Talk:Criticism of Internet Explorer

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Microsoft / Windows (Rated B-class, Low-importance)
WikiProject icon This redirect is within the scope of WikiProject Microsoft, a collaborative effort to improve the coverage of Microsoft on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This redirect does not require a rating on the project's quality scale.
 Low  This redirect has been rated as Low-importance on the project's importance scale.
Taskforce icon
This redirect is supported by WikiProject Microsoft Windows (marked as Mid-importance).
 

Previous Discussion[edit]

The Criticisms of Internet Explorer article came into existence in May 2005, and is intended to eventually provide a neutral, concise, yet thorough summary of the criticism that has been heaped upon the Internet Explorer web browser in recent years.

I would like to iterate that the critics are primarly fire fox or non-ie users and thus, it is biased. I will accept the external link oppinions and only those oppinions.

The article's content was separated from its previous home in the Criticisms section of the main article, where it was growing excessively lengthy and where it had been the subject of much debate. The debate was over specific technical issues mentioned in the content, as well as the arguably less-than-neutral way in which topics were often framed.

Please note that relevant discussion that occurred prior to the creation of this article was not moved from Talk:Internet Explorer. When starting a new discussion thread here, please check the following pages to see if there was prior discussion of the topic:

  • Talk:Internet Explorer - current discussion
  • Archive 1: Discussion prior to January 30, 2005 - topics covered include market dominance; IE for Unix; Mozilla evangelism; Mac OS X image caption; standards; IE features; competing browsers; removal of IE; Microsoft's bug database; edit wars; cost of IE; anti-IE CERT advisory; misc NPOV issues
  • Archive 2: Discussion prior to July 9, 2005 - topics covered include NPOV issues; JavaScript; Microsoft AntiSpyware; Embrace, Extend and Extinguish and vendor lock-in; presence of Firefox-related content; BHOs; web standards and degrees of conformance; languages; article headings & organization; PNG; peer review and featured article submissions; Sandbox and Chrome; the criticism section; impact of Firefox advocacy on IE's decline; ActiveX and eBay

If there is prior discussion, please continue it here on Talk:Criticisms of Internet Explorer, not on the old page. — mjb 05:48, 24 May 2005 (UTC)

Cause I want to[edit]

actually nvm. Whole page is but a series of uncited rants by FF users.

Unsigned msg from when? OK, now, I happen to be a firefox user, an epiphany user, an Opera user, and an IE user. That's because I'm a web developer. The article is not biased, it's NPOV true. Before I was forced to develop twice, once for MSIE, once for all other browsers. Now, since IE7, I'm forced to develop thrice - once for MSIE7 too. And no new functionality added (mixed XHTML/SVG highest on my wish list). Said: Rursus 13:24, 8 June 2007 (UTC)
Cause firefox is better. haaa. —Preceding unsigned comment added by 86.163.241.72 (talk) 20:09, 10 December 2007 (UTC)

Digest Auth[edit]

Something about its messed handling of digest auth would be nice Tc1415 14:39, 12 May 2006 (UTC)

General Comments[edit]

It's possible to summarize the article by saying "IE sucks." CoolGuy 19:44, 19 May 2005 (UTC)

This would do well to demonstrate the ill-informed standpoint of the anti-IE, pro-OSS zealot. Go ahead --Beachy 17:07, 20 May 2005 (UTC)
How is it "pro-OSS zealotry" when people using non-OSS browsers (Opera for example) think "IE sucks" too?
How is it "pro-OSS zealotry" when one the engineers who created IE switch to Firefox and says "IE sucks now"? -- Anon 11:39, 21 Sep 2005 (UTC)
I guess you are trying to say that not everyone who is anti-IE is pro-OSS. Fair enough. But the IE articles here on Wikipedia have a long history of being vandalized by people with an agenda, and more often than not, these people valiantly uphold the stereotypes that Beachy made reference to.
You also undermine your argument by misquoting an article that you most likely read about on slashdot, whose collective anti-MS, pro-OSS bias is unparalleled. And for some reason you didn't want to mention that the engineer in question criticizes Firefox in his article, or that he switched mainly out of disappointment in IE's feature lag rather than for its security, standards support or general usability. He was not vitriolic or particularly scathing of IE at all — "IE sucks now" is your embellishment, not a quote he made. — mjb 21:01, 21 September 2005 (UTC)
It's not possible - for the developer it's necessary to state how version X sucked and how to work around it. Said: Rursus 13:27, 8 June 2007 (UTC)

