Realism? On Hacker News? I thought we had all agreed to act as if IT consists entirely of web start-ups with no such thing as legacy, even though that's probably 99% of the pie.
Why can't Microsoft simply take the IE6 codebase and refactor it into a static, standalone "Legacy Intranet Browser" that, similar to Mozilla Prism, will generate a locked-in "site launcher" for each site it must be used on?
However, both of those solutions ignore the branding aspect. You don't want to make your company use "Internet Explorer 6 for Application Compatibility", because that sounds like an acceptable, normal choice to make. You instead want them to use "WARNING BIOHAZARD ISOLATION AIRLOCK BROWSER FOR KEEPING SUCK AWAY FROM COMPUTERS," or something equally poignant. Using it, or even looking at it, should convey the "stay away" signal just as strongly as spots on frogs or stripes on snakes.
The problem is that you can't run IE6 and IE8 in parallel (though it sounds like adbachman may have a way around this). So you keep IE6 for applications in your intranet that require it and use Firefox for general web access.
But the point is that that's possible now, but corporations aren't doing it. As long as IE6 exists on the computer as a general web-browsing solution, and is required for something, they'll be loath to install and maintain "another, redundant" web browser for anything else. To stop this practice, we have to take IE6-the-browser away completely, and in its place give IE6-the-hardwired-intranet-client.
In most of the companies I've worked for in the last decade, including a fair bit in systems administration I've see the following used as the solution.
USER: I have a problem.
TECH: Let's see...hey! Why do you have Firefox installed?
USER: It works better than IE6. Why are you guys using old crap like that? I thought we were a modern company...
TECH: We are modern, except for IE6.
USER: So I have to uninstall it?
TECH: I can't "officially" support it.
USER: Aaaah.
TECH: OK, your print queue is cleared.
Where I work the conversation would go something like this: TECH: Who gave you permission to use uncertified software? USER: Uh, noone TECH: I'll have to make a report about this. MANAGER: You're fired.
That would be the user wearing the boot up his ass. Running unapproved software (e.g. portable FireFox on my locked down machine) would be grounds for immediate dismissal in this org. IE 6 forever (sigh)
Personally i don't like Prism. The idea of taking an app not specifically designed for desktop use and pretending it is a desktop app seems like a waste of time or temporary fix.
I thought that too, but I've been using it for unfuddle and I prefer it this way. It gives me a separate window and dock icon from my regular browser, and just works well with the self contained system that is Unfuddle.
Yea I can see if it has utility right now but it's really a temporary fix to the problem. I think it detracts from the better goal of making rich web apps that run without a browser and the gradual blur between desktop and web apps.
Or they could have an IE6 compatibility mode built into IE8. Then all those companies that designed IE6 apps instead of web apps in the past could specify with a code change that the pages are not www pages but IE6 pages and MSIE could display them appropriately (and addons for FF could use that to hook in an IE6 view, etc.).
The difference between IE6 and IE7 is like the difference between a grain of wheat and risen dough, neither is a loaf of bread [standards compliant browser] but the later is not far off.
The best bet for a company that has sites that require IE6 right now might be to keep IE6 and install Firefox with IETab set to display the problematic sites through IE.
There are unfortunately ALOT of corporate programs that will only run on IE6. At work, our main app is optimized for IE6 but we are trying to ween our customers off of it and onto IE7... but all IE7 bugs are P2s at worst because IE6 is what our customers are using.
At the day job, legacy installations mean IE6 compatibility is a requirement and all other compatibility is a luxury. I prioritize accordingly.
Its not like I can tell my customer "Look, I know telling you to upgrade your browser is going to break that 3 million dollar CRM system you've been using for the last 8 years... but you'll get transparent PNGs!"
How does it use IE6, why was it not designed for the web in the first place? What other single point of failures were chosen in the design phase?
That feature matched CRM, or probably a better one, may be available as OSS now for nothing but the cost of a few developers to create some shims and massage things into place?
At my previous company. A well know global airlines we just put out a tender for cargo management and loading systems. All the vendors were ie6 compliant but did not certify for ie7 or any other browser. I tested many of the demo applications on ie7 and on firefox and they totally rendered incorrectly.
Even at this time corporations are not upgrading and i guess they wont as long as vendors continue to provide ie6 compliant applications without support for any others. Goes to show how slow corporations move these days, even for something as simple as a browser (in comparision to other upgrades they make).
Absolutely. Now if you're a small vendor, you have no choice but to downgrade all your computers to IE6 in order to enable your employees to access business critical information. Because of how complex the configuration is, running IE6 in parallel with IE7/8 with hacks will not work. And there is no other way to access this data. So now even though other browsers might be installed, most users will use IE6 for regular browsing. I want IE6 to die but I'm afraid it'll be a while.
I run IE6 (an a virtualbox VM with WinXP) alongside IE8 and FF, Saf, Op, Chrome, etc., for testing purposes. It even runs IE5, IE4 but I don't have to bother with them except when I want a laugh.
If it's "coded to Firefox" then they're doing it wrong. If it's coded to web standards and only FF is supported, that's a different matter.
Suppose a patent problem came up (as with GIFs in the past) and FF couldn't be used legally any more. If you code to FF you'll probably find you're OK, but if you code to standards then any standards compliant browser should be a drop-in replacement.
It's coded so that we only have to check if it works in Firefox. It's not like we're explicitly trying to leverage Firefox-only markup or CSS. I believe the idea is to reduce waste of development time and effort. I think I'm going to call YAGNI at this point re: future patent problems. I don't see that happening soon, and I'd gladly take the time savings now, and cross that bridge when we get to it -- if it comes at all.
I work for a major law firm (think top 6 worldwide) and we run a DM system that requires IE6 only (ie, doesn't work in IE7 even) so every office worldwide is tied into IE6.
Only recently have we started rewriting it to work in newer browsers ... and I don't expect a working prototype before middle of next year!
The docs and metadata are in a DB, just have a new frontend written to match the old UI but with knobs on. Should be done and ready to test by next weekend (inside joke ;0).
Or just get an OSS DM system and munge it together with the old DB. Perhaps knowledgetree would be a starting point for that.
I can understand there are problems - layout issues, popup windows blocked, activex controls denied permissions. However all these seem like things that are either totally cosmetic or could be fixed with relatively minor hacks - most of them with just registry configurations or an alternative stylesheet here and there. If people could name actual technical reasons why apps break then I'm sure a group of interested parties could quickly form an open source project to develop a compatilibity layer / plugin / something that would bridge the IE7 - IE6 gap. But nobody ever seems to get down to that level of specificity. I have to think that it's really mostly about corporate inertia and lack of desire rather than actual real technical challenges.
They're not "taking advantage" of IE6 any more than the financial industry and the Social Security Administration are taking advantage of COBOL. It's simply that the software exists and has been coded and tested against IE6. For a company as large as Sprint (and many other like them), the cost to refactor or rewrite huge swaths of their codebase are extremely prohibitive.
The benefit of the upgrade does not outweigh the detriment. This is why companies like Citrix exist. Virtualization is a good market to be in right now.
simply this: It's not possible to run IE7 or IE8 on windows 2000, which is still massively installed: many companies don't want to upgrade for a few reasons. (mostly: legacy software and cost).
Microsoft could solve this tomorrow in myriad ways- but it would appear that it's not on their radar.
A lot of things about IE6 are not standards compliant, so it's either programs for both or program for just IE6 and there is never enough time/resources to develop for both.
HTML worked fine. It's the documents which are not compliant, because they use non-HTML frills and were not correctly designed to degrade gracefully without them.
