SitePoint Sponsor

User Tag List

Page 2 of 2 FirstFirst 12
Results 26 to 34 of 34
  1. #26
    Non-Member bronze trophy
    Join Date
    Nov 2009
    Location
    Keene, NH
    Posts
    3,760
    Mentioned
    23 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by Silver Firefly View Post
    I use an analytic app to determine whether I need to be catering for IE6 at all. So far, there hasn't been any need to cater for IE6 or below.
    Which "analytic" would that be... if you don't support it, how can you trust any tool to tell you if you need it since your own demographics are probably only listing bounces.

    I see that one all the time -- "I don't see any real traffic from IE6, so I don't support it"

    Ever occur to you that you don't see any real traffic from IE6 because you don't support it?

    I still just don't get where people think supporting IE6 is 'hard'... but then I don't understand how people can vomit up 50k of markup for 5k of plaintext either.

  2. #27
    Non-Member bronze trophy
    Join Date
    Nov 2009
    Location
    Keene, NH
    Posts
    3,760
    Mentioned
    23 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by rpeg View Post
    Can you recommend any books to get a novice (such as myself) up to speed. PDFs or anything that cover the level of details you're referring to.
    Sadly they are far and few between. Forgetting everything you THINK you know about markup and reading Build Your Own Web Site The Right Way Using HTML & CSS is a good start, if you ignore like the last third of the book where he suggests pointless crap like Google Analytics and so forth.

    Really though there isn't THAT much to supporting legacy IE if you follow a few simple rules and learn one or two minor fixes...

    1) Avoid declaring width and/or height the same time as padding or border. Even though haslayout fixes the box model in theory, you still cannot rely on it.

    2) use a simple reset that just turns off margins and padding on common elements and borders on IMG and FIELDSET. Do NOT use the "universal reset" as that shtups form elements, and do NOT use fat bloated train wrecks of idiocy that may as well be a framework like "reset reloaded".

    3) Don't use CSS frameworks, by definition they use presentational classes defeating the POINT of using CSS in the first place.

    4) Avoid putting comments BETWEEN elements as this can cause rendering errors in IE and FF. This means pointless comments like:

    <!-- start content -->
    <div id="cc">

    Is a miserable /FAIL/. Not only for the vague/meaningless classname but for the comment that would be unneccessary if the ID name made any sense.

    <div id="content">

    All that's needed. Opening a tag is the start of a section? Wow.

    Likewise on the close you see this idiocy all the time too.

    </div><!-- end content -->

    Closing a tag ends a section of code? Who'd have thunk it?!? Putting the comment after could put it between floats, where all sorts of bad things tend to happen. Solution? Don't waste time saying end, and move the comment before the close perhaps saying if it's a class or ID you are closing.

    <!-- #content --></div>

    No rendering bugs, clear as a bell what's going on!

    When comments can trip rendering issues, changing your formatting and commenting style is NOT arbitrary formatting, no matter how much the lazy folks who just sleaze out pages any old way claim it is.

    5) Learn haslayout -- not just it's triggers, but when it's going to be needed... When is it needed? If you are absolute positioning an element inside a relative parent -- that parent needs haslayout. If you are using a bottom margin on a float the element AFTER the float needs haslayout... and if you have a display:block element surrounded by inline-level elements inside a parent with a background, that inner block may erase the background when scrolling unless the parent "haslayout"

    That covers 99% of the times you "need" a haslayout trigger.

    6) margins doubling on floats? Display:inline on the float gets rid of it. Floats are inherently display:block and changing the display: value cannot change that -- but for some reason display:inline makes IE's margins on floats behave properly.

    7) Do NOT waste your time on outdated/half-assed float clearing methodologies like clearing DIV or worse, clearfix classes. Pretty much every case where float wrapping behavior is desired can be handled either by floating the wrapper and giving it a width, or by using overflow:hidden with a haslayout trigger.

    8) You MAY have to resort to an "expression" to fake one or two missing properties like min/max-width/heights. Really that's the only time things like that should be used.

    From there a few simple things like "never put text that goes to multiple lines in a fixed height container" and "always declare your content in a dynamic font" (%/em or even pt) can go a long ways not just towards making IE behave, but also making your page accessible and not broken just because a user is on a different system metric. (aka default font size).

