1. Advertising
    y u no do it?

    Advertising (learn more)

    Advertise virtually anything here, with CPM banner ads, CPM email ads and CPC contextual links. You can target relevant areas of the site and show ads based on geographical location of the user if you wish.

    Starts at just $1 per CPM or $0.10 per CPC.

IE 6.0 - not displaying page correctly!

Discussion in 'CSS' started by Fafles, Sep 9, 2007.

  1. #1
    Edited for privacy.
    SEMrush
     
    Last edited: Mar 18, 2015
    Fafles, Sep 9, 2007 IP
    SEMrush
  2. Stomme poes

    Stomme poes Peon

    Messages:
    3,195
    Likes Received:
    136
    Best Answers:
    0
    Trophy Points:
    0
    #2
    If I were you, I'd immediately begin looking for a new webmaster. For a business you need to run, you either need to hire someone who knows more than your other guy, or learn yourself (which can be done, but takes a lot of time and I haven't met a small-business-owner yet who wasn't already strapped for time).

    One simple easy fix MIGHT be removing the space at the very top of the html code. IE will go into quirks mode even at something as simple as someone hitting the [enter] key before typing in the <!DOCTYPE line...
    ...meaning you need to do this, but it may or may not fix the problem(s).

    The site looks simple and pretty but took a suspicious amount of time to load. I'm used to pages loading in under 2 seconds usually, and yours took a good 10-15 seconds... I dunno it's it's all the scripts or what.

    It looks to me like your site's code could be really simplified, and I also noticed some things like using images in your lists (home, newsletter) instead of text. That can look nice, but if someone's browser has images turned off, they won't see anything at all! There are many techniques out there for having both text with images sitting on top of them-- check out the last 2 examples at http://www.mezzoblue.com/tests/revised-image-replacement/.
     
    Stomme poes, Sep 9, 2007 IP
  3. Fafles

    Fafles Member

    Messages:
    2
    Likes Received:
    0
    Best Answers:
    0
    Trophy Points:
    36
    #3
    Hi!

    Thanks for replying.

    I tried removing the space you mentioned and it did not seem to work. Any other suggestions? :confused:

    Again, thank you so much and kind regards,
    Ana
     
    Fafles, Sep 9, 2007 IP
  4. Philopoemen

    Philopoemen Peon

    Messages:
    704
    Likes Received:
    19
    Best Answers:
    0
    Trophy Points:
    0
    #4
    I recommend the same thing as [Stomme poes] did. You can't run such a website without at least an average webmaster with some knowledge. It will be faster and safer for you.
     
    Philopoemen, Sep 10, 2007 IP
  5. qube99

    qube99 Peon

    Messages:
    75
    Likes Received:
    1
    Best Answers:
    0
    Trophy Points:
    0
    #5
    I build ecommerce websites and always use a separate stylesheet for IE 6.

    Here's the header I use:

    
    	<link href="http://website.com/mm5/css/stylesheet.css" rel="stylesheet" type="text/css">
    	<!--[if IE 6]><link rel="stylesheet" type="text/css" media="screen" href="http://website.com/mm5/css/ie6.css"><![endif]-->
    
    Code (markup):
    The first calls stylesheet.css which is for compliant browsers. The second calls ie6.css which is the stylesheet for IE 6, naturally.

    Get it all working in compliant browsers first, and get it validated. Then copy the stylesheet and revise it for IE 6.

    You can tell from the mm5 directory I work with Miva Merchant. The actual code has a token for the domain name which switches from non-secure to secure as needed. If the stylesheets are not linked correctly you can screw up your site security, so be very careful here.

    Anyways - that's a pretty foolproof method for dealing with IE 6 parsing bugs. It has the advantage that it will not break in future browser releases the way the star hack did when IE 7 was released.

    I give my customers the contact info for 2 other HTML savvy Miva developers they can call on if I'm not available (I serve as their backups, too). Perhaps your webmaster hasn't been "burned" enough for full corporate engagements. The war doesn't stop just because your webmaster does, eh? In the modern world webmasters are simply temporary plug-n-play components of your overall system.
     
    qube99, Sep 10, 2007 IP
  6. Stomme poes

    Stomme poes Peon

    Messages:
    3,195
    Likes Received:
    136
    Best Answers:
    0
    Trophy Points:
    0
    #6
    You're certainly not the only one to do it that way, qube, but it really is better when possible to write the whole site semantically and then deal with IE6 while still using the one stylesheet. No, I don't really worry that the browsers have to download not one but two stylesheets-- no real biggie there, it's just that for every change, it must be done twice and as you said about security (I assume you mean when someone purchases something on her site and must be taken to an https)...

