IE6 Was Unsafe 284 Days In 2006

An anonymous reader sends us to the Washington Post's Security Fix blog, where Brian Krebs has toted up the total vulnerability days for IE6 users in 2006. From the article: "For a total 284 days in 2006 (or more than nine months out of the year), exploit code for known, unpatched critical flaws in pre-IE7 versions of the browser was publicly available on the Internet. Likewise, there were at least 98 days last year in which no software fixes from Microsoft were available to fix IE flaws that criminals were actively using to steal personal and financial data from users... In contrast, Internet Explorer's closest competitor in terms of market share — Mozilla's Firefox browser — experienced a single period lasting just nine days last year in which exploit code for a serious security hole was posted online before Mozilla shipped a patch to remedy the problem."

I hope stuff like this makes the paper

Then it might affect people who don't already know it.

Hazards of monoculture

Consider that this would be less of an issue if IE weren't used by 70-90% (depending on where you look) of web surfers. Most-used and least-secure is a disastrous combination. This is why alternatives are important [alternativ...liance.com]. If the space broke down at, say, 30% IE, 30% Gecko, 15% Safari, 15% Opera and 10% random, malware authors would have to go to a lot more effort to exploit the majority.

Nothing to see here...

you know the drill.

My bet is that the number that COUNTS is probably larger (also larger for FF), the number of days where there was a vulnerability that was known by malicious groups, just not publicly posted.

all a matter of perspective

IE6 Was Unsafe 284 Days In 2006

Of course the flip side of this story is that IE6 was safe for 81 days in 2006.

Obviously, the solution is to shorten the year to 81 days.

There are three main factors for this

1. IE != OpenSource - many eyes are better than few for finding & fixing defects.

2. Desktop integration - across Windows 98, ME, 2000, XP and to a lesser extent Vista.

3. Application integration - there are tonnes of apps writen either embedded in IE, or using IE as a view-port to data, screens, etc.

All of the above (and more) make IE6 a bitch to keep updated quickly and easily. Breaking not just a browser, but OS shell, and tied-apps with a dodgy patch isn't an option for Microsoft and they know it (despite the odd rogue update that slips through the net).

Re:There are three main factors for this

In my opinion, one of the biggest problems Microsoft faces is that web-page structure and syntax is not handled the same way a C++ program's structure and sytax are (as an example); you can make hundreds of syntax and structural mistakes in HTML, CSS and Javascript and IE will still attempt to display your page. I could be wrong, but I heard a couple of years ago that the majority of code in web browsers was not dealing with displaying correct HTML but was dealing with correcting mistakes to display a page. If IE could simply not display incorrect HTML and CSS the code base should be far smaller, which in turn should make it easier to maintain and probably more secure.

Dealing with broken code

If IE could simply not display incorrect HTML and CSS the code base should be far smaller, which in turn should make it easier to maintain and probably more secure.

True. Unfortunately, we've got a decade and a half worth of web pages that were built sloppily. Not all of them, but enough to be an issue, especially since many of them are effectively abandoned and don't have anyone to fix the errors. If it had been designed that way from the beginning, it would be feasible, but there's all that legacy data to deal with. Any HTML browser designed to run on the web, and not just on, say a local set of help pages, has to do something with those pages. Dave Hyatt (of Safari fame) made some interesting comments [mozillazine.org] on the issue when discussing XML error handling in browsers -- basically, learning from the consequences of that decision to tolerate HTML errors without specifying how to recover from them.

Things are a bit better with CSS, as there are explicit rules for how to handle broken code (basically, ignore it and skip to the next line). The bigger problem there is handling code that was written to older, broken implementations -- the IE5 box model, for instance -- and trying to determine whether a page was built for the spec or for the broken implementation. This gets into quirks mode, and doctype sniffing, and things get kind of hairy.

(Then there's the fact that HTML and CSS are both designed with extensibility in mind... any unfamiliar tags or attributes in HTML are supposed to be ignored, so an HTML 3.2 browser can still do something useful with an HTML 4.0 page. But that's a slightly different issue.)

That's nothing

My truck was unsafe 365 days. I could have been in an accident on any one of those days!

