Web Devout tidings


Microsoft continues their self-destructive campaigns

Microsoft finally removed the more insulting lines from their Australian $10,000 campaign, but they also released a “Get The Facts” campaign apparently comparing a future version of IE 8 with some phantom versions of Firefox and Chrome invented by Microsoft for this campaign.

Security

Yes for IE 8, no for Firefox and Chrome. Why? Because IE 8’s phishing and malware protection is supposedly “better” than the one that Firefox and Chrome inherit from Google. Microsoft calls this a “fact”.

Now, I’m sure Microsoft could come up with some studies to support their claim (although they didn’t), just as Mozilla or Google could come up with studies to show that they’re better. That isn’t the point. The point is that all three browsers feature rather similar anti-phishing and anti-malware protection, and yet Microsoft used those features as the sole justification for giving IE 8 the only check mark in this category.

Privacy

Yes for IE 8, no for Firefox and Chrome. This time, it’s because of IE 8’s unique Incognito mode.

Er, sorry, I meant InPrivate mode. Incognito is the name of Chrome’s equivalent. But yeah, Microsoft decided that Chrome doesn’t deserve a check mark. Also, despite Firefox’s wealth of privacy features, it doesn’t earn a check mark because the InPrivate equivalent is only available in Firefox 3.5.

Ease of Use

Yes for IE 8, no for Firefox and Chrome. Because everyone knows that a bunch of confusing features that trigger at unexpected times makes the product more usable, and that Chrome’s streamlined experience makes it a hard-to-use piece of trash, right?

I forget, is this called “Get The Facts” or “Get Microsoft’s Opinions”?

Web Standards

Yes for all three. Considering we’re talking about a simple yes-or-no rating system here, I’d say that’s fair. They also accurately point out that Firefox and Chrome have better support for the upcoming HTML 5 and CSS 3 specifications, and they avoided claiming that IE 8 has perfect standards compliance.

Developer Tools

Yes for IE 8 and Firefox, no for Chrome.

Um… what? Chrome actually comes with more developer tools than Firefox out of the box. Of course, Firefox beats both once you add extensions like Firebug and Web Developer, but I don’t understand the basis on which Microsoft decided that Chrome is the one without a check mark.

Reliability

Yes for IE 8, no for Firefox and Chrome. The sole basis for this was the combination of both tab isolation and crash recovery. In my experience, though, IE 8 and Chrome crash much more often than Firefox, so it boils down to whether you’d rather have fewer crashes with worse crash isolation or more crashes with better isolation. I prefer the former, especially when the browser also has good crash recovery (restoring the tabs as they were before the crash), so I’d pick Firefox as the winner in this category. But again, this is a subjective issue, and Microsoft is just expressing their opinion… in a campaign about “facts”.

Customizability

Yes for all three, with a nod to Firefox’s large extension collection. Good so far.

But wait, Microsoft goes further: “many of the customizations you’d want to download for Firefox are already a part of Internet Explorer 8 – right out of the box.” Okay, they said “many” and not “most”, so technically this is accurate, but it’s misleading. According to Mozilla’s add-ons site, the most popular extensions are FlashGot, AdBlock Plus, NoScript, Video DownloadHelper, and DownThemAll. Out of the box, IE 8 doesn’t support the main functionality of any of those extensions. I mean, sure, you could disable JavaScript support in IE 8, but that isn’t close to the same thing as using NoScript. MS Paint technically supports the same editing functionality as Photoshop, but it would be a joke to honestly compare the two.

And yes, I’m aware that IE also has its own type of add-on system, but I’m addressing Microsoft’s “right out of the box” claim.

I’d also like to point out that the level of customizability that IE 8 and Chrome currently support doesn’t come close to Firefox’s capabilities. From the massive amount of settings in about:config, to the unchallenged wealth and diversity of extensions and themes, to lesser-known features like userChrome and userContent, there’s just no contest here. Considering the tiny feature differences that Microsoft used in other areas to deny check marks to other browsers, if Microsoft were being consistent they’d give Firefox the sole check mark in this area.

