WebStandards and WebDeveloper's life
-
Soon the CSS 3.0 will arrive. What the perspective to developers? None... :^)Why? The WebStandards are done to exist some default of browser implementation and developer experience. In theory, do you write some in CSS and all browsers will render the same. Then the download of the page will be more quick (why the css will be stored in a separated file). You will easily change completely the layout of the websites, just changing some css selectors, and one plethora of . But in practice is a hard work to do.:zzz: When IE7 was developed (years and years of development), all the WebDeveloper world was with hopes of a compliant browser, what not arrived...:sigh: Microsoft developed C# and .Net, and posted to ECMA; The ASP.Net 2.0/Visual Studio 2005 generate a lot of Xhtml compliant, and appeared what was MS finally was worried in standards. This is not a rant against MS, but against WebStandards... One group of persons unite to create a standard, but then does nothing to these standards to be correctly interpreted by the implementors (each vendor interpret the standards as wishes). It not really a standard... X| If you stop to think, is a ridiculous situation. :doh:In a programming language have not space to ambiguities. "int someValue = 3;" someValue will be 3, even if you use Borland C++ Builder, MS Visual C++, Intel C++ compiler etc... But in CSS, "height: 100%", means a lot of different things, depending of browser what you are using...:~ I know what a lot of MS employes read CP Lounge... Why, oh why MS wish not take a LOT popularity offering a standards compliant browser?:confused: I know what all browsers needs follow this path, but Internet Explorer own +85% of the market, then this iniative will really help webdevelopers, and really will take a lot of aplauses to Microsoft. Wow. I really feel myself better now:cool:. Regards.
For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life.(John 3:16) :badger:
Clickok wrote:
But in CSS, "height: 100%", means a lot of different things, depending of browser what you are using...
Yeah, and completely worthless by "the standards" if you use a margin or padding with a percentage. If I want a div to reach 100% width, I do not want to calculate what border and padding I am use on it and drop the width manual to 99.94% ;) Opps.. There is one of my touchy points, standards that should have NEVER been :)
Rocky <>< Latest Code Blog Post: Vista for Web Development, Read this first! Latest Tech Blog Post: Blog changed to Subtext!
-
I think you are missing the point. Extensions beyond the standard aren't that bad, but when you implement a feature defined in the standard it should match the behavior defined in the standard. If you can't make it work the way it is supposed to then it should be considered a bug that should be fixed.
Using the GridView is like trying to explain to someone else how to move a third person's hands in order to tie your shoelaces for you. -Chris Maunder
In my experience, if the standard is clear and there is no conflict with another part of the standards document, then it seems to be implemented correctly in most browsers. It's places where the standard is left open to interpretation that you see issues. I also see a lot of issues where programmers have used standard elements, but their meaning is confused because they did something else somewhere in their code, which made it impossible for the browser to resolve the situation. In addition to that, the CSS 2.0 standard for example, is so complicated a document that it's not practical to implement everything exactly and the programmers make a choice, in most cases implementing the most popular or most important parts of the standard exactly right, and allowing some freedom in the other parts. Since web browsers have to be back-compatible with standards and some conflicts exist, I think they're doing pretty well. It would be nice if everyone could implement the standard exactly, but that's only possible in cases where a single entity controls both the standard and the implementation. That's the reason PDF works so well, and why browsers will probably never be that consistent.
"Quality Software since 1983!"
http://www.smoothjazzy.com/ - see the "Programming" section for (freeware) JazzySiteMaps, a simple application to generate .Net and Google-style sitemaps! -
In my experience, if the standard is clear and there is no conflict with another part of the standards document, then it seems to be implemented correctly in most browsers. It's places where the standard is left open to interpretation that you see issues. I also see a lot of issues where programmers have used standard elements, but their meaning is confused because they did something else somewhere in their code, which made it impossible for the browser to resolve the situation. In addition to that, the CSS 2.0 standard for example, is so complicated a document that it's not practical to implement everything exactly and the programmers make a choice, in most cases implementing the most popular or most important parts of the standard exactly right, and allowing some freedom in the other parts. Since web browsers have to be back-compatible with standards and some conflicts exist, I think they're doing pretty well. It would be nice if everyone could implement the standard exactly, but that's only possible in cases where a single entity controls both the standard and the implementation. That's the reason PDF works so well, and why browsers will probably never be that consistent.
