers and online markup validators to parse my own DTDs so I can decide which attributes should and should not be supported/are and are not valid.
That you can already do in xml. Define your own DTD, and you can in fact do just as you said.

Of course, the #1 browser maker doesn't implement/parse xml correctly, so it won't work on 80%+ of browsers.

Worse, IE7 doesn't fix it yet. Maybe you'll get your wish in IE8, if Microsoft follows the standard. :)
redmonkey
Forum Regular
Posts: 836
Joined: Thu Dec 18, 2003 3:58 pm

Post by redmonkey »

I did try experimenting with XML and DTDs previously... it seemed, although I didn't look to closely into it that the common behaviour was to just completly ignore my DTD, fallback in 'quirks' mode and do the best to render a page with an unknown DOCTYPE.

Agreed, that if a browser developer states that they adhere to a particular standard then they should do their best to comply with that statement. Although I do disagree that by only having one standard it should be regarded as 'the' standard by default.

In almost every industry there are products developed which are considered to be a break from the norm/standard and to me, that's what breeds innovation. I'm involved with development of a product at the moment that breaks a lot of the standard rules of the way of working from all previous products, and one of our competitors has essentially thrown out the rule book completely and taken an entirely new approach forcing end users to completely re-learn their entire working style. It's early days yet but both products are being well received. I say more power to browser developers to do what ever they want :D

With regards the car analogy, I could be wrong as it's not an industry I'm involved with but, I'm not so sure there is a standard as such which defines that the layout of pedal controls must be a certain order. I reckon that's just the order that makes most sense and has been adopted by all manufacturers of their own free will. I think that's more down to user expectation more than adhering to a standard, just like the 'back' button of a browser is always at the top left-hand corner of the window.

I've seen motorcycles with the gear lever and rear brake lever on the opposite side of the machine from what may be regarded as the standard, I've seen motorcycles with reverse shift patterns (I've implemented a reverse shift pattern on one of my bikes), doesn't make it wrong, just different and with a good reason for doing so too.
Roja
Tutorials Group
Posts: 2692
Joined: Sun Jan 04, 2004 10:30 pm

Post by Roja »

redmonkey wrote:I did try experimenting with XML and DTDs previously... it seemed, although I didn't look to closely into it that the common behaviour was to just completly ignore my DTD, fallback in 'quirks' mode and do the best to render a page with an unknown DOCTYPE.
Your description sounds more like the behavior of browsers dealing with xhtml that it doesn't understand well, but it wouldn't surprise me. Frankly, with the exception of Firefox, I find all browsers fairly lacking when it comes to xml implementations (including Safari and Opera). (I should note that it is getting substantially better in Opera9, and in the latest nightlies for Safari). Of course, that does leave a rather important browser (IE) fairly far behind the pack.. :)
redmonkey wrote:Agreed, that if a browser developer states that they adhere to a particular standard then they should do their best to comply with that statement.
Exactly. Honestly, I think all of the manufacturers make a good faith attempt to do their best to do so. Unfortunately, Microsoft's failure to release an updated engine for IE in over 5 years has put a severe dent in 'standards' support. Thats the danger of a defacto monopoly - without competition, if the vendor decides not to release an update, there is little we can do. Thankfully, Firefox, Opera and Safari are keeping the heat on Microsoft, and we're finally getting an update.
redmonkey wrote:Although I do disagree that by only having one standard it should be regarded as 'the' standard by default.
Your choice of the word "Should" implies intent. Whether it should or shouldn't be regarded as the standard is not a choice any one body gets to make. It *is* the standard, and no others have stepped forward to challenge that position. The market determined the best choice. Further, its such a niche field, with such high requirements, that the opportunity cost would have to be very large to draw a competitor. Since the w3c is 'good enough' for the vast majority of both browser makers and users, its unlikely we will ever see a true competitor.

But thats not a bad thing. In fact, having competiting standards bodies can deeply fracture the ability to have a standard solution. For example, multiple standards bodies have endorsed various IM protocol designs, including a variant of XMP (Which powers Jabber). Because multiple IM protocols have been endorsed by multiple standards bodies, there is no true "standard". If there were, it might result in an increased push for interoperability. (Although market forces says otherwise in my opinion).