    She should (I think) find a good webmaster who knows how to write semantically (can write the site in XHTML), knows the major browsers well, and properly uses scripts and CSS. The guy who half-built this site didn't think once about people with images turned off, didn't notice he has a half-gazillion scripts doing who knows what (nothing wrong with scripts per se, but if they're doing work that HTML and CSS can already do, lose the scripts). S/He should be aware of hacks, but use them sparingly. He should know common tricks, and use them sparingly as well. The only good thing I can say about her web guy is that her site is quite pretty. I like the design a lot.

    I personally consider two stylesheets a sort of last resort, for instance when the site MUST be online like tomorrow and there's no time to re-think the CSS... I'd use the two sheets temporarily until I could straighten things out.
     
    Stomme poes, Sep 12, 2007 IP
  7. qube99

    qube99 Peon

    Messages:
    75
    Likes Received:
    1
    Best Answers:
    0
    Trophy Points:
    0
    #7
    I agree that using 2 stylesheets could be viewed as a pain by someone who doesn't use the technique. In practice it is the fastest, most efficient and most accurate way to handle IE 6. I think if you studied up on how dual stylesheets work and how precedence is used on them you'd have a far different view on the matter. Every change is not made on both stylesheets, in fact very few require this and once you understand IE 6 bugs you can predict which one will require different CSS.

    My suggestion would work for someone who had no understanding at all of IE 6 bugs and dual stylesheet precedence rules. It's a blanket solution for a very fast, quick fix that's also accurate and reliable.

    So let me ask you about using 1 stylesheet -

    If one browser adds padding inside a frame (IE 7) and another adds it outside the frame (IE 6) what CSS do you propose to give matching results? The Holly hack is out because it doesn't work in IE 7 and people that used it had to recode their websites when IE 7 was released. In fact, all those types of hacks are poor coding because they are unspecified in the standards.

    Also - IE 7 properly handles hover in UL menus but IE 6 does not. What valid CSS could you possibly write that would work in both browsers? I've never seen nor heard of any. The only solution is to write separate CSS for each browser and additionally separate HTML. Just like we used to do during the Browser Wars.

    I could also bring up the failure of IE 6 to parse floats correctly and a large number of other non-compliancies and parsing bugs.

    Please don't propose using Quirks mode to invoke the Triton engine. That breaks all the other web browsers out there.

    The way you use the word "semantic" indicates to me that you may have an incorrect understandng of it. Semantically written code absolutely will not "fix" even a single IE 6 problem in any way, form or fashion. I don't care how you semantically theme a collection of divs or any other elements, it won't make IE 6 compliant or fix it's parsing bugs. Semantic theming won't alter its behavior one whit.

    IE 6 will be gone in a couple of years. I can hardly wait.

    I totally agree that the so-called webmaster is an amateur with no understanding of professional websites and broad audiences. He also has little understanding of how businesses function and how they are interrelated to their websites. The business owner should replace him in order to protect his income.
     
    qube99, Sep 12, 2007 IP
  8. deathshadow

    deathshadow Acclaimed Member

    Messages:
    9,375
    Likes Received:
    1,853
    Best Answers:
    245
    Trophy Points:
    515
    #8
    BULL. Utter and total NONSENSE!!!

    Because before you dive for browser specific nonsense you should first try a different technique be it reworking the HTML or the CSS... there are usually four to ten different ways of doing EVERYTHING in HTML/CSS - chosing the one that works cross browser without hacks instead of diving for the hacks to get buggy code to actually work is NOT the solution.

    And when you DO need to actually hack, it's better to put the hack NEXT to the code in the same file, instead of scratching your head a month down the road wondering why changes to your main stylesheet isn't picking up in that extra file... 99.9% of the time the only thing you end up hacking for can be corrected by dealing with haslayout, box model correction, or using height instead of min-height, all of which can usually be hidden behind 'easy selectors' like * html.

    No, it's a way to make bloated confusing code and to increase the number of handshakes increasing server load too.