"Quality Software since 1983!"
http://www.smoothjazzy.com/ - see the "Programming" section for (freeware) JazzySiteMaps, a simple application to generate .Net and Google-style sitemaps!Jasmine2501 wrote:
in most cases implementing the most popular or most important parts of the standard exactly right
Yes, but IE gets enough of the most important parts wrong to be a huge PITA. At least web browser quality level has improved to the point where the debate is about how well the standards are implemented instead of which tag combinations will crash which browsers, like it was when I started doing web development.
Using the GridView is like trying to explain to someone else how to move a third person's hands in order to tie your shoelaces for you. -Chris Maunder
-
Clickok wrote:
But in CSS, "height: 100%", means a lot of different things, depending of browser what you are using...
Yeah, and completely worthless by "the standards" if you use a margin or padding with a percentage. If I want a div to reach 100% width, I do not want to calculate what border and padding I am use on it and drop the width manual to 99.94% ;) Opps.. There is one of my touchy points, standards that should have NEVER been :)
Rocky <>< Latest Code Blog Post: Vista for Web Development, Read this first! Latest Tech Blog Post: Blog changed to Subtext!
That is more a case of using the standard in a way different to what is intended. HTML was for simple documents, not user interfaces. The box model as specced and implemented by compliant browsers works better for documents than for UI. As IE implemented it in older versions was more along what you want.
regards, Paul Watson Ireland & South Africa
Shog9 wrote:
I don't see it happening, at least not until it becomes pointless.
-
Jasmine2501 wrote:
Vendors should support the standard, but innovation outside of that standard leads to progress.
Yes, but it does not avoid what by example, c++ compilers talks the even same languages and follows the standards. Since what follow the minimal defaults, all compilers have freedom do innovate, as stated by CG in the previous reply-post. IE does not follow the standards to innovate, but because is buggy... it is sad but is true. Don't take me bad, I'm not against IE. I'm against not follow the minimal standards. Even Opera has some inconsistencies with the standards too...
For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life.(John 3:16) :badger:
Try cross-platform development in C/C++ then get back to us :D
Todd Smith
-
That is more a case of using the standard in a way different to what is intended. HTML was for simple documents, not user interfaces. The box model as specced and implemented by compliant browsers works better for documents than for UI. As IE implemented it in older versions was more along what you want.
regards, Paul Watson Ireland & South Africa
Shog9 wrote:
I don't see it happening, at least not until it becomes pointless.
Yeah I think you hit it right on the head there. The standards don't address all possible documents, but the implementation has to.
"Quality Software since 1983!"
http://www.smoothjazzy.com/ - see the "Programming" section for (freeware) JazzySiteMaps, a simple application to generate .Net and Google-style sitemaps! -
Try cross-platform development in C/C++ then get back to us :D
Todd Smith
I don't think the comparison to cross-platform C++ is a good-one. Think about Java programming. Java runs on a VM (Same as JavaScript, CSS and HTML run in the browser). Think you had to write Java code like:
if System.OS.equals("Windows") // Do this... else if System.OS.equals("Unix") // Do exactly the same, just a little bit different...It doesn't make much sense, does it? -
There are absolutely parts of the C++ standard that are not defined, leaving a compiler free to deal with them as it chooses.
Christian Graus - Microsoft MVP - C++ Metal Musings - Rex and my new metal blog
Yeah, for example function decoration is not specified. MS used this as an argument to develop COM (back in the 90s), because they wanted binary compatibility, not only source compatibility. They claimed they had asked for a standard, and the request was denied. So C++ developers were punished with over a decade of COM programming...
-
Try cross-platform development in C/C++ then get back to us :D
Todd Smith
He, he... cross-platform? So far the best answer to this problem is... Windows :) if (document.layers) { // One browser } else if (document.all) { // Another one }
-
That is more a case of using the standard in a way different to what is intended. HTML was for simple documents, not user interfaces. The box model as specced and implemented by compliant browsers works better for documents than for UI. As IE implemented it in older versions was more along what you want.
regards, Paul Watson Ireland & South Africa
Shog9 wrote:
I don't see it happening, at least not until it becomes pointless.
yeah, that is the problem, they all looked at anything to do with presentation and called it "suggestions". We live in a world of browsers trying to implement user interfaces for a standard that was based for "document" hanlding. A lot like using Excel for you next programming IDE, you might be able to do it, but who would want to. The standards box model does not work for anything, it was poorly designed and should have been dropped in favor of a MS box model long ago. The current box model cannot handle percentages and other presentation elements at the same time. But developers have to pay the price for people that clearly did not pay attention or worse yet, had pride issues as their "document" standards was becoming the basis of a "user interface platform". Anyway, still a sticking point with me, I have spent hundreds or possibly thousands of hours of work of the years making sure different browsers can display information that same. Compound that by the hundreds of thousands if not millions of developers that have had to deal with the same issue. That is a lot of wasted time for a bad standard.
