Hello, I bet you know well the error console in Firefox browser, right? Well, i am in the middle of finalizing and releasing a big script application and so i was making sure it is error free. The Firefox error console reports nothing, not even a single notice about CSS or javascript.. nothing. When the site is loaed with IE 6, the statusbar is filled with "Done, but with errors" text. It says page has like a million JS errors. I trust the FF error console more, but still would want to have IE reporting nothing too. What do you think i should do? Should i care at all for IE? Thanks, Ruslan
Does your script function the same way you are expecting it to be in IE with errors?? If it does then probably you are OK but I suggest you to track those errors and fix them. You never know when it may come back at you. Because IE is still used by around 75% or more of the market. (Don't blast messages to me for any IE/Firefox percentage usages, this is based on my stats)
ie is used by about 60% its usage has dropped but as ie7 was released it started growing again yes you should try to optimize it for ie also
Was just going to do that really.. My stats show the opposite. And yes, everything is 100% functional the way it should be in IE. Those reported errors are not affecting the usability of the script. Thanks, Ruslan
As long as it displays and functions correctly in both Firefox and IE, I would not worry about the IE JS errors.
I say code it correctly and run as many test , checks and validators that you can. Then test in as many browsers as possible. Opera, Firefox, etc. Then if IE is the only one acting up ignore it as IE is a piece of shit.
It sounds like you're running afoul of Internet Explorer's document object model/JScript implementation. At which point you have two choices. Ignore it (using just the W3C DOM) or go ahead and modify your scripts so it will play nice with IE's DOM and the W3C DOM. (Off-Topic - Note to Microsoft: We love you too... or at least we would if you created and marketed quality products instead of crap.)
It's far more than 60%. Closer to 85%. But still, they're the majority even if everyone you know uses Firefox - I'm having similar issues at the moment too, but with this in mind, I'm gonna rewrite the code. It sucks - we're back to the days of catering for two browsers..
Sad, but true. It's browsers war again. Anyway, thanks everybody, i will do my best and thanks for your opinions and suggestions. Ruslan
It's really not a browser war per se, just a company that keeps acting like it's 1998 is all. Heck, if I acted like it was still 1998 I'd probably be locked in an insane asylumn (I was a senior in high school at the time).
I designed my site considering IE and it got all messed up in Firefox, so i had to redesign some of its pages again to make it work on Firefox. Compatibility issues, nightmares after we start loving our site in one browser and have to change it to make it work on the other one as well.
While we all must code for IE, it is best to code in Firefox. Firefox is at this time the most W3C html/css and W3C DOM compatible browser in overall terms. IE has its own DOM and a proprietary scripting language (JScript), but it will manage to support the standard DOM and ECMAScript (javascript) for the most part. There are some problems, but until they negatively affect the page, ignore IE warnings. cheers, gary
Dan, the Acid2 test not withstanding[1], I find that per objective testing, Firefox2 has an, oh, so slight margin of compliance over Opera9. That aside, I find that Opera is frustrating as a development tool; seemingly having variations in rendering from build to build. I have made pages that failed in 6, worked in 7, failed again in 8, then worked half-assed in 9, with variations within minor revisions. (Opera 8.x was a royal PITA on that score) Inconsistencies seem to center around margins and around undimensioned float elements. I don't know that they harm the user experience, but these variations harm the developer experience. Of course, YMMV gary [1] Konqueror/Safari pass the acid2 test, but I know they are far from fully compliant; proving, to me, at least, the test is far from representing a pragmatically useful reference.