34 Comments
- lunar, on 10/12/2007, -2/+24"#4) When your site works in internet explorer without you trying."
This one is my favorite. - cyberfelon, on 10/12/2007, -3/+12Testing your site in Internet Explorer AFTER you've finished coding it is ridiculous. A good designer would never leave that until last.
- saifatlast, on 10/12/2007, -2/+7What are you talking about, I do this all the time. Code it using normal, standard code for nice browsers like FF, then go back and add hacks in as necessary to fix IE.
- phildo, on 10/12/2007, -0/+5Man... all this time, I thought the warm fuzzy feelings I get while doing web design work were from the the vodka and Red Bull.
- appleswitch, on 10/12/2007, -1/+5Best Digg I've dugg so far :D I spend ~a week with some great CSS and when I gritted my teeth and launched IE... It worked! WFF ill remember for years
- Cglass, on 10/12/2007, -2/+6char, you should definately start with firefox first, since firefox actually respects the standards that most browsers (except ie) tend to follow, god I hate IE
- ManInTheBunker, on 10/12/2007, -0/+4It's not like you have to choose. Just learn the 2 or 3 really major ways the IE f-s up web standards and just avoid them.
1. Never use padding on an element with a declaired width
2. Always use the redundant "display: inline" on floated elements or suffer the "double margin" bug
3. Always remove white space inside ul's and td's
That really is the vast majority of it.
By the way, anyone notice that the "recent features" sections blows up with a text zoom or two? ironic... - cadavreexquis, on 10/12/2007, -0/+3I think it's a preference. Some prefer to debug at the end. I like previewing in IE periodically and fixing bugs and writing hacks as I go along.
- DaMacGamer, on 10/12/2007, -4/+7dugg it for the kitty saying "mmmm, firefox" on the page :)
- awilke, on 10/12/2007, -1/+4Developping for Firefox and hacking for IE is ALOT easier than developping for IE and hacking for Firefox in terms of coding.
- rishimaharaj, on 10/12/2007, -2/+4What about "IE displaying your properly-served XHTML page correctly"? Oh wait - that's impossible.
- sedgemonkey, on 10/12/2007, -0/+2I would assume you would want to test thoroughly in the browser that the majority of internet users use -- currently IE. Put all your silly biases aside and think about it.
- shadedream, on 10/12/2007, -1/+3"from my exprerience it's usually
1. spend half an hour making the thing work in IE
2. spend the rest of the week getting it to work on firefox and opera"
Took a look at your site, you're using a table based layout as opposed to XHTML/CSS based layouts (web standards). Thats why it works in IE so well for you, using DIVs and CSS gets to be a hassle in IE because of its broken CSS implementation (not to even touch on its proken PNG implementation...) - Yossarian78, on 10/12/2007, -0/+2"I’m going to assume that most web designer / developers are web-savvy people and for that reason use a more capable browser as their staple and for that reason probably create websites using that browser as their main testpad. "
Are all other web designers really such elitist snobs that they really work this way?
Do managers really let their developers work like this?
I'm all with mrfoos on this one. If you've got an actual business or product you are running, you should be developing in IE. IT IS THE MOST USED BROWSER! I don't get why so many web devs get on their soap boxes about Firefox and then proudly go around saying "I don't spend nearly as much time making my work perform well on the browser that 95% of your clients are going to use". That is asinine to say the least.
It is the same thing as if Blizzard wrote WoW for the Mac first, and then went back and ported it to windows. Ideally you can get versions out for both at nearly the same time, but you absolutley would NOT miss launching with the windows version just because your devs liked playing it more with their Macs.
Design in IE first. Get it working for the largest market. If you think IE sucks so badly, then you'll have gotten all the hard work out of the way. The firefox and safari versions will then fall in line much more quickly. Putting off the hardest part of development till the end is just bass-ackwards and means you aren't going to hit dealines very well. - nic777, on 10/12/2007, -0/+2" IE is the browser of choice for 95% of the visitors across all domains"
Its not a choice for these users.... - Squinky, on 10/12/2007, -0/+1This is true in some cases, but there are a few designer-coder hybrids out there. Hell, the only reason I got into coding was so I could add some interactivity to my imagery.
Of course then one thing lead to another and now I'm a coder by profession ... - sensibledriver, on 10/12/2007, -0/+1That list is more like "Things you had better be doing to keep your job."
Oh, and designers should never touch code...ever. - quesera, on 10/12/2007, -0/+1And that's exactly why management should not be making uninformed technical decisions for developers.
IE compatibility may be the most important acceptance criterion for a design. That makes sense.
But it makes zero technical sense for IE to be the primary development and testing platform.
It's simple: IE is buggy and weird. Test equipment (your browser) should not have a long list of intermittent quirks that are triggered by obscure and seemingly unrelated things. Experienced designers know how to avoid most of them, but they are still there.
Using Firefox as a development browser gets you two immediate advantages: first, it's as close to correct in rendering HTML and CSS as you can get. You may still have to add hacks to fix other broken browsers, but you know you're starting from a base that is correct. It's always easier to start from a known good state, and address the quirks individually for different browsers. (The smartest second step, by the way, is to test in Safari or Konqueror: a completely different rendering engine that is also very very close to correct...so if you find a discrepancy (which is largely unexpected except for differences in supported bits of CSS2), you have reason to suspect your markup.
The second reason to develop with Firefox (and the main reason to prefer Firefox over Safari/KHTML) is that you have access to a reasonably good javascript debugger. No sane javascript coder would develop in IE. Ever. Test, yes, QA, absolutely. But not develop. It would be a total waste of time.