    Uhm, use a valid doctype so IE isn't in quirks mode? That's quirks mode behavior... Much less the entire RoW (Rest of world) uses the behavior you describe to IE7 so hide the corrected width behind the * html 'easy selector'. One line of code in the CSS right next to the rest of it instead of three lines in your html and a wasted handshake and file transfer.

    csshover2.htc - end of problem... Especially since anyone bright enough to turn javascript off is likely smart enough not to still be using IE.

    Sorry, but your arguement is utter rubbish. Instead of diving for hacks and wasting time building an entire separate file, try using code that works in all browsers or inlining the one or two minor fixes that are needed to bring IE6 into compliance.

    Most of the time when people resort to separate stylesheets they are just bloating out the page when all they need to do is rewrite the HTML and CSS to something cleaner/simpler. Oh Noes, you might need to use a extra 'meaningless container' like DIV or Span. (or you might need to use three or four LESS DIV's, SPAN's and Classes - which seems more common in my experience)

    Now that on the other hand I can agree with wholeheartedly. That code is a train wreck from hell - typical of the projects I usually end up on where I have to tell the person I need to throw it all out and start over from scratch.

    Because quite frankly, that page is broken on my machine in every browser except firefox. Opera, Safari and IE7 don't like it one damned bit either. (Opera and IE7 I'm fairly sure it's because I'm running 'large fonts' here)
     
    deathshadow, Sep 12, 2007 IP
  9. qube99

    qube99 Peon

    Messages:
    75
    Likes Received:
    1
    Best Answers:
    0
    Trophy Points:
    0
    #9
    Yes, valid code without using any non-compliant hacks.

    You just contradicted yourself.

    IE 6 has a serious parsing bug with the star selector. This is not cross-browser compatible code.
    www.info.com.ph/~etan/w3pantheon/style/starhtmlbug.html

    Once again you advocate the use of a broken selector that won't work properly in IE 6.

    As I mentioned before, you don't understand the use of multiple stylesheets. There are no extra HTML calls on the server, each browser only downloads one stylesheet.

    And yet another non-compliant IE hack plus you want to tell your customers to go get another browser. That's total suicide in ecommerce.

    It's obvious you don't comment your code or use style guides. You'd get fired for that kind of amateur crap in our shop.

    In many common ecommerce projects the main code is structured like a plate of spaghetti with all sorts on nested, looping conditionals. The ONLY way to keep it all sorted out is to (1) use multuple stylesheets. Not 2 of them, but sometimes 15 or 20 and even more. (2) Use industry standard code comments and (3) use a professionally prepared styleguide. Without these quality controls it might take someone like you days to locate the correct selector for a link style or a div. It might take you days to figure out where the link goes because of all the conditionals and dynamic scripting that generates the link.

    So yes, this powerful technique can be disastrous when used by an amateur. In professional projects it is required, not optional. I work on a very large, international website that uses over 50 developers from around the world. I have no idea how many pages it has but it's certainly over 1 million. The project is successful because of the abovementioned quality controls. This thing has more stylesheets than most websites have pages.

    And there's not a single star selector or Holly hack in it anywhere.
     
    qube99, Sep 12, 2007 IP
  10. deathshadow

    deathshadow Acclaimed Member

    Messages:
    9,375
    Likes Received:
    1,853
    Best Answers:
    245
    Trophy Points:
    515
    #10
    Did you actually READ? You should pay particular attention to the words 'before' and 'try'. You should TRY before hacking. There are some things like a lack of min-height that are the reason you need to send IE 6.x and earlier different code.

    And you apparantly also didn't even understand the page you linked to. It's BECAUSE of that bug that we can use it as an 'easy selector' since such code is ignored by other browsers, but is still considered valid CSS. That's the whole point.
    http://www.webdevout.net/css-hacks

    Read up on the section called 'easy selectors'

    Actually no, you apparantly don't understand the * html hack - it only works in lte IE6 - all other browsers ignore it.

    Example:
    body,html {
    height:100%;
    }
    #container {
    min-height:100%;
    }
    * html #container {
    height:100%; /* treated as min-height in lte IE6 */
    }

    The #container min-height declaration fires in FF, Opera, Saffy and IE7 - but does nothing in lte IE6. The * html #container declaration only works in IE6 and earlier... which is EXACTLY what we want there (since lte IE6 treats height as min-height)

    IF you use the full four conditional sets, and have each different stylesheet have FULL copies of the CSS for the page, then yes. Enjoy your extra 400 bytes of code for something that could be inlined. Otherwise you have a minimum of TWO sheets being sent per client. BOTH ways it's a waste of code, sloppy programming and a maintennance nightmare that you end up spending five times the manpower to maintain than should be needed.

