Microsoft Windows AOL

it'll happen...

The unfortunate part of all of this is how AOL strings along the Mozilla community. For how long now AOL has teased at using Gecko for their rendering engine, even going as far as putting it out to their beta testers (to what most beta testers considered and absolute success) only to withdraw it in the next beta.

My personal feeling is that AOL has used Netscape as a leveraging point to extort what they want from MS. They could have realistically introduced the Gecko engine at any point and used the beta tests to show MS just how easy it is to swap the engine out "look it's not in 6 - now it's in 7 - oh look we took it out of 8". They never really intended to swap the engine out or it would've been done already.

Nevermind the fact that Gecko was never ready for prime time, not rendering many popular sites as well as IE and being a resource hog...

eh... that's a questionable statement. I've seen plenty of sites that render identically under the top browsers (IE5/6/Mozilla/Netscape6/Opera6/7) (more and more each day). Almost every one of those sites if you look at the underlying code it's standards based. The real problem is that many developers, even/especially "professional" ones, tend to code for what looks good in IE but they fail to run the code through Tidy, define a doctype, or check it with a validator. To them if it works in IE it must be okay - not true.

IE will do it's best to render even the crapiest code. Often other companies point this out as THE flaw in IE. But if the developer has actually chosen a doctype IE will render in standards mode instead of quirks mode.

Mozilla (gecko et al) don't have the same kind of "quirks" in their quirks mode so pages that get shoehorned one way under IE will not necessarily be mangled the same way in Mozilla. It's not really a fault of the browsers per se but the developers of the sites not creating better code or worse letting programs vomit out code for them.

It's difficult to compare apples to apples in the case of Mozilla/gecko rendering engine and IE, as IE gets partially preloaded with the OS and gecko does not (I believe it can be but...) how do you really determine resource usage when components of the app are split up like that.

In my experience, it's often not bad HTML that messes up the rendering on Mozilla, it's perfectly valid HTML that the Mozilla camp for some unknown reason chooses to render differently from IE.

One case of this especially annoys me: Mozilla interprets CSS font sizes differently from IE (and no, it's not the other way around because IE had CSS support first). It's darn near impossible to make the font sizes consistent between the browsers without resorting to deprecated <font size> tags or using fixed point sizes (which creates accessibility problems).

As far as resource usage of the rendering engine, it's easy to see the difference if you load a large HTML file, such as the 6 MB PHP documentation page. Take a look at Task Manager after loading the page, and notice that the IE process is consuming ~50 MB of memory, while the Mozilla process is consuming ~100 MB. 50 additional MB of memory usage cannot be attributed to its components not being preloaded; it is due to less efficient use of memory by its rendering engine.

It's also interesting to note that the same page takes 2 seconds to load completely in IE on my machine, but a whopping 15 seconds to load on Mozilla. Really goes to show how contrived those claims of Mozilla's speed prowess were. For complex CSS rendering it may be good, but for basic text layout it's embarrassingly bad...

Actually I code pages daily and have not noticed any difference in font sizing using CSS between IE5/6/moz/opera. I'm working on an example page right now to show another developer how to stop using the FONT tag and start using CSS and flipping back and forth between the pages in IE6/Mozilla I see no difference at all.

Again I would point to bad coding if this is what you are seeing. If you could be more specific with an example I might be able to be more specific in how the example is flawed or may, in seeing the example, admit I'm wrong. Either I have no problem doing.

Similarly I can't dispute the memory hog claim because I cannot find the 6 mb PHP documentation (from PHP.net I only found a couple of files over 250k).
I can say that right now looking at my test document which is about 30k IE is at 14mb and mozilla is at 17mb of memory (both freshly loaded and the test page the first thing to go in). So yes Mozilla is 17% higher memory consumption for this small document if you ONLY look at the Mozilla/IE executables. I think loading a 6mb html file would not be the norm loading a 30k document probably would be the norm and in loading that 30k I see no noticeable difference in speed between the two (but I am working locally on a decent machine, mileage may vary).

I think part of what may contribute to Mozilla's increased footprint is the UI and depending on whether you are using a beta version (which contains debugging) or a release version. The UI issue I think has been remedied in the latest version of Firebird, although I don't have it up to test right now.

Like you, I design pages (almost) daily and--since adopting standards--I have had little problems with compatibility among browsers.

(One font size problem occurs if you use tables in a "quirks"-rendered document, in which case the table cells do not inherit the font size of the body. This is to emulate bugs in the behavior of older browsers. Putting a doctype that will render the page in standards mode will fix this--as will not using tables except for tabular data...)