Firefox ok, Opera not?[edit]

Why is a link to a Firefox-promoting article all right but one to an Opera promoting not? The latter was removed promptly by Minghong. Yet the "Firefox Leaves No Reason to Endure Internet Explorer", "Internet Explorer Is Too Dangerous to Keep Using", "How to Protect Yourself From Vandals, Viruses If You Use Windows" and "Securing Microsoft Windows (for Home and Small Business Users)", articles which all recommend / promote Firefox and Firefox only, are ok to link to!?

I'd prefer removing all browser-specific links instead of adding more browser. I.e. get rid of footnotes 12-15.


Comment for "Firefox ok, Opera not?": I agree. The footnotes link to a glowing Firefox review by a journalist who is known to be an eager Firefox fan, but linking to Opera is not allowed? Look at secunia.com, and you will see that Opera has no known unpatched vulnerabilities, while Firefox has several. If Firefox is supposed to be an example of excellent security, why isn't Opera?
I'm sorry but from the title, it was browser-specific. Hence the link was removed. That edit was not a biase against anything. And yes I agree this article needs some cleanups to remove unnecessary references to other browser (aka non-IE browser promotion). --minghong 6 July 2005 16:44 (UTC)

Win XP sp2 only available for Win XP[edit]

Do we really need to assume people are that stupid? —User:ACupOfCoffee@ 18:56, 7 April 2006 (UTC)

I changed the wording on this to "Note that the security features introduced with Windows XP Service Pack 2 (and Windows Server 2003 Service Pack 1) are not available for earlier versions of Windows, including Windows 9x, NT and 2000." Does that sound better to you? Warrens 19:49, 7 April 2006 (UTC)

Article changes[edit]

I've been working on a lot of changes and improvements to the article, to help improve the overall encyclopedic value of the article.

Most significantly, I'm converting the article to use REF tags for references, rather than inline external references. This brings us much more in line with WP:CITE. I'm also using the cquote quoting style, which looks pretty nice, and gets those important quotes into a position of better readability.

Other things I've done:

  • Removed the SecurityFocus vulnerability count; their web site doesn't really give us any easy way of seeing how many outstanding vulnerabilities there are, so the number "27" that was previously in the article was not easily verifiable. Secunia is a respected source, it should stand well on its own.
  • Clarified that the vulnerability that led to Download.ject was specific to IIS 5. Also removed two of the three external links associated with the subject; one was an invalid URL that didn't point to relevant information, and the other contained a lot of technical detail about the virus, which is suitable for the article on the virus itself, but not this one.
  • Added information about UAC and Protected-IE features in Vista.
  • Added a counter-balance to the information about Apache having fewer vulnerabilities than IIS. It's not true anymore! It certainly was the case prior to IIS6's release, but it isn't now. I've added Secunia references demonstrating this.
  • Removed the link to David Wheeler's paper on how to secure Windows. Issues of his being unabashedly pro-Linux aside, there are a lot of technical faults in his paper, largely because information he leans on is years out of date. Apart from that, the paper covers securing many aspects of Windows, and only a small portion is given to Internet Explorer.
  • Removed the term "non-standard" from the description of ActiveX. The ubiquity of ActiveX makes it "standard" in one literal definition of the word, so to describe it as "non-standard" isn't exactly correct. I rewrote the sentence as "ActiveX controls are also Windows-specific applications and are not portable to non-Windows platforms without significant effort.", which I feel is a pretty accurate assessment.
  • In some places where a person or source is cited, I've noted what year it's from. This will help lend some real context to what's being said.
  • The reference for the book "Malicious Mobile Code" has been changed from a Groklaw web link to a proper book citation.
  • I expanded a couple of quotations to balance things out a bit more.
  • I've found that there were a number of mis-represented dates as well, I'm fixing these as best I can.

... and that's just in the "Security" section of the article, phew! I'll be working on this for a while longer; I'll update this with further changes I've made. Warrens 04:07, 8 April 2006 (UTC)


The cartoon style quotation marks are unnecessarily showy, IMO.

