Slashdot Mirror
Internet Explorer 6 Will Not Die
caffeinejolt writes "Despite all the hype surrounding new browsers being released pushing the limits of what can be done on the Web, Firefox 3 has only this past month overtaken IE6. Furthermore, if you take the previous report and snap on the Corporate America filter, IE6 rules the roost and shows no signs of leaving anytime soon. Sorry web developers, for those of you who thought the ugly hacks would soon be over, it appears they will linger on for quite a bit — especially if you develop for business sites."
Sorry web developers, for those of you who thought the ugly hacks would soon be over, it appears they will linger on for quite a bit -- especially if you develop for business sites.
Yeah, IE6 is the herpes of the internet. It appears to be gone after heavy medication but if you look under the first layer of skin, there it is.
...
Oh, and I should point out another untimely mark of IE6: we've all made this hilariously fugly hacks to make crap work in IE6 at some point and those relics of the last millennium are still out there. Which means that browsers still have to support the old rendering ways of IE6. Yes, the doctype will tell the browser what standards to use but I'm betting that the support for rendering HTML 4 is just as annoying as having to patch up old struts 1.x applications and read through nested tables galore in the HTML.
And we all know that 90% of the work out there for developers is maintenance. What a painful irrepressible memory
My work here is dung.
The reason IE 6 won't die is intranet applications that were coded specifically for IE 6 that corporations haven't bothered to make cross-browser. IE 7 (and presumably IE 8) breaks a lot of those sites.
At my current job, we're not allowed to install IE 7 or 8, and don't have the administrator rights to do it. It sucks because as a web developer, I'd like nothing better than to see IE 6 die a quick death.
This space left intentionally blank.
IT departments have no budgets right now. Testing all the in-house apps with IE8 would cost money. Even telling people to press the "render in IE6 mode" button would be quite expensive in terms of calls. So they're just blocking the update.
A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
f. Some of us, for various reasons, are pretty much stuck with using IE6 for browsing /. and are faced with a pile of mis-rendered & incompatible pages
Slashdot doesn't render properly in Safari 4 or Firefox 3.5 beta4 either - the comment titles and scores aren't displayed anymore
Stop doing the hacks, and let IE6 render them ugly and broken, while compliant browsers will render them correctly.
Consider that many users will not realize it is their browser. They will simply decide your site is screwed up, and leave promptly. This is not a mistake to be eager to make in many scenarios.
Reply to That ||
And scare away 50% of potential consumers because of a "broken website" ?
The Wise adapts himself to the world. The Fool adapts the world to himself. Therefore, all progress depends on the Fool.
"I checked the site statistics for my site and IE6 went from 15% of the hits in April to 0% in May."
Well, duh, because no sod can see anything in IE6 - visit once and never come back again.
This is the sort of crap that Opera has thrown at it - email a complaint to MSN, the BBC, any large website about parts not working in Opera (although they all do now), and you only ever got "nobody uses Opera to visit us"... OF COURSE NOT! BECAUSE IT DOESN'T BLOODY WORK!
It's like saying "Since we started banning unhappy people, our store recorded that 100% of customers in the store were happy with us!"
doesn't slashdot have any bug reporting tools for us to use?
i doubt CmdrTaco is reading anything below +5 insightful ;)
There seems like a pretty clear free-market solution to this problem: developing sites that support IE6, with all the requisite hacks and workarounds, is harder. It takes longer, and should cost more. If developers just attach an appropriate premium to this extra work, businesses start having a financial incentive to stop demanding it.
"Well boss, I got a quote for that intranet app we need developed, and it turns out our IE6 requirement adds 35% to the total cost." "Hrm.. and what's your estimate of the cost of migrating?" "Migrating would cost us more than the 35% on this one project. But looking a year or two out, paying that kind of premium on all future development contracts, switching is way cheaper, and will probably reduce IT expenses for security issues to boot." "Right. Start working on that."
Think about this: if you have a legit copy of Windows XP you're HARASSED to upgrade to the latest version. If you have an illegal copy, you're either smart enough to ignore the harassment, or you constantly fail the required product validation before upgrading.
I think this proliferation of IE6 is because it was the last upgrade that didn't require validation. It lives on through piracy, which also promotes insecure computers that don't have the latest updates.
The answer remains: fuck ye!
Administrator's response: Fuck executables outside %SystemRoot% and %ProgramFiles%.
The comment titles and scores are being rendered (highlight the page with ctrl-a), the problem is with the CSS - the background image that runs the length of the div element containing the title is being overwritten. This:
get's overwritten by this (appears further down the document):
You'll notice the issue doesn't occur on some of the alternative stylesheets (Ask Slashdot, YRO, etc). In the meantime, you can hit 'change' in the threshold form to set things straight.
If I say I don't believe in you, will that make you disappear?
I have one application sitting here right in front of me that is comprised of over 5618 files (about half of which are ASP or HTML) that were orginally built around IE5. When IE6 came out they broke. When IE7 came out, they broke. IE8 won't even render half the site.
The people who were commissioned to build it were done and gone years before I started working here. I have no documentation, the code is laced with inline SQL, .HTCs, and, in some places, 7 or 8 layers of includes. The database is undocumented, I'm the only person in the company who understands any of it.
COULD it be fixed? Yes. But it would take months for me to do it, and it would cost too much to hire someone else. Scrapping it and rebuilding it is the only viable option, but management spent a ton of money on this app and nobody will admit that it's a disaster and a $1 million+ mistake.
Whether you admit it or not, a lot of early web code out there was written by a lot of people who never had any business being anywhere near the profession. It's not going away any time soon.
Not to mention that it's 2009, but Slashdot can't even be bothered to work with Unicode yet.
Here is an em-dash: â"
Here is some Japanese: æ--¥æoeèzãã
See?
He who lights his taper at mine, receives light without darkening me.
I suggest you find a new job. That is a time bomb. Any management who won't admit that in 5 years a important part of the business logic is not going to work. Microsoft is going to stop supplying security patches for IE6. It's a fact, at that point you are going to have to run a very insecure browser while you do what you are saying is too expensive to do. Only now you have even more risk then starting the project before it's an emergency.
What happens when new hardware simply will not run XP and you have no choice?
My company just went though this. Luckily they listened to me and were proactive. We had tons of PHP4 code, a lot of it incompatible with php5. I pointed out plans from several projects we use to drop PHP4 support and the fact PHP itself was getting ready to drop support.
So we got approval to start the project. It took us 2 years of modest work in addition to our normal projects. We also made sure all new projects were fine with PHP5. While we were at it, we rewrote everything to conform to a standard that worked in all major browsers at the time IE6, firefox, and safari.
We also came up with a unified plan for the future. Doing things like putting an end to little access databases and random mysql servers. Unifying that took even more work as we had to reverse engineer work from developers long gone.
Now we have a very flexible framework to work in that allows us to quickly change directions as trends change in our field. Boss wants a site to work on his blackberry, no problem. He suddenly switches to an iPhone, again no problem. He goes bonkers and moves to linux, guess what, no problem.