    If it validates, which all of what I'm referring to does - and is future proofed since unlike SOME people I've never had to run around like a chicken with my head cut off because of a new browser (see IE7 - have fun revisiting all your sites to code a new stylesheet did you?) then where's the problem? I'm not sure where you're geting 'non-compliant', or how it's suicide. Suicide is cutting your profit margin to nothing by wasting manpower by making something twenty times more complex than it needs to be.

    I comment code that NEEDS commenting - frankly if you write your HTML properly with semantic code and meaningful classnames that say WHAT things are, and INDENT properly (I prefer Wirth style, but AT&T style is acceptable too) most of the time you don't need comments unless you are doing something out of the ordinary (like a holly hack, IE 5.x centering tricks, etc)

    Quite frankly if you are doing **** like this:
    <!-- start content -->
    <div id="content">

    /FAIL/ at intarweb.

    and here's the PROOF that you need to back away from the keyboard:
    Am I to understand you are ADVOCATING Spagetti Code? Mein gott no wonder most of these off the shelf packages are such total CRAP on their HTML output. What's next, going to tell me the client gets billed by the K-LoC? What is this 1978? You working in Fortran or something?!? Oh wait, my bad. It has to be DiBol.

    So pretty much ignoring ALL of the advantages of CSS (like precaching), ditching any attempt at code simplification and minimalist markup, and in general churning out crap 1997 style. BRILLIANT.

    You mean like say... Wirth style indenting? Nun with a wooden ruler? There's a reason a lot of 'style guides' make me want to puke - this new fangled variation away from AT&T STyle or Wirth Style is utter rubbish... Possibly because most of them are repackagings of AT&T style that some jackass feels they can actually charge time for developing when they copied it out of a 1980's programming book.

    Because of course having to churn through 20 different CSS files just to find the one you are looking for, and then to have to edit twenty separate files just to change the one is SUCH a time saver.

    You do include debug flags, right?

    No, it's a disaster when used by anyone. Sorry.

    BULL. You're just making MORE work for yourself.

    Well, we can settle this right quick. URL? Let's see what type of steaming pile of code this is.

    Hell, it's probably WHY it needs fifty developers instead of two ;) - Ok, not entirely true - I could see two back-end coders, one layout coder (since most back-end coders can't write HTML worth ****), and one additional per language
     
    deathshadow, Sep 13, 2007 IP
  11. GWiz

    GWiz Peon

    Messages:
    359
    Likes Received:
    17
    Best Answers:
    0
    Trophy Points:
    0
    #11
    Holy shit? Why don't you two pull your head out your asses and realize what you're arguing about, some extra bytes and lost efficiency? Who the fuck cares?

    The days of 56k are gone, most people these days have broadband, and bandwidth is becoming cheaper every day, I don't think a few extra bytes has ever blown up a server.

    And what this about a poor code being a disaster? Well what's it to you? What do you care how someone else runs their business? If it's the way they run things, and have been doing so successfully for years, than good for them. If they don't know any better, and could improve their efficiency, don't you think that's their problem, not yours?

    Stop fucking looking for an ego boost, you're like a bunch of kids trying to prove who's the the bigger e-penis. At the end of the day, nothing will get resolved and you aren't proving anything to anyone.
     
    GWiz, Sep 13, 2007 IP
  12. deathshadow

    deathshadow Acclaimed Member

    Messages:
    9,375
    Likes Received:
    1,853
    Best Answers:
    245
    Trophy Points:
    515
    #12
    Wow, strike a nerve?

    Sorry, but no...
    http://www.websiteoptimization.com/bw/0708/

    Still almost 48% of the US market and unlikely to actually change any time soon. MOST of the smaller towns within 20 miles of my current location STILL don't even have dialup ISP's that are local calls, much less broadband... and I'm in New England. Get into the american midwest and there's not one ISP in the world that's going to spend half a million dollars to run 50 miles of cable to pick up 100 customers. NOT GOING TO HAPPEN. For close to 40% of the contintental US the option of broadband is still a decade off or more.