Then, when things are working correctly, check them in IE. Check them in Opera. Fix what you have to. If you're new to this game, do it early and often so that you can catch the major stuff before you build too much dependency into your layout decisions. If you're confident in your abilities, it's completely reasonable to wait until the end and then spend a few hours adding the usual hacks to convince IE to pretend to be a modern web browser.
The opposing position: "IE is the most popular so it should be first!1!" is the same backward argument that says "the answer is correct, so the code must be good". It's often rationalized with fun little pithy non-statements like "I have a business to run. I just want it to work" or "we make money, not art", etc. Great. If your product is an internal thing, or if you make widgets goddamnit and you just need a damn site on the interwebs, then fine. Do it the marginal way. Lots of used car dealerships have crappy websites, and no one really cares. Because no one really cares about used car dealerships.
But if you actually want to do business on the web, your site will require care and feeding. You'll monitor your site and see what works and what doesn't. You'll make changes. You'll probably hire a smart person to handle all of this for you. This is fine. But that person, if they are smart enough, will promote standards based design because they know they will have to live with the results, with future compatibility, with iterative changes, and with everyday maintainability. I'm not talking about a one-shot "if only we had a website..." business. They can use FrontPage and ActiveX. It doesn't matter. But if you treat opening a website the way you would treat opening a new storefront, you'll do things the right way.
The first and smartest step on that path is writing standards-compliant code. It's akin to making sure the bricks and mortar buildout has all the necessary permits and inspections and licenses. It's not that hard, it's not that expensive, and it will haunt you for a long long time if you try to take shortcuts. - cadavreexquis, on 10/12/2007, -0/+1Excellent article! Sadly, number 4 is right - I've yet to develop a complex, standards-friendly site that didn't need some IE hack or work-around.
- Arramol, on 10/12/2007, -0/+1It's so true. I recently had to design a web page for a class project. I just threw it together quickly in Dreamweaver since it's not a computer class and wasn't a big part of the project. I didn't even think to preview it in IE, just did it in Firefox. I get to the class, and find that Firefox doesn't have the Flash plug-in, and I don't have enough permissions to install it, so I had to load it in IE. Imagine my shock when it loads as a mess, despite it just being a basic table layout. Luckily we ended up using another group member's laptop for the presentation and she had FF with Flash, and it worked fine.
- yongfook, on 10/12/2007, -1/+2not LAST, but I think it's safe to say that most web designers use a more standards compliant browser as their testing browser. In the same way that it makes no sense to test in IE last, it makes no sense to use IE as your testing browser either. So one way or another you're going to be making little adjustments for IE along the way or make them all at the end - which way you do it is up to you, they both achieve the same ends and both ways probably end up taking the same amount of time, but personally I find it a neater practice to do all the IE fixes at the end, rather than piggybacking it onto the development for "kinder" browsers.
- JonGretar, on 10/12/2007, -0/+1Nice Article...
You can actually use Writely with Safari though. They say it in the error message. The URL is just www.writely.com/BasePage.aspx?browserok=true as it is still in testing. - inactive, on 10/12/2007, -1/+2All you IE haters out there, get ready to thumb me down... but I gotta say it. I managed over 100 small business, public organizations and church domains here in the USA. None of these are tech sites. They are websites for regular people. IE is the browser of choice for 95% of the visitors across all domains. I'm a business man. I don't give a crap about supposed standards if they aren't helping me with 95% of my customer traffic. Saving IE for last? If I were playing the odds, I'd never even view the sites in Mozilla, Firefox, or Opera. Ok. Slay me.
- Thorandor, on 10/12/2007, -0/+0Hehe, very nice article! Its just great when you work on a site, it looks great, it feels great....and then you realise that it's not working on the client's browser (which is a Win95 machine with IE 4.x), ah the joy!
DiGG++ - inactive, on 10/12/2007, -4/+4The IE one couldn't be more true. Tons of true stuff that I didn't even realize before! Nice article
Jeff - http://jeffeh.com - inactive, on 10/12/2007, -5/+5from my exprerience it's usually
1. spend half an hour making the thing work in IE
2. spend the rest of the week getting it to work on firefox and opera
maybe i should start with firefox first.... - DrColossos, on 10/12/2007, -0/+0"Internet Explorer is the family member you don’t like but have to deal with"
ha ha that is my favorite quote here... very creative comparison - Cglass, on 10/12/2007, -2/+2I.E. should be banned, it's using wallhacks and aimbot.
- theaftercrash, on 10/12/2007, -1/+0i like to debug as i go along, so i can keep a steady eye on how things are progressing both in firefox & i.e. this way, i can make quick fixes before i go and do something else i would otherwise have to un-do later.
but this article was a good read, it's the little things that were mentioned (like bad forms) that aren't given much thought, but have a negative impact on a website. - Monoboy, on 10/12/2007, -1/+0I don't like to take the time to stop writing markup/CSS to look if it's working correctly in IE. Heck, half the time I wait until the end to preview it in any browser.
After a while it's easy to get used to how browsers work and I usually have a good idea how something is going to turnout before previewing it. Sure there might be an error here or there, but nothing that's hard to go back and change one line of markup/CSS. - thegsa, on 10/12/2007, -10/+7Great Article
Digg++ - Splitt3rxx, on 10/12/2007, -7/+2it won't load
- Alphabet, on 10/12/2007, -28/+7ugh, their website design is terrible. Kind of ironic ...
- verucasalt, on 10/12/2007, -34/+3How did this garbage get on the front page?


What is Digg?
Check out the new & improved