Acid 2 Test[edit]

Microsoft has publicly stated that the next version of Internet Explorer (version 7) will not pass the Acid 2 test. Could we elaborate on this?

See the Acid 2 page for the lowdown.

5 years of development, no results[edit]

Perhaps a mention of the fact that it has taken Microsoft 5 years to add PNG alpha support, tabs, and pretty much nothing else, is needed? They haven't done anything (besides PNG) to ease the ridiculous amounts of extra code needed to make a standards-compliant site work in IE. PNG support and tabs seem like they would take a week to implement and a couple months to test, which means the IE team has been standing around with their heads up their asses for 5 years. Pitiful. I'd add it myself but I could never write about this in an NPOV way. - Noclip 12:36, 25 July 2006 (UTC)

Because Internet Explorer 7 has no problems with PNG and JPG, shouldn't we delete that part about nonsense of Internet Explorer not supporting png?

The reason why it takes so long for MS to develop something for IE is prioritization. In the field of computers, .NET is the highest priority, then Windows, then Office then many many others and then comes IE. Jancikotuc 18:10, 8 September 2006 (UTC)


IE7 has inconsistent support of PNG. One PNG format image will look fine on a colored background or on top of a GIF background, but if you stack 2 PNG images on top of each other (such as an IMG over a background), the transparencies won't work. So basically, 5 years of development, and STILL nothing.Applesanity 18:28, 28 January 2007 (UTC)

NPOV[edit]

This large article contains mostly descriptions of things that should be references in the main Internet Explorer#Criticisms article. Criticisms by definition are opinions and should have no place in Wikipedia. Whatever is legitimate in this article should be distilled and merged with the Criticisms section in the main article, which is already quite detailed. GreyWyvern 15:41, 10 August 2006 (UTC)

I think that all the Cricitism pages (IE, Opera, Firefox) should be deleted, they are not Encyclopedic. Things that can be merged should be merged in the general article (not under a "Criticism" header). That's my 2 cents on the subject. -- AdrianTM 22:07, 10 August 2006 (UTC)
If the article only contained statements such as, "Joe Blog doesn't like Microsoft," I would agree with you. However, the criticism pages for IE, Opera, Firefox, etc, point out factual shortcomings and contain valuable information for web designers and web users alike. --Dbolton 17:14, 11 August 2006 (UTC)
Much of the information on this page seems out of date. Some of it is even inaccurate. For example, the claims that IE's rendering engine has not had any major changes since version 5. Even if you discount the recently released IE7, IE6 had significant rendering changes, such as fixing the faulty box model rendering of IE5 in standards mode. Before this information gets merged, i'd suggest a lot of changes... 12.207.87.61 20:21, 26 October 2006 (UTC)
Actually, IE7 still doesn't understand the box model. It cannot differentiate between how margins and paddings should interact when stacking block elements. It still doesn't understand that the width and height attributes define the width and height of your content, not the entire box. It still doesn't understand how floating elements should behave inside containers. It still doesnt understand the differences between static elements and relatively positioned elements. I could go on.....
I've had much experience with most of the well-known browsers, MSIE, Firefox, Safari, Netscape, Konqueror, and Opera. No matter how you do the math, IE7 falls far short of its competition. The thing this page is lacking in is proper citation - but common knowledge works just as well. The only reason IE commands market share is because practically all PCs come with IE pre-installed. And just so we're all clear, I'm an Opera user.
For anyone who has used MS' "Small Business" services, one of the most surprising things to notice is that IE will not recognize its own Microsoft SSL Certificate.Applesanity 10:15, 27 January 2007 (UTC)

Dumb error message: "The Page Cannot Be Displayed"[edit]

That message is so annoying that Internet Explorer is now notorius for it! No wonder I'm now using Firefox for all my internet browsing needs... and on a Mac, too! To think I actually used to use Internet Explorer back when we first got our old Windoze XP machine. Heck, I actually liked Micro$oft! But then I realized how bad it was (damn you pop-ups! DAMN YOU!!!) I got Firefox (for Windoze) and later an Apple MacBook (The white one, not the black one. I highly recommend it for those old Windoze users who would like to make "the switch") and Firefox for that. I just couldn't get into Safari! Share your Firefox/Non-IE stories, please.