This is why I used SetSAFER

Nothing like a quick Software Restriction Policy to "disallow" the use of IE :-)

I also have to admit, that since FireFox 2.0, I can trictly tell my browser which to sites to masquerade as IE.

Quite handy if I do say so myself...

What does this mean?

If IE6 was unsafe for nine months out of the year, what did it give birth to? Inquiring minds want to know...

Moo

For a total 284 days in 2006 (or more than nine months out of the year)

Yep, it took them nine months to get that baby.

284?

Were there only 284 days in 2006? http://developers.slashdot.org/article.pl?sid=06/1 2/21/1836240/ [slashdot.org]

Out of how many?

Out of how many? Uh?

Lobbyist hat on

At MS it is our commitment to better our security on all our applications. In 2006 we spent over 284 days researching and developing a series of bug fixes for our IE product line. This gave us over 98 days where IE was impenetrable to attackers and didn't require the need for any patches. Mozilla would like to claim that there product is safer than ours, yet they admit themselves that they had a period of 9 days where their browser was highly vulnerable to hackers and exploits. IE offers a web experience unsurpassed by any other browsers, compatible with every major website online today. If you choose to use an alternative browser it will still have flaws, but MS Windows allows you to choose, and having choices is what MS is all about. Would you really not want to have a choice in web browsers? Would you really want to only have Firefox and that be the end all be all to browsers? People need to have a choice, that's part of why this great country of America was founded.

I'M A WINDOWS GUY

I use IE for everything and I've never once been hacked by these supposed security holes. I do all kinds of stuff like online banking, eTrade, eBay, online shopping, the works! And it's totally secure because it's all encrypted. Sure, I've had something like $24,000 worth of charges applied to my credit cards that weren't mine, but that wasn't because of IE. That was because I made the mistake of dealing with a few companies that use Linux or some Unix variant (heh, sounds like a disease we're talking about here instead of an OS) for their web portals and they probably got rooted. Open source software is just not safe. The hackers are all over it since it's all out in the open. Once they get a chance to look at how it works, they can easily make it do their bidding. At least Microsoft has the sense to keep stuff private. NO hackers in the entire world could figure any of that stuff out because there just isn't any single person as smart as Bill Gates and his crack team of developers. I wouldn't touch Firefox with a ten foot pole since it's open source. Although they only report the bugs they think they've found, there are probably billions more than MS has in IE because the hackers have a roadmap with open source. It says, "Here's the keys to the kingdom. Come hack me". I Trust MS products because MS is all about making great, innovative software that is secure and robust and stable.

NOTE: The above post is merely a parody of the Windows user who's "got religion". A reasonable Windows user knows better. A reasonable *nix user knows better. Let the games begin...

Lies, damn lies and statistics

Hmm... all hail the Washington Post for very neutral reporting on this one. Although I am BY NO MEANS a defender of Microsoft I feel we have to put this in perspective. How much market share does Microsoft hold vz Mozilla? I would imagine that the people trying to find security exploits are for the most part looking at Internet Explorer... not only does it still hugely command market share, but it's also the choice for less savvy users. A little like a virus comparison between say Macs and PCs... its all a matter of statistics.

Let's not forget...

That TFA can only document "safe" status regarding known vulnerabilities for IE or real browsers.

Someone needs to report that IE (6 and 7) has had craptastic standards support for 2195 days of this century (as of 4 Jan 2007).

Thanks for the money, folks

I made thousands of dollars -- more than half my company's gross revenue -- cleaning up spyware in 2006. A lot of it, probably 30% or 40%, was on fully patched machines with current anti-virus software. Almost every time I read about exploit code becoming available for a zero-day vulnerability, my phone starts to ring.
I have one customer who gets hit three or four times a year. Each time, I get $75 to $150 for booting his system to Windows PE and cleaning off the pests. He's running McAfee Enterprise 8.0i (from his job) with all the "Unwanted Programs Policy" settings maxed out, and he still gets hit, and I still get paid. (I think it may be due to his Web surfing habits, but I don't ask and he doesn't tell).

If Microsoft ever delivers a really secure OS and browser, I may need to go get a job... after all the XP machines die off, that is. Since I still see Windows 98 and ME boxes running (some plugged directly into Comcast cable modems), I suspect that will be a few years yet.