Compatibility

Yes for IE 8, no for Firefox and Chrome. “Internet Explorer 8 is more compatible with more sites on the Internet than any other browser.” Yeah, when you illegally use anticompetitive practices to shut out other browsers (practices for which you have been convicted in a court of law), and thus hold over 90% of the market for multiple years, effectively forcing web developers to cater to the most nonstandard implementation of web technology in significant use, you tend to end up with the most compatible browser. It’s a bit like a politician bragging about the number of votes he has accumulated in rigged elections.

Manageability

Yes for IE 8, no for Firefox and Chrome. I don’t have experience in the area of enterprise-wide deployment, so I can’t really comment on this aspect.

Performance

Yes for all three. In fact, Microsoft calls it a tie. You know how every single speed test you’ve seen in the last few years has shown Internet Explorer to be the slowest by far (with IE 8 being only a minor improvement over IE 7, and IE 7 being only a minor improvement over IE 6)? Microsoft calls that a tie.

I’ll tell you what, Firefox and Chrome are a lot closer to each other than Internet Explorer is to either of them, and I can definitely notice that Chrome is faster than Firefox at most drawing operations. You don’t need slow-motion video to tell the difference, you just need to look at some of the next-generation web applications that haven’t reached the mainstream yet due to browser performance limitations!

Conclusion

The Internet Explorer development team earned a lot of credit with me on the standards support quality of IE 8. The recent series of misleading and sometimes insulting marketing campaigns has almost completely used up that credit. I assume that other web developers have felt similarly, and if so, Microsoft is doing a lot of damage to themselves by refusing to engage in the kind of honest dialog that Mozilla has particularly excelled at.

6 Responses to “Microsoft continues their self-destructive campaigns”

  1. Rob Says:

    When I first got into the web business, about five years ago, I started questioning Microsoft’s practices and was laughed at. Four years ago I started questioning Microsoft’s browser and was made fun of. Three years ago I said IE would become the poor cousin over the next few years and got banned from a forum.

    Microsoft’s pathetic attempts are so blatantly obvious it gives the impression the boat is leaking but you can’t tell the passengers.

    That’s my current observation. Please don’t laugh at me or ban me.

    Posted using Mozilla Firefox 3.0.11 on Windows.

  2. Tomasz Grajewski Says:

    “I can definitely notice that Chrome is faster than Firefox at most drawing operations.”

    I wouldn’t say, that Chrome is faster than Firefox in drawing operations. It’s slower, at least on this graphics heavy page: World of Merix. And Firefox isn’t so fast either, as it renders mentioned page slower than Opera… Of course IE is still far behind, and its increase of rendering performance is neglible or none (if it really is an increase, as in case of IE8 it’s rather decrease).

    Posted using Opera 9.64 on Windows.

  3. David Hammond Says:

    Tomasz: As usual, it depends on what exactly you’re doing. I was working on a JavaScript-based game engine (a scrolling overhead-view RPG thing like classic Final Fantasy games), and Chrome and Opera totally kicked Firefox’s butt on that.

    In that case, I had a large table of cells that had positioned background images, overlaid with a few small absolutely positioned blocks that also had positioned background images. The scrolling mechanism would generate a new row or column of tiles in the off-view buffer, then position the whole enchilada a couple of pixels at a time until it had scrolled a full tile, and then remove the newly off-view row or column. The row/column generation and removal took a negligible amount of time, and I verified that the bottleneck was in fact the browser’s performance in handling the incremental positioning.

    Chrome and Opera were smooth as silk, with no significant speed difference when the view was scrolling versus when it wasn’t. In Firefox, everything slowed down to about 80% the normal speed while scrolling. IE 8 slowed down to about 50% the normal speed while scrolling.