This is not a forum. However, even some web servers mimic the IE's error message... Yuk, you can even get with Opera under Linux, e.g. if you mistype a website URL on such server.

If I recall correctly, you can disable "friendly error messages" and show more detailed ones. This option is hard to find, however. —The preceding unsigned comment was added by 67.184.87.2 (talk) 20:26, August 23, 2007 (UTC)

The IE blink when changing page[edit]

I have not seen any discussion of the irritating feature of IE - including the latest version - whereby whenever there is a change of page the screen blinks for a noticable period of time, even when most of the page content is unchanged. This does not happen with Firefox and some other browsers. Why is this, and why has it not been addressed in the latest IE version?

Geoff97 22:02, 12 September 2006 (UTC)

I doubt that could be considered an Encyclopedic fact. Remember that this is an Encyclopedia not a forum where to complain about I don't know what obscure problem with your browser. We also need to avoid original research. -- AdrianTM 10:41, 15 October 2006 (UTC)

You are right. Point taken. Geoff97 19:40, 18 October 2006 (UTC)

IE7 tabbing[edit]

Actually the user can define wherether when opened another tab, it has to display a blank page or the homepage in IE as of IE 7.0 RC1. So i had to delete the following message as inaccurate: ===IE7 tabbing=== While beta versions of IE7 include a useful feature for opening multiple web pages in a "tabbed" layout the "open new tab" command opens a blank page rather than the default startup page

Illuminati 23:56, 7 November 2006 (UTC)

Hoaxes[edit]

To describe this section as OR is a misunderstanding of that policy. The source contains crticism and highlighting this is perfectly OK. TerriersFan 21:21, 20 November 2006 (UTC)

I motivated my revert this way: I don't think "hoaxes" is encyclopedic material, also you'd have problems proving that hoaxes represent a form of criticsm. -- AdrianTM 21:27, 20 November 2006 (UTC)
The text of the hoax manifestly contains criticisms. Whilst the hoax may not be encyclopaedic the concept of using a hoax website as a form of criticism certainly is. TerriersFan 21:43, 20 November 2006 (UTC)
Let's wait for a 3rd opinion on this issue, I remain at my opinion that's not something encyclopedic and a hoax is a hoax, not a criticism. -- AdrianTM 23:07, 20 November 2006 (UTC)

NPOV (Again)[edit]

This article is completely POV. It talks about problems that have already been resolved (such things would only be appropriate in a "History of Criticisms of Internet Explorer" page). Also, a criticisms page is inherently POV. After this article is cleaned up, it should be merged with Internet Explorer. For the record, I am a FFX evangelical (see my user page). -Kingy 20:36, 28 December 2006 (UTC)

ECMAScript Implementation[edit]

The article currently states that MSIE provides full implementation of ECMAScript. Isn't E4X also a part of ECMAScript too that no version of IE currently implements? In any case, someone with sufficient knowledge please include E4X discussion in the article. —Preceding unsigned comment added by Mdoubledragon (talkcontribs) 06:01, 6 November 2007 (UTC)

ActiveX/COM[edit]

I think the section on COM should be split into something on COM and something on ActiveX.

COM is just the component model for IE; there is nothing directly harmful from it. Where it is weak is that Browser Helper Objects, BHOs, are plugins (yes, just like FFX plugins) that can fit in on the http request chain. I dont know what their original intent was, but they get abused by spyware to track browsing patterns, and by malware to intercept submissions to HTTPS sites. It is no less insecure than the plugins for any other browser, given that there is no automatic download of BHOs.