In my latest adventure, I redesigned my main website page in HTML 4.01 Strict with CSS. It looks almost exactly like the old one, yet I changed a bunch of markup. I believe my HTML is now structurally correct (I have a "breadcrumb" trail marked up as a nested list and a "common links" bar below it also marked up as a list; CSS modifies them to display inline, and if IE supported CSS elements of generation, I could be even more structurally correct.)

However, it displays great in Opera 7 (even Opera 5!), Internet Explorer 6 (will test older versions), and Netscape 7. Navigator 4, of course, mangles it, but I didn't need to tell you that, did I? :) Luckily, it doesn't understand @import, so I'll have to use that to hide some style from it. And older browsers? They'll display content just fine, but won't get the formatting if they can't handle it. A Lynx user doesn't need formatting, so there's no sense in feeding it to him embedded in the HTML. Navigator 3 can't understand CSS, and so it'll also just get content; anyone still using that browser is probably more after content, anyway.

I believe that coding to standards will enable one to develop sites that display in any browser. (These days, you might still encounter a few bugs you'll have to work around.) It is, after all, not 1998 anymore. :)

One example I had in mind was:

<span style="font-size: small">
hello
</span>

Mozilla renders this smaller than IE.

But after reading RobertM's comment, I now see that putting in an "HTML 4.01 Strict" <!DOCTYPE> tag (and not "Transitional") makes IE's text as small as Mozilla's. Go figure. With this knowledge, maybe I can finally make the move to CSS. :)

BTW, the PHP documentation page in question can be found at
http://www.php.net/distr.../php_manual_en.html.bz2

"'Maybe it is because the Netscape technology sucks,' Microsoft's Cullinan previously remarked." This doesn't even sound like something any company who *wants* to be respected would say. As far as page rendering is concerned--which is the main thing it would do in AOL--I think it's *better* than IE. (Go ahead, just try to use advanced CSS 2 or the standard DOM in IE...)

That aside, I think this article is nicely written. Definitely better than the CNET one, which "predicts" the demise of Netscape as a result of this action. Speaking of which, does AOL's new IE license mean that they *can* use IE or that they *must*? In previous agreements I thought it was "must," but this one sounds like "can."

Cullinan said that based on the reviews of Netscape 6.x, so he did have a point. But I agree, these days Gecko is great and I use it as my primary browser.

I believe the license allows them to use it, not necessarily a "must include." But AOL's Parsons made clear in today's conference call that AOL has no intentions of ever switching because everything works perfectly right now as it is.

I see--it was a Netscape 6-era quote. (In that case, it's hard to tell how a page displays when the browser won't start or crashes when you try to do anything...) :-)

As far as the IE license, I really didn't think you needed one to use it. Since it's just an ActiveX control Microsoft intentionally made available for use to programs in any development language that supports it (I'm guessing AOL is written mostly in Visual C++) and since AOL has been using it for the last couple years despite their freedom in this regard ... I was wondering if this agreement was a "can" or a "must," although--like you--I think their statements make it seem like they'll use it anyway no matter what the wording. :)

In regard to the "working perfect with IE" statements, I was part of the AOL with Gecko tests, and it worked very well for me--and that was even before I was a Netscape 6.-anything-but-0 convert. But in other ways, I could see why they'd be hesitant (especially when so much of their content, some primarily from third parties, is going Web-based).

I guess this means we'll have to wait until 2010 to see anything exciting in this regard, unless you count AOL for OS X. :) (Which *looks* good, but from what I've heard from people who know people who use it--I don't know any--is that it's a bit unstable.)

It's nice that AOL gets to use IE for another seven years. It will certainly make things easier for them, but what will they be using seven years from now? They should probably keep that Netscape thing in their back pocket.

And I’m surprised to see MS and AOL getting along so well, now.

That doesn't matter, seeing where as nearly every page you load will still look better in IE.

Of course, I choose MozillaFirebird for my browser, simply because I really love having tabbed browsing, pop-up blocking, and my little google-bar.

Huh? I rarely if ever see a difference in page rendering between the two anymore.

Could you give a link to your Cullinan quote, I don't seem to be able to find anything even remotely similar available through a search.

His quote is in the article above. (You can Edit | Find for it, or in Mozilla, use Type-Ahead Find.) :-)

found it http://www.betanews.com/print.php3?sid=1016168410... So in the summer/fall of 2001 project manager Cullinan, in what sounds like a candid conversation vs. interview for the media says, based on what he's heard from the reviews that Netscapes technology sucks. I assumed you were quoting something more recent. It sounded harsher than it really was and for that time I would also have to agree with that statement. Today I feel differently just took a hell of a long time to get here.

The quote was originally referenced in this article as well, but was removed after confusion over when it was stated.

do u think aol will ever merge icq and aim networks together....