    Or take a look at the minor suburbs of major cities like Boston or Chicago where broadband is STILL >$50/mo... There's a reason those people are still on dialup.

    This arrogant attitude 'everyone has dialup' is utter and total crap - and frankly being a bigger ass than either of us were being.

    We can't all live in South Korea or Monaco (and even in those places you've still got 12-15% still on dialup)
     
    deathshadow, Sep 14, 2007 IP
  13. GWiz

    GWiz Peon

    Messages:
    359
    Likes Received:
    17
    Best Answers:
    0
    Trophy Points:
    0
    #13
    Ofcourse, 56k users still exist, but what % of those users are going to be visiting your website anyway? Depending on your demographic, you may likely end up getting less than 1% of 56k users due to your target community being on broadband.

    If you're targeting gamers, most gamers aren't going to be caught on 56k. So of course you shouldn't be concerned about saving them 3 extra seconds of load time for those poor 56k users.

    I won't deny that yes it's good to have good code, but its not an absolute necessity, and at the end of the day, no one is going to die if code messy.
     
    GWiz, Sep 14, 2007 IP
  14. Stomme poes

    Stomme poes Peon

    Messages:
    3,195
    Likes Received:
    136
    Best Answers:
    0
    Trophy Points:
    0
    #14
    Me too, but everyone can't just download IE7 (though they CAN just download FF and O). I know you have to have XP with SP2, but I've also heard of problems with people getting IE7 on top of that.

    I've got in my footers "Upgrade to..." with the big three. Too bad Opera doesn't have a download page in Dutch. I don't think many of our customers will use O if they can't read the directions on how to get their language files : (

    Fafles, good luck with your business, but your site needs more than a bit of tweaking. Unless someone like deathshadow, Dan Schultz or VimF decide to recode your site for free, you really need to hire someone to do it. If s/he looks at the code and say "Holy Crap!" then they likely can do what you want.
    Or, you could stick around here and learn if you insist on doing it yourself. You learn a lot here, and at SitePoint and a few other places.
     
    Stomme poes, Sep 14, 2007 IP
  15. qube99

    qube99 Peon

    Messages:
    75
    Likes Received:
    1
    Best Answers:
    0
    Trophy Points:
    0
    #15
    The point I'm trying to make is the use of 2 stylesheets for solving IE6 parsing bugs is very easy, fast and efficient. It also avoids the use of hacks such as the star selector and the Holly hack. The code is totally valid, allows Strict doctypes, and never has to be revised due to a browser release. It has the added benefit that someone with only a basic knowledge of HTML/CSS can get it all to work very quickly. There are no hacks whatsoever used in this technique.

    The .htc hack is an interesting one since it only executes the jscript in IE6. Many HTML coders probably don't want to learn Microsoft jscript to write a web page, and many webmasters would not be able to troubleshoot it. The 2 stylesheet technique using only HTML/CSS would be a far easier, simpler and faster method for those who do not know jscript.
     
    qube99, Sep 14, 2007 IP
  16. deathshadow

    deathshadow Acclaimed Member

    Messages:
    9,375
    Likes Received:
    1,853
    Best Answers:
    245
    Trophy Points:
    515
    #16
    Unless it's not available where they live. Are you going to tell 90%+ of the state of South Dakota they don't count?

    In a LOT of places it's not a matter of want, or 'gamers' or 'nobody who cares about computers' - it's a matter of availability... and flat out in a LOT of areas across the US it is either priced out of peoples reach or flat out just not available - and not looking to be available any time soon.

    ... and being you can play UT 04 just FINE on dialup, I really don't see the 'gamer' thing. Sorry, but no.
     
    deathshadow, Sep 19, 2007 IP
  17. Dan Schulz

    Dan Schulz Peon

    Messages:
    6,032
    Likes Received:
    434
    Best Answers:
    0
    Trophy Points:
    0
    #17
    Hi Ana (I'm the Dan Schulz that Stomme referred to, in case you're wondering). Sorry to hear about your developerr's sabattical, it can happen to the best of us when we least expect it.

    Unfortunately, I do have to agree with the others here, this code is rotten to the core, and needs to be scrapped wholesale. Fortunately the front-end (HTML/CSS/JavaScript) for the most part is pretty easy, since you have a two column layout. I'm not particularly fond of the fixed width since it really goes against my coding style, but whatever floats your boat as they say.