BHO are same as FFs Extensions. The former are compiled binaries the latter are usually written using JavaScript. They can be used for various tasks like screen reader, flash blocker, content filter, ... -- 62.178.136.129 19:42, 2 January 2007 (UTC)

ActiveX, on the otherhand, is a security disaster and the citation (#15) is wrong. Here's why

No more then other browser extensions that consume data from the web (i.e. Firefox extensions, Gecko plugins, Opera plugins, ...). -- 62.178.136.129 19:42, 2 January 2007 (UTC)
  • Originally, Ole Control Extensions, OCX objects were written, with the goal being the 32 bit successor to VBX, Visual Basic Custom Controls. OCXs were designed to be reusable lumps of code that developers using Visual Basic, MS Access, Visual C++ and other dev tools could use to implement complex bits of the app. Drag it from the list of objects to the gui, fill in some attributes, write event handlers and you have something integrated with your app. IT was such a good design that it got used extensively by Windows applications. You'd make your core app a set of components and glue them together. To redistribute your app you'd include the OCX controls (as DLLs) with your app to make one big installer. Downstream users would not get the .LIC file containing a developer license and so not be able to go into design mode with the controls, but still use them in the apps they came with.
  • ActiveX took that idea and hacked it onto HTML. Any HTML page could declare a OBJECT link to use an activeX control. If the link specified a later version, or the special version tag to force a download, the remote file would be downloaded and installed. A signature would be checked to see if it was valid. If the control was already on the system, it was trusted.
The exact behaviour depends on IE configuration & version, Windows version and user rights. It won't work that way since Windows 2000 (at least for LUAs). -- 62.178.136.129 19:42, 2 January 2007 (UTC)

This leads to the following problems

  • All installed AX controls can be instantiated. That includes all OCX controls with the right COM interfaces.
That once again on depends on IE configuration and version -- 62.178.136.129 19:42, 2 January 2007 (UTC)
  • There is no sandbox for the code; it runs (pre IE7 on Vista) with the full rights of the user
Windows XO offers a protected mode as well (Run as.../"Protect my computer...") -- 62.178.136.129 19:42, 2 January 2007 (UTC)
I suppose you could run IE in a different user account, though it's pretty are. SteveLoughran 19:13, 6 January 2007 (UTC)
  • Because all original OCX controls were written with utter trust in their callers (who were embedding them in apps), they usually offered broad read/write capabilities to the host PC.
While true this is not a problem of ActiveX - the ActiveX specification does not demand this. -- 62.178.136.129 19:42, 2 January 2007 (UTC)
The ActiveX specification added an optional interface in which a component was told in which zone it operated in. I don't know of any commercial controls which implement this, though I have done so in some that I have implemented. It's the exception, rather than the

rule SteveLoughran 19:13, 6 January 2007 (UTC)

  • Newer ActiveX controls may have been written with emphasis on safety, but the tendency of buffer overflows (possibly amplified by a brittle Win32 API), means that they can still be a source of buffer overflows.
This is true for any extension mechanism (Gecko plugins, ...) and not specific to ActiveX. -- 62.178.136.129 19:42, 2 January 2007 (UTC)
Agreed. Some languages and APIs do appear to be more prone to this problem than others, with C and pre-STL C++ being particularly bad.SteveLoughran 19:13, 6 January 2007 (UTC)
  • Spyware authors can buy a certificate and sign their apps. There is no automatic check for certificate revocation.
Unfortunately yes and not specific to ActiveX. The same is true for Java applets / certificate revocations are checked, iirc -- 62.178.136.129 19:42, 2 January 2007 (UTC)
Signing Java applets is optional. By default applets are untrusted. Even signed Java applets are not that prone to buffer overflows, though there is a font on unbuntu that crashes java1.5 every time I try and render it.
  • IE's Zone model gives 'trusted' sites the ability to download and install ActiveX controls unprompted. Any zone privilege escalation model can let something malicious force in an ActiveX control. For example, when IE4/5 would save cached HTML files to well known locations/filenames, malicious sites could force a load of their cached page (which is now on the local system and trusted) and so trigger an automatic download of an ActiveX control. This is why modern browsers use less predictable naming schemes.
The exact behaviour depends on IE configuration & version, Windows version and user rights. Not a problem in IE6 and later -- 62.178.136.129 19:42, 2 January 2007 (UTC)
IE6 is still vulnerable to 'zone escalation', if you can convince IE that data comes from a site with more trust.

The signing of ActiveX says that a developer is either well meaning (but potentially incompetent) or malicious and yet prepared to pay for a key. Without a sandbox to run untrusted code, signed things get unlimited rights. Which means the HTML page using the activeX control may have unlimited rights too, if the ActiveX control has some security defect in it. And, because you can always force a download of an older version of a library, there is no way to reliably revoke an old version of the app.