Point being, multiple standards bodies can be detrimental, and in the case of webstandards, a single standards body has been highly beneficial. The only time you don't want a single standards body is if they are holding back innovation. With the presence of WhatWG, that problem has been solved as well. The best of both worlds. :)
redmonkey wrote:In almost every industry there are products developed which are considered to be a break from the norm/standard and to me, that's what breeds innovation.
Sure, and thats why groups like WhatWG are pushing the boundaries out. You are unfairly associating standards bodies with preventing innovation. It may be true in other fields, but it most definitely is not true with the Web.
redmonkey wrote:It's early days yet but both products are being well received. I say more power to browser developers to do what ever they want :D
Unfortunately, thats the mindset that got us marquee tags, blink tags, and a range of inconsistencies so bad that you literally couldnt code a page consistently across even three different browsers five years ago. The world grew up since then, and realized that *billions* of people working together need at least the basic understanding of what a webpage should be. Without those standards, we wouldnt have the web and the features we have today.
redmonkey wrote:With regards the car analogy, I could be wrong as it's not an industry I'm involved with but, I'm not so sure there is a standard as such which defines that the layout of pedal controls must be a certain order.
There are actually several. Of course, not every manufacturer follows them, but the vast majority do. The layout came about originally through market forces, and became defacto standards. Thats the same process powering the WhatWG features like Canvas and SVG, which will become part of the w3c standards as well. (SVG is already mostly complete in the standards)

The current situation is really fairly close to ideal, which is what I was getting at originally. We have a standard, which reduces confusion. We have a working group that helps drive innovation forward beyond what the specs cover, faster than they can document the concept. We even have a process that takes those innovations and helps formalize the industry consensus into a standard after the fact.

Put another way, I don't see much they could improve.
redmonkey
Forum Regular
Posts: 836
Joined: Thu Dec 18, 2003 3:58 pm

Post by redmonkey »

Roja wrote:
redmonkey wrote:It's early days yet but both products are being well received. I say more power to browser developers to do what ever they want :D
Unfortunately, thats the mindset that got us marquee tags, blink tags, and a range of inconsistencies so bad that you literally couldnt code a page consistently across even three different browsers five years ago. The world grew up since then, and realized that *billions* of people working together need at least the basic understanding of what a webpage should be. Without those standards, we wouldnt have the web and the features we have today.
Hmm, yeah, perhaps that statement was a bit on the lose side.

To clarify, I have no problems with browser developers implementing proprietary tags/syntax. I have found (again in other industries) that on occassions a manufacturer will introduce a proprietary method which is later adopted by other manufacturers as it is deemed to be useful/make sense. Of course a browser manufacturer may implement 30 proprietary tags, 29 of which may be deemed useless by others, but one may have some value and may be adopted by other browser developers and make it into future specifications so I think there is value from allowing proprietary tags/syntax. Also, just because it's there doesn't mean you have to use it.
Roja wrote:Put another way, I don't see much they could improve.
I assume you are reffering to the working ethics/approach of the consortium and not the actual spec? I don't know enough (actually anything) of the inner workings of the consortium to be able to comment, although, with regards the specs... there is always room for improvement. :)
User avatar
m3mn0n
PHP Evangelist
Posts: 3548
Joined: Tue Aug 13, 2002 3:35 pm
Location: Calgary, Canada

Post by m3mn0n »

Roja wrote:
redmonkey wrote:Personally, I dont agree. As far as I'm aware, the W3C produce 'a' standard not 'the' standard. I'm sure their publications used to be branded 'recomendations' not standards, not sure if this is still the case?
Correct, it is a standard. However, as to it not being the standard, what alternatives compete?

The answer is none. The browser manufacturers produce applications that render html according to the spec. Whether you think the spec is flawed (or not) doesn't change one fact: The browser manufacturers claim to support the standard.

With all the browser manufacturers supporting a single set of standards (the w3c's standards), that makes them the standards.
Agreed.