Rocky <>< Latest Code Blog Post: Vista for Web Development, Read this first! Latest Tech Blog Post: Blog changed to Subtext!
-
yeah, that is the problem, they all looked at anything to do with presentation and called it "suggestions". We live in a world of browsers trying to implement user interfaces for a standard that was based for "document" hanlding. A lot like using Excel for you next programming IDE, you might be able to do it, but who would want to. The standards box model does not work for anything, it was poorly designed and should have been dropped in favor of a MS box model long ago. The current box model cannot handle percentages and other presentation elements at the same time. But developers have to pay the price for people that clearly did not pay attention or worse yet, had pride issues as their "document" standards was becoming the basis of a "user interface platform". Anyway, still a sticking point with me, I have spent hundreds or possibly thousands of hours of work of the years making sure different browsers can display information that same. Compound that by the hundreds of thousands if not millions of developers that have had to deal with the same issue. That is a lot of wasted time for a bad standard.
Rocky <>< Latest Code Blog Post: Vista for Web Development, Read this first! Latest Tech Blog Post: Blog changed to Subtext!
Fair enough, poor implementations of the standard around poorly detailed points of the standard have cost us all a lot of time. Though it has gotten better, I haven't had a major browser rendering difference issue in about a year now. It has become easier. Not to boast at all, I am sure you have found it too, that skill makes a huge difference. New people to HTML and CSS struggle a lot with browser issues while those of us who have been doing it a while tend to instinctively avoid the major ones and can easily fix the smaller ones. I just haven't seen anything better yet. No other implemented standards exist and even the unimplemented ones don't seem very good. I am quite amazed sometimes that this whole world wide web thing works what with the hundreds of millions of different computers out there using it. HTML and CSS are crap but, strangely, they work. I'm not sure any other standard would have done a better job. It seems to me to be more a human fault in setting and using the standard.
regards, Paul Watson Ireland & South Africa
Shog9 wrote:
I don't see it happening, at least not until it becomes pointless.
-
Fair enough, poor implementations of the standard around poorly detailed points of the standard have cost us all a lot of time. Though it has gotten better, I haven't had a major browser rendering difference issue in about a year now. It has become easier. Not to boast at all, I am sure you have found it too, that skill makes a huge difference. New people to HTML and CSS struggle a lot with browser issues while those of us who have been doing it a while tend to instinctively avoid the major ones and can easily fix the smaller ones. I just haven't seen anything better yet. No other implemented standards exist and even the unimplemented ones don't seem very good. I am quite amazed sometimes that this whole world wide web thing works what with the hundreds of millions of different computers out there using it. HTML and CSS are crap but, strangely, they work. I'm not sure any other standard would have done a better job. It seems to me to be more a human fault in setting and using the standard.
regards, Paul Watson Ireland & South Africa
Shog9 wrote:
I don't see it happening, at least not until it becomes pointless.
Yeah, I agree it works but it takes work to make it work.. That's a lot of work ;)
Paul Watson wrote:
I haven't had a major browser rendering difference issue in about a year now.
You are mighty lucky! I went round-and-round not long ago with Opera since they decided to render a couple pixel border on their "form" tag. That took days to find. Of course that was not due to standards, but it sure was a browser rendering difference. As far as CSS goes, as long as you stay away from the Quirks Mode, most late model browsers will render close to the same. But then you are stuck with the stupid box model standard on all browsers :)
Paul Watson wrote:
I'm not sure any other standard would have done a better job.
I do not know about "another" standard, but just changing the box model in the CSS standard alone would have greatly improved the standards. Oh well, guess we are heading for ActionScript or XAML site eh? :)
Rocky <>< Latest Code Blog Post: Vista for Web Development, Read this first! Latest Tech Blog Post: Blog changed to Subtext!