    It looks to me like the entire site was built around the shopping cart system, which is a tactic I try to avoid at all costs, since the shopping cart system should be an enhancement to the site, not the Web site itself. If there's any part of the cart system that needs to be "global" so to speak, it should be the code that allows customers to add items to the shopping cart, and that should only be on the product/checkout/confirmation pages (and in your case the home page).

    Bear in mind though that eCommerce is not my area of expertise, as I focus almost exclusively on content sites, though I have worked on Web applications in the past (in one case I was able to reduce the code needed by a whopping 75%, while enhancing the functionality of the site and making it easier for people to use, regardless of browser or abilitly).

    If you could kindly excuse me for a moment, there's a lot of misconceptions and downright dangerous lies in this thread that I have to dispel so the record can be set straight.

    You shouldn't need to do this at all in the first place. Not only are you causing the server to request another file when IE 6 is used, but nine times out of ten the style rules in the stylesheet wouldn't be needed in the first place if the Web site was coded differently.

    Unlike some people, I actually consider Internet Explorer 6 to be a (mostly) compliant browser. Sure, it's lacking a few things like PNG alpha-transparency support, min/max-width/height, :hover on elements other than anchors and fixed positioning, but these can be overcome, and served via the * html (Star HTML) hack, which uses IE 5/6's "root element" bug to serve those four browsers (IE 5/Mac, IE 5/Win, IE 5.5, IE 6) alternate code. Yes, * html does affect Mac-IE, which is why the backslash comment hack was developed for those situaitons when Win-IE needed code that Mac-IE didn't need (due to it being more compliant than its Windows brethren at the time). Thankfully though, the days of Mac-IE are gone as Netscape 4, and it can safely be relegated to the dead browser pile.

    It may be foolproof, but as I said before it's unnecessary. And for the record, the * html hack was not broken in IE 7; the IE team at Microsoft happened to fix the "root element" (which came before the HTML element itself) bug, which caused developers to scramble to "fix" their code for IE 7 (these same people will be in for a rude awakening when IE 8 comes out in a couple years - trust me on this).

    Actually Stomme, it's really better to keep an eye on IE 6 at the same time as IE 7, Firefox, Opera, Konqueror and Safari. Most people code their sites using the latest version of Firefox, then check in Safari, completely ignoring Opera and Konqueror, then check in IE 7. Then they'll check IE 6 as an afterthought, and start diving in the dumpster for hacks they wouldn't have needed in the first place had they been paying attention to IE 6/7, Firefox, Opera, Konqueror and Safari as they were going along, but I digress. ;)

    Oh yeah, most definately. Accessibility, usability, cross-browser compatibility and on-site SEO as well as the proper application of clean (and this does include hack free as much as possible), minimal, semantic and valid markup are essential these days, as are the complete separation of structure (HTML), appearance (CSS) and improved interaction/functionality (JavaScript). If a developer isn't aware of these things, they need to learn. The learning curve may be a bit steep depending on how deeply the "old ways" are entrenched, but as they say, you can't make an omelette without breaking a few eggs.

    Simply put, if a person cannot use, much less even access, a Web page, then the developer has failed to do his or her job. Whether out of lack of awareness or ignorance does not matter. Fortunately though, the former is far easier to correct than the latter.

    That would be about it as well, though as I said earlier, I'd just use the * html hack and be done with it (or the * html first-child selector hack for IE 7) so the job can get done today so I can worry about fixing it on a local environment tomorrow while the client's site starts doing what it's supposed to do out in the wild blue yonder.

    This is actually far from the fastest, most efficient method, and is wrought with more problems than a person can shake a fist at. Especially when you consider that when it comes time to remove the conditional comment - how many pages have to be edited to remove it? Of course a person could use the over-glorified $400 text editor called Dreamweaver, but I haven't touched that program in three years now. As for includes? Might as well forget it, unless the CC is delivered via an include that contains other content, otherwise you have to edit the PHP templates as well, and let's face it, how many people duplicate their PHP templates with includes, rather than using a single file then calling the page content dynamically depending on the page that needs to be served?

    It's more like a hack really. Which I really detest.

    Easy, collapse the padding using the universal selector. Of course, if you're going to be dealing with forms a lot, then this wouldn't be such a good idea, but some CSS (such as a reset) can get around that.

    It's called a combination of unobtrusive JavaScript (or a .htc file embedded into the stylesheet, preferably imported into the main stylesheet and hidden from other browsers using the * html hack) and a proper fallback option for those who don't have scripting enabled in IE 5/6.