Even if it is not the official standard, it's a recommendation/specification that has become the de facto standard.

And to not go into too much detail here, I personally dislike the cross-client incompatibility we web developers have to deal with and how much work people put into making things work on many different browsers. This of course isn't that big of a deal in XHTML/CSS as it is with JavaScript, but it is still something that is so unnecessary and it's frustrating because of how complex it makes things when it could be so very simple.

Instead of it being, "okay, this is how you do this." It's "okay, this is how you do it on this client, this is how it's done on this client, and make sure to include this because this client needs it, and make sure not to include that because it's not supported on this client". I guess that's why I love backend stuff. =P

Anyway, back on track...
redmonkey wrote:with regards the specs... there is always room for improvement.
Agreed. My personal belief is we don't need a new web dev language, aside from the current ones. I wouldn't mind them just continuing the way they're developing.
Gambler
Forum Contributor
Posts: 246
Joined: Thu Dec 08, 2005 7:10 pm

Post by Gambler »

IMO, HTML and CSS are far from ideal. These two languages are used far beyound their originals scope, and that alone makes them inefficient. Besides, people already use Flash, Java and PDF to do things that are theretically doable in HTML.
RobertPaul
Forum Contributor
Posts: 122
Joined: Sun Sep 18, 2005 8:54 pm
Location: OCNY

Post by RobertPaul »

redmonkey wrote:As far as I'm aware, the W3C produce 'a' standard not 'the' standard. I'm sure their publications used to be branded 'recomendations' not standards, not sure if this is still the case?
But surely when all the major browser makers are members ... well, there's something to be said for it, as far as the standard goes...
Gambler
Forum Contributor
Posts: 246
Joined: Thu Dec 08, 2005 7:10 pm

Post by Gambler »

Do a little experiment. Take 10 random links from that list and give it to HTML validator.
foobar
Forum Regular
Posts: 613
Joined: Wed Sep 28, 2005 10:08 am

Re: Just out of curiosity, what would you change in XHTML/CS

Post by foobar »

Roja wrote: The problem isn't web language at all. Its *one* browser manufacturer (who happens to have 80%+ marketshare) not following those standards, and not listening to customer feedback for over 5 years now.
Ironically, M$ cooperates with the W3C. I remember one of the people that drew up one of their web standards was from Microsoft. And I don't think that was a unique case either. But alas, if someone's got their head up their @$$, then there's not much you can do... :?
Roja
Tutorials Group
Posts: 2692
Joined: Sun Jan 04, 2004 10:30 pm

Post by Roja »

Gambler wrote:Do a little experiment. Take 10 random links from that list and give it to HTML validator.
Here's my 10:

http://www.microsoft.com/ - Valid
http://www.opera.com/ - Valid
http://www.mozilla.org/ - Valid
http://www.adobe.com/ - Valid
http://www.att.com/ - Valid
http://www.cern.ch/ - Valid
http://www.loc.gov/ - Valid
http://www.umich.edu/flash.html - Valid
http://www.progeny.net/ - Valid
http://www.verisign.com/ - HORRIBLY invalid

Now, I grant you that I didn't truly choose random choices, because I was most interested in how the large companies (especially the browser makers) were living up to their talk. With the exception of Verisign, I wasn't surprised much. Lots of big companies are coming around to the value of it, and many w3c members are "walking the walk". Nice to see.
foobar wrote:Ironically, M$ cooperates with the W3C. I remember one of the people that drew up one of their web standards was from Microsoft. And I don't think that was a unique case either. But alas, if someone's got their head up their @$$, then there's not much you can do
It wasn't. Microsoft has a long history of endorsing and embracing the w3c standards.

Releasing new versions of their browser was also something they did regularly. However, in the last 5 years, they've fallen behind, because they felt their monopoly position made further development unneccesary. Witness the shift of developers from IE for Mac and Windows to WebTV, then years later closing off IE for mac development entirely.

Its only this year, with Firefox and Opera breathing down their neck that they felt challenged enough to spend development dollars playing catch up so they didn't lose any more users.

To be fair, I have virtually nothing nice to say about Microsoft.
Post Reply