    It's not that hard, and if you work that way from the start supporting IE6 or even 5.x on the same codebase as everything else -- and that's WITHOUT resorting to that stupid malfing useless IE conditional crap bloating out the markup on every page. Most every time I've seen someone resort to IE conditional comments for CSS or markup, it's to hide how poorly written the markup and CSS are in the first place!

    Quote Originally Posted by rpeg View Post
    Thanks and please don't chew me out. You seem to be in a bad mood.
    Still don't get where people get that either... I'm in a cheery mood most of the time -- I just happen to call things manure when it smells like a farm in Vermont.

  3. #28
    SitePoint Evangelist Karpie's Avatar
    Join Date
    Jul 2007
    Location
    Perth, Australia
    Posts
    445
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by deathshadow60 View Post
    7) Do NOT waste your time on outdated/half-assed float clearing methodologies like clearing DIV or worse, clearfix classes. Pretty much every case where float wrapping behavior is desired can be handled either by floating the wrapper and giving it a width, or by using overflow:hidden with a haslayout trigger.
    This is the only one of those I disagree with. If you want to float an element to wrap it's children floats, where does the floating end? What if you want to clear the floated parent? You'll end up floating half your page for no reason at all.

    Overflow:hidden I'll use for very simple things, but when you start adding things like hover elements on top, it quickly becomes unfeasible.

    I quite like having a clearfix class (though I don't call it that, it's a stupid presentational name), it adds nothing to the markup and it works perfectly.

    Anyways back on topic...

    I've just been given the go-ahead for the very first time to not worry about making a new site pixel perfect in IE6. It still has to be usable, functional, and accessible, but apart from that, no big deal.

    This may just be one of the happiest days of my life.

  4. #29
    Non-Member bronze trophy
    Join Date
    Nov 2009
    Location
    Keene, NH
    Posts
    3,760
    Mentioned
    23 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by Karpie View Post
    This is the only one of those I disagree with. If you want to float an element to wrap it's children floats, where does the floating end? What if you want to clear the floated parent? You'll end up floating half your page for no reason at all.
    No reason? Float clearing is no reason? Besides, MOST of a website ends up floated (or should) once columns are invovled, so ONE measly extra declaration in the CSS is a problem HOW exactly?

    Quote Originally Posted by Karpie View Post
    Overflow:hidden I'll use for very simple things, but when you start adding things like hover elements on top, it quickly becomes unfeasible.
    Which I suggest avoiding hovers in the first place, or using some other method to avoid the need for float clearing.

    Quote Originally Posted by Karpie View Post
    I quite like having a clearfix class (though I don't call it that, it's a stupid presentational name), it adds nothing to the markup and it works perfectly.
    bloated pointless OUTDATED presentational markup -- at that point you might as well start using moronic classes like "left", "right" or "bigtext".

    Quote Originally Posted by Karpie View Post
    I've just been given the go-ahead for the very first time to not worry about making a new site pixel perfect in IE6. It still has to be usable, functional, and accessible, but apart from that, no big deal.
    Pixel perfect is a myth that should NOT be in the planning on any site in the first place -- though try to convince the PSD jockeys with their "Accessibility, what's that?" attitude.

  5. #30
    I solve practical problems. bronze trophy
    Michael Morris's Avatar
    Join Date
    Jan 2008
    Location
    Knoxville TN
    Posts
    2,023
    Mentioned
    63 Post(s)
    Tagged
    0 Thread(s)
    IE 7 is our current floor, and I'm beginning a year long presentation layer rewrite that will set the floor at IE 8. Part of that rewrite is removing all the blasted layout tables and pulling the inline css out of the controller layer when it appears.

    IE 8 is the new IE 6 -- the bare minimum floor we have to support for the next 10 years thanks to the fact it will take that long for all the XP machines to get out circulation.

  6. #31
    SitePoint Wizard
    Join Date
    Oct 2005
    Posts
    1,832
    Mentioned
    5 Post(s)
    Tagged
    1 Thread(s)
    Quote Originally Posted by deathshadow60 View Post
    Do I dare look at the other links? Oh look, the second one doesn't even mention the easiest fix for the duplicate characters/dissapearing content bugs... move the comment INSIDE the preceeding float instead of putting it in-between them.

    </div><!-- end id column1-->
    <div id="column2">
    == FAIL

    <!-- #column1 --></div>
    <div id="column2">
    == No IE or FF bugs.
    Still a bug nonetheless. And not one that most people would figure out right away. Comments should be allowed to be placed anywhere without any problems. And I've never had a problem with this in Firefox.