    Most of which are caused by "hasLayout" and are easy to get around. The only other real bugs that one has to worry about are the double-margin float bug (use display: inline; alongside your float without a hack) and the three pixel jog (which can be avoided by properly floating elements and clearing floats in the first place).

    More like using quirks mode to invoke the Trident (and yes, it's Trident, not Triton; the Mac-IE rendering engine was called Tasman) rendering engine's error-handling (which isn't great anyway). But yeah, other than the "it breaks all other Web browsers out there" comment I do agree with you. Stick to standards mode at all times.

    Why do I disagree with that "breaks all other browsers" statement? Two reasons. One, most of the other browsers will default to their "error-handling" as well (Opera has its own way of doing things - it's way or the highway). As for the second reason, well I already touched upon that. Opera doesn't have a quirks mode. Only standards mode.

    Actually, it can. HTML is a structural markup language. Use the elements as they were intended to be used (and this does include a complete and proper DOCTYPE) and Internet Explorer WILL render it properly. The misconception that it doesn't only enters the equation when you toss CSS into the mix.

    Don't count on IE 6 being gone any time soon. Look how long we had to put up with IE 5.01 and Mac-IE after IE 6 came out and Microsoft stopped supporting Mac-IE. I think it's going to be another four to five years (an eternity in Internet time) before we can finally write off IE 6 (since you have to use Windows [edit: 98SE], 2000 or XP to use IE 6, and you need a legal and verified copy of XP with Service Pack 2 or Windows Vista to legally use IE 7, despite a standalone IE 7 being available at tredosoft.com).

    Actually the bug is IE 5/6's insistence that there's an element called "root" above the HTML element that makes the * html hack possible. This was corrected in IE 7, which makes the hack safe to use in the wild (and thus negating the need for a conditional comment targeting IE 6).

    Huh? The * html hack works in IE 5.01 Windows, IE 5.2 Mac, IE 5.5 and IE 6 BECAUSE of the foolish insistence that there's a "root" element above HTML in the DOM - and yes, CSS does interact with the DOM, since HTML elements are objects that can be manipulated by styles - heck why do you think there's an entire subset of JavaScript that lets you write styles in JavaScript (such as document.style for example) in the first place?

    Complete rubbish on so many levels. For starters, <link rel="stylesheet" type="text/css" href="ie6.css" media="screen" /> IS HTML (well, actually XHTML in this case), it's just being sent ONLY to IE 6 due to the conditional comment, so yes, there is an extral call to the server. That call is being made to request that the server send a stylesheet that has no business existing in the first place.

    Second, and this is more important than the first, if you're using more than one stylesheet in a document, you're doing something wrong. The only exception to this is when importing one stylesheet into another, and even this should be done sparingly, if at all.

    The * html selector works perfectly well, and it validates as well, so your non-compliant arguement just got shot down like an old Soviet MiG fighter in a dogfight. And (in this particular case) who said anything about telling people to get a better browser? if the coder had done his job properly, the Web site would work properly regardless of what browser, assistive device or other user agent was used in the first place.

    No, he doesn't comment his own code. With over 20 years of programming experience, and being that he's responsible for the sites he builds and maintains, he knows fully well what he's doing. When he writes code that others have to maintain, then, and ONLY THEN will he document it. (And yes, I can say this, I've known him for nearly five years now).

    And besides, I don't think your shop would be able to afford him anyway. ;)

    At which point the code should be tossed out and re-written from scratch. If that means using object-oriented instead of procedural code, then so be it. As for the front end, where the stylesheet (yes, stylesheet, as in the old saying "There can be only one." from the movie "Highlander") comes into play, if the HTML that is produced by the back-end isn't up to snuff, then re-write it so it is. It may cost more to re-write, but any real businessman will be able to recoup that "loss" through increased sales of a much higher quality product, especially if it's a Web application that can be re-sold (or licensed).

    Which makes it a failure since you might as well be using spaghetti code instead of a properly written rock-solid application.

    Which doesn't mean squat. Bad code is bad code, no matter how many beers you've had or how many bottles of Chanel #5 you've poured all over it. Another way of looking at it is saying that a snake that changes its name to a lilly is no less a snake.
     