This is even true for Java applets (signed applets usually have unlimited rights) and not specific to ActiveX. Automatic download of ActiveX controls depends on on IE configuration & version, Windows version and user rights. -- 62.178.136.129 19:42, 2 January 2007 (UTC)

So yes, ActiveX is a design disaster. Java has a sandbox, though Java Web Start drops to the 'give signed apps unlimited rights'. Microsoft's .NET framework remembers where things came from and always gives them the rights of their originating zone.

Signed Java applets usually have unlimited rights, too. Every browser extension technique (plug in, ActiveX control) that is working with external (internet) data is a security risk. This is not specific to ActiveX. -- 62.178.136.129 19:42, 2 January 2007 (UTC)

Returning to ActiveX, the first main discussions of its flaws were in the ACM risks digest.

  • Risks18.61 March 1996, Richard M. Smith discusses how built in ActiveX controls could be used for denial of service attacks.
  • risks18.82 Feb 1997: first (unsigned) hostile AX control demonstrated.
  • risks18.83 March 1997: first issues discussed
  • risks18.18 March 1997: Microsoft Response "ActiveX is safe"
  • risks20.44 June 1999: Steve Loughran provides first public demonstration of a microsoft-signed AX control bundled with IE5 being a security vulnerability (my work:)
  • risks20.50 July 1999 Richard MS mith buys a retail Win98 PCs

shows it is full of bundled activeX controls that provide access to the machine from untrusted web sites.

Old stuff, very old stuff. None of this works on fully patched windows system (at least with Windows XP SP2 and default configuration). Nevertheless most of this not specific to ActiveX even Java applets can be used (and even unsigned applets that are exploiting errors on the byte code verifier). -- 62.178.136.129 19:42, 2 January 2007 (UTC)


One important difference between native AX controls and Java, the CLR, Flash and other VMs is that if any of those tools have a VM weakness it is the duty of the runtime vendor to issue a high priority alert. Its only one place to fix, and once the patch is rolled out then the problem is gone. Whereas with AX the tendency to overflow the buffers is intrinsic. That said, because the VMs are slightly more secure, they get trusted more, flash is turned on widely, Java less, but still often allowed against untrusted sites. Whereas nowadays, nobody in their right minds should be downloading an activeX control off the network. Its so dangerous that people are cautious. That is: the likelihood of a VM having a problem is less, but the potential for exploitation of a hole greater. SteveLoughran 19:13, 6 January 2007 (UTC)
Regardless of the IE7 status, the ongoing problem should be covered. I agree that signed Java applets are also a weakness, but we are fortunate in that they are not that common. There has been at least one recorded spyware vendor abusing them to get spyware in through mozilla, however. Maybe we should have a "security issues of mobile code" page, covering the whole idea of executing code from places you don't trust. SteveLoughran 19:13, 6 January 2007 (UTC)

I'd give Richard M Smith credit for hypothesizing the idea of bundled ActiveX controls being inherently unsafe, together he and I provided the proofs of concept. When I look back at that time, I cannot get over how naiive we all were. It was very much a hypothetical problem, because spyware was nonexistent, rootkits to relay spam and host phishing sites inconceivable. If we hadnt worked to show ActiveX up as a mess, I wonder whether things would be better. Equally importantly, if we had shown it bad in 1997, rather than waited until 1999, how would things be different.

Most spyware is spread through email or im using social engineering, nowadays. Nothing would have changed, I guess. -- 62.178.136.129 19:42, 2 January 2007 (UTC)

anyway, that's the main history of ActiveX security, go [ http://catless.ncl.ac.uk/php/risks/search.php?query=ActiveX search] the risks digest site for more context. I'm going to point out my role in the process, but not cite myself on the page, and think that Richard M. Smith put a lot more effort it. If he doesn't have his own page, he deserves it.

