views:

356

answers:

15

Why big companies and even US government still have Internet Explorer 6 as their recommended browser? I'm working at Cisco Systems and their recommended browser is IE6 which makes my life, as a web developer, miserable. I have to spend three times more time debugging problems for IE6 than for any other browsers and I think that they (my boss and few people for whom I'm developing applications) are thinking that I'm writing my code/css not intelligently. What can I do to persuade them to switch to FF or at least to IE7.

+5  A: 

You need to get to work showing them why IE6 is a poor browser. Give them nice examples of "this is web-standards code in current browsers, this is web-standards code in IE6...see how crappy you are for requiring IE6?"

You should probably put it a little kinder than that.

Pselus
Two problems: the assumption that they don't know this already and that you can convince them when their own internal people can't.
annakata
+1  A: 

Aren't there huge security improvents in IE7? This should appeal to big / government companies.

Edo
+2  A: 

IE6 already did his job 8 years ago, however because Windows Vista didn't spread as it should have done, people are stil sticking with IE6, if Windows Vista was the main stream it would have been a peaceful transition.

I wish Windows 7 will push these enterprises to IE8 which is a better browser at least better than IE6

bashmohandes
but still... IE8 is far far away from the concurrency... And as a developer I really don't know why!
balexandre
That's why I said at least better than IE6 :)Personally I use IE8 as main browser and I think it is enough, but my Dev browser is for sure FF
bashmohandes
A: 

1 Simple answer: People don't want change. And that's probably why they still use that old ancient IE 6 browser. link text.

Roland
+6  A: 

Large companies are like cargo ships. They know where the icebergs are, they can clearly see them: it just takes them a while to actually turn.

These kind of organisations have massive amounts of smart people who's job it is to be interested in browser security (and other such issues), it's not that they aren't aware of the problem, it's pure logistics and bureaucracy. It takes testing, approval, and a considerable effort to actually roll out. Consider that a common platform is much easier to manage for centralised IT support, and consider that most such companies will handle roll outs for new software when they roll out new hardware, which is expensive and done on a alrge scale where they can maximise bulk buy deals.

And be thankful it's just IE6. I had to do work for NN4 for a major bank not so very long ago.

annakata
A: 

Many companies still use IE6 because:

  1. If it ain't broke, don't fix it. (Sure, we know many ways that IE6 is broken, but the average user doesn't know or care about web standards and knows nothing about security.)
  2. Their existing web apps were built for it. In many cases, these apps break in IE7 or FF.

Arguing with about web standards is unlikely to sway anyone, even if it involves your time efficiency. Your best bet is to push security.

Geerad
+2  A: 

A lot of large companies have custom intranet apps that were coded to work only in IE6, and depend on some weird activeX control, with the reasoning "It's a controlled environemnt, I know the browser is IE6, I shouldn't need to bother testing on any other browser" (an attitude I've seen displayed by some people on stack overflow, which should seem obviously wrong, and yet I was unable to argue convincingly against it)

It's called vendor lock in. The cost of staying with IE6, and continuing to develop on it, is less than the cost of switching, and then updating/porting the company's HR programs to work in IE7.

Breton
This is exactly the problem at my current company. We have applications integrated with not only IE6, but a *specific version* of IE6. I work on the intranet and I can't tally the hours I've spent making things look "right" in IE and yet still be up to par with standards. I can't even write robust CSS because virtually every rule uses the descendant selector. I just feel to limited...
Zack Mulgrew
I actually feel that the above attitude is morally wrong, because it means inflicting that kind of pain on future developers.
Breton
+3  A: 

Just to note a few arguments that do not work. With stupid management answers that I have heard in the past.

  1. If you stop requiring support for IE6 developers you will save tons of money on internal software development. Ok, so we save how much? 200k/year? And how much does the upgrade cost? So you say that our helpdesk will have to install new browser on every single computer? That is thousands and thousands of computers and will certainly cost more than 200k a year.

  2. IE6 is an old browser that does not scale to today requirements. Son, does Google work on IE6? Oh, if they can run their awesome mind blowing technology on IE6, why can't you? Are you any worse?

  3. IE7/8 or Firefox are much safer than IE6. Well, we haven't had a breakage yet, have we? That means that we're safe. And why are we safe? Oh, because we haven't upgraded to something that is not tested.