Only 284?

I thought it was in the 360 range.

As Long As IE Runs ActiveX

it's unsafe.

Which means it was unsafe for the last 365 days of last year.

I just did another five hour spyware cleaning last night (which still isn't complete). A fifteen-year-old kid managed to bring a Dell PC to its knees over just a few days of browsing the wrong sites.

The kid was visiting the client. The kid has an Apple at home - so he didn't know what he was doing was death to Windows...:-)

Comes as no surprise

The real surprise was that there actually days when the browser was safe. I would like to see what the stats on IE7 will be at the end of 2007 http://www.cybertopcops.com/ [cybertopcops.com]

The news states wrong.

The author must have forgotten that 2006 was 365 days long, not 284.

Worth noting...

Mozilla's Firefox browser -- experienced a single period lasting just nine days last year in which exploit code for a serious security hole was posted online before Mozilla shipped a patch to remedy the problem."

It's worth noting that I'm betting that nine days was only how long it took for Mozilla to ship the "official" patch to "official" places...I'll bet a number of distros had downstream patches available (at least for submission) within 24 hours.

For anyone doubting ESR's written claim about FOSS's superior ability to squash bugs, you only need to take note of examples like this to know that he was right. Given enough eyeballs, all bugs are indeed shallow.

*Dodges tomatoes headed in my direction with cries of "slavish, unquestioning fanboy!"* ;)

Guess what folks

Guess what folks! Connecting your computer to the internet was unsafe 365 days last year!

Phew!!!!!!!!!

Thank god it wasn't a leap year!

Re:I wonder

Are you basing that on anything scientific? No. Just an uninformed opinion.

Re:Imagine that..

While normally I'd agree with you, the article is from the Washington Post, and is very well supported. Not to mention that there is little "bashing" and much more statistical support.

I am by no means a Microsoft hater. I use many of their products (specifically Windows and Office) because they are simply better than the alternatives, even the free ones. However, I am also not a Microsoft zealot, and realize the company has it's flaws (not talking about business practices, just software) and IE is one of them. I have been with Firefox for several years now, and while that is not perfect either, it is far superior to IE. That isn't intended to be MS bashing, just the cold, hard truth.

This article is absurd

I wonder what windows would add up too

IE and windows are really one big insecurity mash-up that is hard to see individually. Remember the Netscrape lawsuit over bundling IE? When M$ was arguing in court that taking something as insecure as a web browser and tightly integrating it into something that is supposed to be secure like an OS was required for their continued innovation.

Anyway, I think this is absurd. IE6 had a patch available. It was IE7. M$ released IE7 as a "high priority security update" via their built in update process. In the same way that the patch for Firefox was distributed as a later version of the browser through their built in update process. I fail to see the difference. I can see this ending up on slashdot, but the Washington Post really should know better.

The washington Post should know better. As

Re:Firefox... never safe in 2006?

Therefore, only applications using the Apple Java plugin is affected by this vulnerability.

Re:Show me the code

You amaze me, years are 365 days each, sometimes more.

Re:Show me the code

I don't get it. You can't be this clueless, so it must be a joke. Could someone explain it to me?

Re:So?

Try this.

Assuming firefox (2.0.0.1) is open, you are reading this post.
Check memory consumption (windows XP, currently FF consuming 37Mb)
Cntl-click on reply to this 21 times (giving 22 open tabs, 57Mb)
Open each tab, scroll on page.
Close each of 21 tabs (leaving 2, 45mb)
repeat (52mb)
repeat (58mb)
repeat (60mb)
Now I couldn't claim this as somehow exploitable, but it does highlight the behavior during browsing does have an effect on the memory usage. Especially when even a quality product has a memory leak.

Re:Blame the NSCA, not Bill.

Spyglass wrote it. Spyglass licensed Mosaic from the NCSA, but Spyglass Mosaic was a from-scratch implementation. Spyglass then licensed their from-scratch implementation to Microsoft for a percentage of direct sales, which turned out to be a very unfair deal, since IE was never sold.

Re:Nobody Gives a Rat's Ass

Flamebait? Predictable. Pfftph. LMAO @ koh for making me his virtual foe.