SteveLoughran 23:14, 31 December 2006 (UTC)

To sum up:
Every browser extension technique (plug in, ActiveX control) that is working with external (internet) data is a security risk. This is not specific to ActiveX.
Every browser that downloads code (ECMA Script, Java applets, .NET WebControls, Flash applets(*) ) is insecure.
(*) Flash is a VM and byte code language
-- 62.178.136.129 19:42, 2 January 2007 (UTC)


Those that run code outside a sandbox (AX, Signed Java applets) can do most damage, and those that are written in languages prone to pointer errors and buffer overflows are the most exposed. ActiveX is not only the combination of the two, but because of all the build in AX components, retail PCs have often been vulnerable out of the box.
As I said above, maybe we need a 'risks of mobile code' page, which could include Lisp eval() statements, .NET CLR, etc, and look at their different features and risks. Certainly the [Mobile_Code] page needs some expansion. SteveLoughran 19:13, 6 January 2007 (UTC)

Unclear Error Messages[edit]

1) This section needs sources.

2) IE tries to make technical error messages more user friendly. These friendly messages do not help experts (webmasters, administrators, etc) but do not hinder average users (they would not have any clue what went wrong if shown more detailed error messages)


I'm not sure what kind of source you're looking for. It's as much of a fact as anything else, not something theres really going to be a huge reference for. I could take some screenshots if that would help without violating the no self-research rule, but really I don't see the point. It's just a 'feature' Microsoft forced upon us with the only way to disable it being a registry edit.

I'd also disagree with your statement that they don't hinder average users. I was an "average user" once. Thanks to there always being things over my head always making me want to learn more and advance my knowledge, I'm now your typical techie. I know plenty of others like that. No formal computer schooling, no book reading, just enough not being pampered that eventually you just start learning things on your own. By babying users you remove that possibility and hinder them by putting a bag over their head and telling them theres nothing outside of it. Kind of like cars with the "service engine" light instead of actually telling you whats wrong. Cars have the excuse of not having proper means of display and it costing a lot more to feed real debugging info to you though, computers don't.

Sorry for the long rant, but they don't call it a talk page for nothing. --Semi 18:44, 19 March 2007 (UTC)

Cleanup needed, now need to say IE6 explicitly in many places[edit]

Cleanup tag placed on article. Please do not simply remove, rather volunteer to pitch in.

A lot of the material is now dated as parts of it relate only to IE6 or earlier, and the article needs to say so explicitly where appropriate. Need to change from present tense to past tense in many instances. Volunteers to help clean up these points? CecilWard 13:06, 3 April 2007 (UTC)

Proof of detremental effects on PNGs?[edit]

"The lack of support for PNG alpha channel in versions prior to IE 7 has resulted in a reduced usage of the PNG image format on web pages."

Is there any proff of this?

Pod 11:57, 27 April 2007 (UTC)

Removed vague section about IE7[edit]

I commented out the subsection about how IE7 doesn't support standards. The entire section consisted of a single sentence saying that IE7 doesn't support standards. There was no mention of which standards, no further explanation, nothing. I would support uncommenting this section, but only if it can be greatly expanded. Otherwise it is meaningless. Roachmeister 15:02, 16 May 2007 (UTC)

Removed section about download size[edit]

I removed the section about download size. Wikipedia:No original research is Wikipedia policy and this section was nothing more than an exercise in presenting a Wikipedian's research on the subject. -/- Warren 15:45, 2 June 2007 (UTC)

SVG[edit]

I think it's worthwhile that some mention be given to Internet Explorer's lack of support for another open standard, SVG. --Rktur 17:03, 23 June 2007 (UTC)

Skewed[edit]

This page is completely skewed and biased towards IE and does not even scratch the surface of it's far to numerous problems!

Whoever made it should be ashamed of being so biased and dishonest.

Criticism of Firefox is banned, why not IE?[edit]

All criticism of Firefox is immediately removed and not allowed. In fact the entire "Criticism of Firefox" article was permanently removed and protected so that no one can enter any criticisms. So why is this rambling article allowed to exist? Couldn't be because wikitards are biased... no of course not.