To me IE6 is like mainframes and cobol. You could dream of replacing them, but it just doesn't work all that easy.

Marcin
+4  A: 

I'm going to be brutally honest, and will likely get voted down for this, but:

If you really are spending 3x more time debugging in IE6 than in other browsers, your boss is right: you are not writing your code intelligently.

It's probably not your fault. IE6 has its quirks, and it takes time and experience to learn those quirks. However, you can preemptively solve many IE6 issues -- and sidestep many of the remaining issues during the design phase of the project.

For example:

  • In CSS, always couple a float with a display: inline.

  • Use clearfix for float clearing.

  • Learn to use zoom:1, and how and where hasLayout applies.

  • Don't mix margin and padding on the same element.

  • In JS, either learn the ins and outs of browser differences, or use a library like jQuery that normalizes those differences for you.

There are lots more. You'll learn them in time. But realize that since it's unrealistic to convince many big companies to move from IE6 in the near future, you have to adapt your coding practices to save yourself -- and your clients and company -- time and money.

thedz
A: 

I think part of the problem is related to Vista. Everybody knows that Vista is heavy and don't aggregate many changes that can make companies to change. So, most of the companies prefer to keep XP on their machines with its IE6 installed. The other reason I think is the difference between the HTML implementation on IE6 and IE7. Companies have their intranet (and internal systems) made to run on IE6 and migrate to IE7 can bring a lot of issues to solve... As Geerad said: "If it ain't broke, don't fix it".

A: 

I just went though this exact same thing with a global brand. Initially I agreed with you, and wrote a whole bunch of angry emails to try and state my case, at the end of the day IE6 was clearly stated in the spec and I had no choice but to get it done properly to the clients requirements.

The biggest issue I had with IE6 was not layout, that can be fixed relativelky easily, but with the lack of support for transparent PNG's. Even that had a clever JS solution in the end and now the page looks identical in all browsers. My devs have all learned their lesson the hard way and I dounbt we will be sitting in this position again. What I try and do now is get the CSS template pages done first and tested across all platforms and browsers using a service like browsercam before handing the project over to the coders.

When the project went live it was discovered that there were issues with the new IE8 too. Since IE8 wasn't in spec because it didn't exist at the time there was talk about it not being our problem, but with the amount of money these guys pay us combined with the moral and possibly legal obligation to future proof our work we fixed that too.

Here's my take on an old business principle: since I appreciate the contracts and the money they generate, the clients are always right, even if they are stubborn and ignorant and sometimes wrong :)

Gary Benade
A: 

Internet Explorer 6 is the newest IE available for Windows 2000, an operating system that may be dated but is still widely used in corporate environments, sadly. Companies move very slowly, and Windows 2000 is "unfortunately" still good enough for many things companies usually do, which is Office work, Communication and usage of some internal Applications.

Also, there are some third-party programs that require IE6 due to some ActiveX controls that do not work with IE7 and newer.

Michael Stum
+1  A: 

The problem with large companies is that they have dozens and dozens of little internal applications that were written years and years ago and nobody maintains or updates them. While switching to a modern browser would save time and headache, it's a very large investment for the company to test and fix all of their existing applications.

KevMo
+1  A: 

There's nearly the same question on server fault. The short answer is, the very thing you desire in IE7/IE8 (standards compliance) is what prevents the adoption - it breaks all those enterprise web apps to have IE not behave like IE (like KevMo said).

Tony Lee
What up with the down vote? Understanding why we're all stuck supporting IE6 is important. True, I intended to be provocative and thus begged for a down vote but, I'd like some recourse to commence. Explaining to your customer that IE 6 sucks isn't useful - I gotta say, at least IE6 isn't IE5.
Tony Lee
+1  A: 

Having been fighting that fight in my company for more than 2 years now, I have learned a lot. For us, the problem goes like this:

IT Governance:

We have 2500 web-based line-of-business applications worldwide. Poll each group and ask them how much time it would take to make sure all their apps work in IE7/8.

Each group:

Some of our portfolio will probably work without modification, but lets play it safe, assume the worst and say it will take 4 man-weeks per application in case we hit some snags.

IT Governance:

It will cost 60 million dollars to upgrade the damn browser? Hell no.

Rex M