Time for Chrome to go

[Rob Grainger]

(Pedantry warning) You're mixing terms here. A lexical analyser does the lexical level: so the spell checker already does that. A parser (grammar analyser) checks that the usage is valid. We need grammatical analysis that is able to check for "close" lexemes (words).

[Rob Grainger]

Its odd then that IE is the browser that consistently scores lower on standards compliance. I know there's a lot of MS developers here, but to try and pretend that because something doesn't render like IE it is broken is shocking. IE is the single browser that breaks web compatibility more than all others, although improving from IE9 onwards.

[giuchici]

Very good! However, it took a while until somebody noticed. I personally just concluded what was meant from context.

giuchici

[Lost User]

How can Chrome be winning a race that doesn't have a finish line? The standards aren't set to even be completed until 2020. That is really all one needs to know about how goofy web development has become.

[gggustafson]

I have just completed an Article in which I dearly wanted to dynamically create a stylesheet in the Javascript function that initializes the web page. I followed the following track:

function create_stylesheet ( )
  {
  var created = false;
  var head = document.getElementsByTagName ( 'head' ) [ 0 ];
  
  if ( head )
    {
    var style = document.createElement ( 'style' );
    
    if ( style )
      {
      var CSS_string = new String ( );

      CSS_string =
        ".green_link," +
        ".revised_green_link" +
          "{" +
          "}" +
        ".abbreviation_alone" +
          "{" +
          "display:none;" +
          "}" +
        ".abbreviation_definition" +
          "{" +
          "display:inline;" +
          "}" + ...

      style.type = "text/css";
      style.rel = "stylesheet";
      
      if ( style.styleSheet )             // IE
        { 
        style.styleSheet.cssText = CSS_string;
        }
      else                                // W3C
        { 
        var CSS_text = document.createTextNode ( CSS_string );
        style.appendChild ( CSS_text );
        }

      head.appendChild ( style );
      created = true;
      }
    }
  
  return ( created );
  }

The two rules, abbreviation_alone and abbreviation_definition were then to be retrieved from the page stylesheets and pointed to by two Javascript global variables. Although Firefox, IE, Opera, and Safari all performed as desired, Chrome did not. I had to revert to an external CSS file. The only reason for the dynamic stylesheet was to avoid having two files (.js and .css) included in the project. Please see DOM Style Sheets for further details. The Security in Depth: Local Web Pages describes a "feature" that has effectively crippled users of earlier Chrome versions. As a result, folks are suggesting a move away from Chrome. HTH Gus

[alexdresko]

And that's why I've never even installed Chrome.

I'm not a player, I just code a lot! Alex Dresko

[beeseearr]

I think you're missing his point. You would not need to go slower and re-read your posts if you used the same high-quality English everywhere. It is because you have the two versions of typing that it requires extra effort on your part to type correctly the first time through. (Granted, mistakes will still be made, but the frequency should be a lot lower.)

[austin hamman]

its rare for me to have something which works on FF but not Chrome, but very common to have something which works in firefox and chrome but not IE. usually i find i develop something in FF (mostly because of firebug which is better than ie and chrome's built in debuggers) then i work on making it work in chrome (usually works right out of the box, sometimes i need a few minor tweaks because i used FF specific things) then IE (which is a nightmare that adds another month to any development project)

[Reelix]

http://html5test.com/[^] According to this site, the latest Dev build of Chrome is more compliant than any other offical / dev build of any other browser ;P

-= Reelix =-

[Narud Shiro]

Why we have to be worried about the browser when there are amazings 3rd part components fully cross browser compatible? Since I use one of them, I only have to think about which browser would runs faster the apps, instead of think about an app could run in different browsers. :-D

[Narud Shiro]

Have you consider use jQuery? It would make you easier the management of the DOM with js, and it's cross browser compatible :)

[gggustafson]

Although jQuery may be an answer, it's not the answer - fix Chrome!

