Hacker News new | ask | show | jobs
by webdragon 5862 days ago
Microsoft obviously knows that the only way to kill IE6 is to say it is unsupported software and to stop releasing security updates for it. Until that happens, IT departments will continue to cling to it as part of the default install of Windows XP.
3 comments

Except, that will never happen before the support lifecycle expires. It would be devastating to their enterprise sales. One reason you buy Microsoft at a large company is because they promise support for 10 years. If they renege on that with IE6, it would get too much exposure and enterprise IT departments would revolt and MS would start losing ground in one of their last strongholds.
Exactly. Microsoft's enterprise business is built on backwards compatibility, and they've already had a few breakages in recent years, with (most things in) Vista, and the recent announcement that Windows Mobile apps won't run in Windows Phone 7.
Well, isn't that ten year support timeline due for expiration in the next year or so? I'm going from memory here so please correct me if not.
That would be for the initial release, but not for the service packs unfortunately. IE6 was still standard (and had some security enhancements) in XP SP2.
If Microsoft were serious about this they'd have released a ChromeFrame-style compatibility mode when IE8 came out which would allow admins to use group policy[1] to load the old rendering engine for poorly supported (i.e. all) enterprise webapps and use the modern engine everywhere else.

This is the same approach which will be necessary to address endemic security and reliability failures caused by bad architectural choices made for single-user desktops in the 90s. Since they finally shipped MED-V I suspect they'll probably use that approach and tell corporate customers that the only way to continue running IE6 is in an XP VM with some sort of on-demand launch scheme.

1. Setting an HTTP header would be easier but seems likely to allow phishers to opt-in to security holes and there's an argument that pushing the support cost off to app owners is better any way.

The problem is also compounded by the fact that a lot of companies (especially insurance and finance) have corporate intranets that are designed to run on IE6 exclusively. For these people not supporting IE6 is not an option, until the intranets themselves get rewritten.