    I tried a variety of different approaches, including the canvas element (which was the absolute slowest approach I tried, by the way). In each approach I tried, Chrome was at least a little faster than Firefox, and Opera was usually the fastest or tied with Chrome. IE was always out of the game.

    So, while I can verify that Firefox on Windows renders World of Merix a bit faster than Chrome (on Linux, Firefox is slower, possibly due to the same regression that initially made Mozilla’s Bespin unusably slow on Linux), it’s hard to look at any one test case and call it a conclusive assessment of the browser’s relative drawing performance.

    Posted using Mozilla Firefox 3.0.11 on Linux.

  4. Gérard Talbot Says:

    Web Standards
    Yes for all three. Considering we’re talking about a simple yes-or-no rating system here, I’d say that’s fair.

    David Hammond,

    I’d say you are making a very bold statement here. Let’s try to discuss it in a structured manner…

    Are you saying that UAAG 1.0, HTML 4, DOM 1 Core, DOM 2 Core, DOM 2 Events and ECMAScript-10262 3rd editions are not web standards? or that the above specifications of techical recommendations shouldn’t be considered accepted and confirmed web standards by web browser manufacturers?

    DOM 1 Core results:
    Internet Explorer 8 67.6 % 161 tests passed out of 238 tests
    Firefox 3.0.9 93.7 % 223 tests passed out of 238 tests
    Opera 9.64 89.1 % 212 tests passed out of 238 tests
    Am I missing something here?

    DOM 2 Events: Internet Explorer 8 scores 0. 0 support. 0 compliance. Am I exaggerating here?

    If png file type is now correctly supported in IE 8, then why can’t I use a png image file as favicon/webpage icon? Why do I must use an .ico file to identify my favicon/webpage icon? .. like this W3C QA “How to Add a Favicon to your Site” tutorial explains/recommends

    Bug 364028 at connect’s IE beta feedback has been closed and has not been reactivated. Is that ok with you?

    Regarding ECMAScript-10262 3rd edition: are you actually claiming that IE 8’s implementation of JScript is compliant with the spec? Have you ever read a document titled JScript Deviations from ES3? Have you ever read this post from Garrett Smith regarding JScript?

    Even if we consider only and exclusively CSS 2.1, there are people (James Hopkins and me and possibly others) who believe IE 8 has a good bunch of implementation CSS 2.1 bugs that other browsers (Firefox 3, Opera 9.6, Safari 4, Konqueror 4) do not have.

    Regarding web standards, security, accessibility, customizability, etc…, each and all of such topics can not be resolved or assessed with a blunt yes or no.

    regards, Gérard

    Posted using Konqueror 4.2.4 on Linux.

  5. David Hammond Says:

    Gérard: In retrospect, I probably should have approached that point the way I concluded the Customizability point. Of course there are large differences between browsers, and IE still stands apart from the crowd as having the worst overall standards support by a sizable margin. They were at least forthcoming about their inferior implementation of newer specifications, as I noted, although they still lag in various other web standards as you noted.

    I agree that what this comes down to is that this black-and-white check-or-no-check system is misleading, no matter where you draw the lines. If I had to use a black-and-white system, I’d be much more forgiving about feature differences than Microsoft was on the points that they believed favored them. I wouldn’t say that IE 8 *doesn’t support web standards*. All things considered, it’s actually pretty good at supporting web standards. The other browsers are generally better.

    Posted using Mozilla Firefox 3.0.11 on Linux.

  6. Magne Andersson Says:

    You know, this list was even worse when it first was published. IE had all the checkmarks while there were only two ties. I read from one person on Twitter that he, through some help from a few friends at Microsoft got it cleaned up a bit, and making it a bit more fair. But the first version (yes, I saw it) was horrible. In this version the comments is modified too, the old ones were even more unfair to the other browsers.

    Posted using SeaMonkey/Mozilla Suite 20090531 on Windows.