[Sasha Laurel]

With all of the statistics and pedantic arguments, it feels like the point is being missed. Chrome is a pain for many of us, and it doesn't really feel like we should have to "tip-toe" through the css just to make things look right. Also, arrogance from the dev team is inexcusable, and merely adds to the stereotype that programmers are prima-donna crybabies. My personal opinion is that Chrome sucks, it thinks it knows best, and it really doesn't. I only have it installed because I have to support it to some degree. The call to google remains, fix it, isn't that why you have the luxury of being perpetually "beta"? Throw the web programmers a bone and support existing layouts before trying to forge ahead with an html 5 spec that is far from mature.

[GenJerDan]

So? A browser can be 100% compliant and it won't mean a thing if the sites visited are not written to the standard and display like kaka because of it. And if you think anyone is going to spend the time and money to fix their websites to make Chrome happy... Sure, if you're starting from scratch creating a site, it's no big deal. But to redo everything you already have?

Build a man a fire, and he'll be warm for a day. Set a man on fire, and he'll be warm for the rest of his life. My Mu[sic] My Films My Windows Programs, etc.

[Mark J Miller]

I'm sorry, but most sites are not developed using VS/IIS. Apache has far more market share than IIS. My experience has been similar to yours, all my jobs since 2000 have been on the MS stack, but there are still more java jobs than C# (Dice survey)[^] and apache leads IIS in market share (http://news.netcraft.com/archives/2011/02/15/february-2011-web-server-survey.html[^]) I don't see how you can make that claim. And anecdotally, I know MS developers who develop using FF or Chrome because IE's developer tools were so late to the game and in my experience haven't been as good as the latter two. So given the apache statistics (the java link is a stretch I admit since that doesn't directly correlate with web pages) and the fact that not all IIS developers develop using IE I'd say that your claims are not factual at all.

Code responsibly: OWASP.org Mark's blog: www.developMENTALmadness.com Bill Cosby - "A word to the wise ain't necessary - it's the stupid ones that need the advice."

[gggustafson]

Maybe I should have written my comments using your comments. They are much more succinct and to the point. Thanks, Gus

[Mark J Miller]

Yeah, but W3 schools targets developers and the link was targeted at the comment that most sites are developed using MS tools, specifically IE. This shows that the default (an assumption based on what they are using to visit the W3 site) browser for most developers is not IE. It might be a stretch but I'm willing to bet that developers use their default browser for development and then follow up with other browsers to ensure compatibility afterwards.

Code responsibly: OWASP.org Mark's blog: www.developMENTALmadness.com Bill Cosby - "A word to the wise ain't necessary - it's the stupid ones that need the advice."

[gggustafson]

I think I said "In my experience..."

[Mark J Miller]

Seems like a lot of work just to avoid including .js and .css files in the project. Considering they are standard web application files and they are static and therefore cache-able why spend so much time fighting it? Also, I don't see the argument for leaving Chrome because users of previous versions had a problem. By that argument, no one should be allowed to use any version of IE because of IE 6.0. Plus, IE users aren't forced to upgrade like Chrome users. Chrome automatically updates without any user interaction, so the only ones who would still be stuck on an old version would be those who have done something specific to prevent the update. I wouldn't even know how to do that nor would it make sense to because even though I may want to target Chrome for development, my users will always have the latest version so targeting an old version is an exercise in futility.

Code responsibly: OWASP.org Mark's blog: www.developMENTALmadness.com Bill Cosby - "A word to the wise ain't necessary - it's the stupid ones that need the advice."

[gggustafson]

Seems like a lot of work just to avoid including .js and .css files in the project. Although both .css and .js files are " standard web application files," by dynamically creating the CSS I avoid having to supply two files to people who want to use, say, Green Links. Now, because I try to be browser independent, I must supply both. This complicates the developer's tasks, unnecessarily. I don't see the argument for leaving Chrome because users of previous versions had a problem. The latest version of Chrome significantly modified the way Chrome works. In the toy programming environment this is not a major issue - not many people are affected. But in the production programming environment, this is a serious issue. If you are the development manager for a company that had tied its development to Chrome, had obtained a large number of customers, and then, because Chrome changed, found that your customers could no longer use your site, you would probably feel betrayed. To keep your site alive, you would probably suggest that your customers move to another browser. Even though I may want to target Chrome for development, my users will always have the latest version so targeting an old version is an exercise in futility. Exactly the point. Good development always means backward compatible development. At least in my world it does.