    Dan Schulz, Sep 19, 2007 IP
  18. Dan Schulz

    Dan Schulz Peon

    Messages:
    6,032
    Likes Received:
    434
    Best Answers:
    0
    Trophy Points:
    0
    #18
    Sorry about that, looks like my mega-post ran afoul of the 20,000 character limit. Oh well, as they say, two [in this case, posts] are better than one.

    Watch your mouth. There's a (ok, at least one) lady on this thread.

    Most? Where do you get your information from? Deathshadow summed it up perfectly (by the way, I'm one of those dialup users, and I live in the Chicago area) so I won't elaborate further.

    Anyone worth his or her own salt building Web sites or programming applications for a living has to know how to code properly. No ifs, ands, or buts about it. It doesn't matter if you work solo or for a company. ESPECIALLY if you work for a company, since you're likely to be tested on your knowledge of industry best practices anyway.

    Ahem, what did I say about proper language? Your obvious lack of respect for the lady who started this thread speaks volumes. Unfortunately it's drowning out anything you're actually trying to say.

    He's responsible for the official Web site of a table-top miniature warfare gaming company that's responsible for an intellectual property that's existed for over twenty years with a world-wide audience. Trust me, he gets a LOT of dialup users. What do you do?

    That's technically not entirely true, Stomme.

    http://tredosoft.com/IE7_standalone

    I wouldn't even do that. I'd just make the Web site degrade gracefully in IE 6 if I couldn't get it to behave. But that's just my own personal decision. As for Opera not having a download page in Dutch, I know somebody who works at Opera Software. I'll see if he can ask around about that for you.


    Unfortunately I'm kinda booked solid at the moment, but perhaps Jason (deathshadow) might be able help out instead? (And frankly, I'd probably ditch that ecommerce package and go with something far more reliable and secure as well - the lack of quality in the HTML and CSS really makes me nervous about the possible SQL injection attacks and other attack vectors - real serious security issues - that may be lurking underneath the hood).

    It may be fast and appear to be efficient, but it's really not since you're paying for a hamburger on Tuesday that you bought today. What's going to happen when Tuesday comes along and you don't have the cash?

    With the Star Selector hack, when it comes time to remove it, all that needs to be done is edit one single file and *voila* - no more hack. Can you say the same about the proprietary and invalid HTML that's hidden in plain sight via ordinary HTML comments on every single page of the Web site?

    Actually, it also affects IE 5 and 5.5 as well. And there's no need to learn JScript either - just drop it in. You will have to set the proper MIME type for .hta and .htc files on your server if it hasn't been done so already, but other than that (which is a real trivial matter anyway) there's no fuss to worry about, plus for those that have scripting enabled, the .htc file will also trigger :active and :focus as well as :hover - improving the usability of the Web tremendously for IE 5/6 users, especially those who use the keyboard to navigate through a Web page/site instead of a mouse.
     
    Dan Schulz, Sep 19, 2007 IP
  19. deathshadow

    deathshadow Acclaimed Member

    Messages:
    9,375
    Likes Received:
    1,853
    Best Answers:
    245
    Trophy Points:
    515
    #19
    Though I don't like to point at that these days since I've not updated the code for the site in over two years, and there's more of 2001 about it than 2007 - but then if you can look at your own code from two years ago and NOT be disgusted, you probably lack the proper demeanor to be a coder.

    I don't know about a lot - but last time I checked the server logs for outbound speeds transfer rates below 128kbps was still pushing 40% or so.

    Which is a good deal of my point on the subject. It is NOT easier to maintain having multiple sheets 'per browser', it sure as shine isn't future proof since you'll just end up making ANOTHER one when the next browser comes along, and is a waste of development time to boot.
     
    deathshadow, Sep 19, 2007 IP
  20. Dan Schulz

    Dan Schulz Peon

    Messages:
    6,032
    Likes Received:
    434
    Best Answers:
    0
    Trophy Points:
    0
    #20
    And what's a waste for the developer (time) is a waste for the client (money) as well.

    I talked about this on SitePoint not too long ago with regard to vector vs raster graphics for logos. I ended up finishing that thread (so far anyway) by asking the print guy why he's coming to me (the developer) for the vector logo that he should have gotten from the designer that I got the rasterized version of said logo for the Web site from in the first place so he could use that logo on a billboard or magazine. *scratches head*

    But I digress.
     
    Dan Schulz, Sep 20, 2007 IP