I am against all "Criticism" articles or sections in general. They attract trolls and fanboys, as we clearly see here, and poor quality contributions, moreover, they are magnets for POV and dubious material and dubious sources (forum posts, opinion pieces, etc). I do see however a difference between Firefox and IE, first of all IE is installed on 90% of computers by default so if there's a problem with it -- that problem is automatically an encyclopedic item, IE also set and continue to sets the de facto "standard" on the Web, pretty much every Web designer has to design for IE, if IE does something nasty is again, unfortunately a thing of notable importance. Who cares if Opera, or Firefox, or Konqueror, or Safary, or any other browser does something nasty? Is it of encyclopedic importance? Is it something notable? If it's something notable it should be explained in the article of the respective browser not in a POVish "Criticism of ...." article, not even in a "Criticism" section, but integrated in the article wherever that issue belongs, I would support the same thing for this article, only that there are too many items that can't be nicely integrated in IE article and as I mentioned many of these things are of notable importance. (and no, I usually didn't respond to unsigned posts, this is for the benefit of other good faith editors) -- AdrianTM 03:05, 14 July 2007 (UTC)
I agree to the fact that IE still is the standard and problems with that browser are automatically an encyclopedic item, firefox currently however also is 10-30% of most website visitors (depends on the target of the website, if it targets old people that percentage is generally much lower) and shouldn't be forgotten in Wikipedia. I agree to you however, critism should not be a part of Wikipedia. Wikipedia is en encycolopedic website and should only contain facts that are backed up by citations.
Source: http://www.w3schools.com/browsers/browsers_stats.asp
+ Alot of google-ing to unprotected awstats Jaapvstr 00:54, 23 September 2007 (UTC)

omg! Does Microsoft pay them to edit this article?[edit]

Can't anyone put this on their whatchlist and prevent it from being... vandalized!? What's with the billion [citation needed] tags? I know I have to assume good faith but adding [citation needed] after a statement that XHTML is the successor of HTML is just Idk... vandalism. Or is it some really evil irony? I think both articles are explicit about the fact that XHTML is the successor of HTML and W3C apparently doesn't care if Microsoft thinks that we should stick to good ole HTML.

What's with adding three [citation needed] tags in a paragraph where the first sentence (having two references) has been explicit about the subject matter? It reminds me of the clause of not using wikipedia policies (e.g. "articles must have reliable sources") to subvert what the policies were actually intended for. 354d 07:40, 27 September 2007 (UTC)

The XHTML one was kind of ridiculous, sorry, my apologies. I've removed it. And no, of course I don't get paid for this. My only reward is the satisfaction from the knowledge that this is one more article that fanboys have a hard time dumping hatred on Microsoft and letting it stand as an unchallenged fact. Matt Yeager (Talk?) 07:08, 30 September 2007 (UTC)
Criticism sections and articles are BAD (they should be avoided in Wikipedia), if somebody doens't take care of them they are magnet of original research and unsourced info. I actlually like to see the info where it says that XHTML is the successor of HTML, even if that might be obvious for some people but as Wikipedia put it if the info is obvious you can surely find a reference for that, if you can't find then that's WP:original research. -- AdrianTM 18:31, 2 October 2007 (UTC)

r u serios?[edit]

u canot be serios. --Henry W. Schmitt 06:24, 30 October 2007 (UTC)

Non-criticism[edit]

It's seems more than half of this article is stuff like "IE does (such and such bad thing)." rather than actual coverage of people criticizing IE. Can someone help me fix this? Josh (talk | contribs) 06:58, 10 November 2007 (UTC)

Merge to Internet Explorer article[edit]

I think the material from this article should be integrated in Internet Explorer page. Ideally it should be integrated organically not in a "Criticism" section that would attract unreferenced and poor quality contributions. Criticism articles and sections should be avoided (that's Wikipedia recommendation). BTW, since I will probably be accused of promoting I don't know what special interest, let me make it clear that I opposed a "Criticism" section/articles about Firefox and Opera too and I just proposed to remove a "Criticism of Linux" link. Software is not a piece of art that requires a criticism and praises or reviews section, if there are notable, encyclopedic downsides they should be mentioned inside the main article, but not in a separate troll-attracting section. Thanks. AdrianTM (talk) 15:36, 1 February 2008 (UTC)

Votes[edit]

WAIT A MINUTE!!! Why merge an article like this into IE's article, but then fail to add the proper section to the article?!--Dullstar (talk) 05:09, 12 July 2009 (UTC)

Wikiproject Computing[edit]

I have removed the WikiProject tag, as this article is either a redirect or deleted. If you oppose, please restore the tag. Thank you, fahadsadah (talk,contribs) 16:09, 30 March 2009 (UTC)