    Quote Originally Posted by deathshadow60 View Post
    Of course, I still don't know where the blue blazes people get it takes "so much time" to support IE6, unless of course they are building their pages with flawed methodologies or aren't telling the PSD jockey where to stick it when it comes to certain non-viable design elements.
    Because it does take time--time better spent doing other things than finding workarounds for a buggy and insecure 10-year-old browser. And unless you are an expert, you sure as hell aren't going to know what to do to fix these stupid IE6 bugs. It is going to take hours for a newbie to get up to speed on crappy IE6. If you started doing web development in 1999, you are going to have grown up in the era of buggy IE and would have had no option but hack for it because IE controlled 90% of the market share. Someone starting in 2009 isn't going to have that experience.

    IE6 came out in 2001. It took years for some of these hacks to be discovered.

    IE5/6 Doubled Float-Margin Bug - CSS fixes and workarounds

    Up 'til now (Jan '04) this bug was thought to be unfixable, and was generally controlled by replacing the faulty margin with a left padding on a non-visible float, along with a nested inner box to serve as the visible box within the invisible float, or by hacking a one-half margin value for IE/win only. It works, but is clunky and messes up the clean source code. Well that's all over now.
    To hell with IE6! I am using a 10 year old computer with only 256 MB of RAM right now and Firefox 4 works great in it. (Firefox 4 is much faster than FF3.) Safari, Opera, and IE7 work fine, too. I don't care about crappy corporations and their outdated software that is dependent on IE6--they can install other browsers in addition to IE6. I don't care about idiots who can't pull their heads out of their butts long enough to spend a few minutes downloading an updated browser. These people are the dregs of internet society and I am not doing a thing to support their antiquated butts.

    I'm certainly not supporting the visits to my sites from IE2, Netscape 0.6, or Safari 1.3.2 either. It's time to move forward. IE6 needs to be taken out behind the barn, shot in the head, and put out of its misery.

    Supporting IE6 only gives people an excuse not to upgrade. If corporations need to use IE6, they can install Firefox or another modern browser and tell their employees to use IE6 only when necessary for work purposes and use the good browser for web viewing. Now we are left with Grandma using Windows 95...

    How come Microsoft didn't just fix the bugs in their IE6 rendering engine?

  7. #32
    SitePoint Wizard
    Join Date
    Oct 2005
    Posts
    1,832
    Mentioned
    5 Post(s)
    Tagged
    1 Thread(s)
    Quote Originally Posted by Michael Morris View Post
    IE 8 is the new IE 6 -- the bare minimum floor we have to support for the next 10 years thanks to the fact it will take that long for all the XP machines to get out circulation.
    Microsoft made the conscious decision not to support Windows XP with IE9 because they wanted to drive sales of their operating system.

    Firefox supports XP, so does Safari, Chrome, and Opera. Estimates are that 45% to 55% of computers are still using XP. The difference between other browser vendors and Microsoft is that Microsoft has a financial incentive in rendering XP obsolete any way possible. Even if that means deliberately making sure that new software will not work with XP.

    I've never used IE7 yet. I'm currently on an XP machine and I've used Vista and was completely unimpressed. I see no reason to change operating systems at the moment. Neither do millions of other computer users. I'm not paying $300 for a new OS just because Microsoft wants my money. I'll wait until I get a new computer for a new OS. And I will miss XP. It's been good to me.

  8. #33
    SitePoint Wizard
    Join Date
    Oct 2005
    Posts
    1,832
    Mentioned
    5 Post(s)
    Tagged
    1 Thread(s)
    Quote Originally Posted by deathshadow60 View Post
    5) Learn haslayout -- not just it's triggers, but when it's going to be needed...
    Thanks for the link. The page at that link links to Microsoft which then links back to that page at satzansatz.de.

    I swear it's almost as if Microsoft has very little idea how their own buggy products work.

  9. #34
    Non-Member bronze trophy
    Join Date
    Nov 2009
    Location
    Keene, NH
    Posts
    3,760
    Mentioned
    23 Post(s)
    Tagged
    0 Thread(s)
    Quote Originally Posted by cheesedude View Post
    I swear it's almost as if Microsoft has very little idea how their own buggy products work.
    Notice they also took a page from Mozilla's book, and call it a "feature".


Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •