The IEBlog recently posted about their efforts to improve reliability in Internet Explorer 8, particularly the idea of “loosely-coupled IE” (or LCIE). The short explanation is that each tab runs in its own process, so if a web page causes the browser to crash, only that tab crashes — not the whole thing. (It is a bit more complicated, but that’s the principle.) Combine that with session recovery (load with the same set of web pages, if possible with the form data you hadn’t quite finished typing in), and you massively reduce the pain of browser crashes.

I’d like to see something like this picked up by Firefox and Opera as well. They both have crash recovery already, but it still means restoring the entire session. If you have 20 tabs open, it’s great that you don’t have to hunt them down again. But it also means you have to wait for 20 pages to load simultaneously. It would be much nicer to only have to wait for one (or, if I read the IE8 article correctly, three).

Edited to add:

On a related note, I’ve run into an interesting conflict between crash recovery and WordPress’ auto-save feature. If you start a new post, WordPress will automatically save it as a draft. If the browser crashes, it will bring up the new-post page, but restore most of the form data you filled in. So the title, the text of your post, etc will all be there. But WordPress will see it as a new post, and you’ll end up with a duplicate.

This wasn’t a major problem when I encountered it — I had to reset the categories, tags, and post slug after I hit publish (since I hadn’t noticed that they’d been reset to defaults), and I just deleted the older, partial version of the post — but I can imagine if I’d uploaded an image gallery, I would have been rather annoyed, since there’s no way (that I’ve noticed) to move images from one post to another. Reuse them, sure, but not such that the gallery feature would work.

Internet ExplorerInstalled the first Internet Explorer 8 beta. Some thoughts:

I’m impressed that it can import settings from Firefox & Safari.

It detected Firefox extensions and even offered to look up similar add-ons. Unfortunately it was a big long search string with all the titles, and therefore a useless list of results for things like cameras (yeah, how am I supposed to install a $1000 Nikon D80 on my web browser?) and the hint book for Splinter Cell.

Activities: My first thought was, “hey, they’re doing stuff with microformats!” Which is key to the underlying support (recognizing types of data and only offering relevant services, like maps for locations but not for book titles). But on the face of it, it’s a lot more like the way Flock integrates with various web services: Set up your blogging provider, and you can easily send stuff to your blog. Though right now they mostly have Microsoft-hosted services.

“Emulate IE7” appears to involve restarting in an alternate mode right now. I assume automatic switching is something planned for later betas.

Other than that, the UI seems about the same as IE7 so far.

It does indeed pass Acid2 (assuming the page isn’t swamped when you try to load it).

So, how else does its rendering differ?

Minor visual glitch: I have CSS-based banners on some pages (W3C validation, for instance), using spans with borders. If it’s on the last line of a page, IE will cut off the bottom border, because it extends past the end of the page. Other browsers show it. I’ve gotten around this in the past by adding a blank paragraph afterward, but now IE8 collapses the empty paragraph. That’s probably the correct thing to do, but it does mean adjusting things a bit. Not a big problem, though, because I’ve just noticed that it handles other pages fine, without the <p></p> workaround, which means that I’m probably already using a better solution elsewhere.

Several cases of re-styling UL lists seem to confuse it. The tabs running across the top of my Flash page, for instance, or the sidebar on the Alternative Browser Alliance. Others appear just as they do in other browsers (including IE7). This will bear investigation. (Edit: 2 different problems; see below.)

Still no sign of generated content. Beta 2? Please? Edit: according to, it does support generated content, but images don’t work (yet?). I’d been using this, progressive-enhancement–style, to add icons for outgoing links on my Flash site. It works in, well, everything else current.

Additionally: I’m surprised to see it so early, and to see it as a public beta and not something that required an MSDN login. And they had the sense to release a version for Windows XP! I was half-expecting it to be a Vista-only release, which would’ve been seriously annoying.

Further updates will be added below as I think of them.

It turns out the problem on the Alternative Browser Alliance menu wasn’t related to lists as I’d thought, but to a change in the CSS parser. For whatever reason, IE8b1 is susceptible to the Caio Hack (/*/*/ place code here /* comment */) normally used to hide CSS rules from Netscape 4. At this stage I should probably be able to remove it and not worry about NS4 anymore. (And it turns out that since I added media types to the link a while back, NS4 doesn’t even read the stylesheet in the first place!)

On the issue with the tabs on the Flash site, it looks like IE8b1 isn’t extending backgrounds beyond the text line on inline elements (oddly, also like NS4). This is probably what’s really going on with the CSS buttons I mentioned above. I’ll have to check which behavior is correct, but my money would be on the Gecko, Opera and WebKit interpretation. If so, this will probably be changed before the final release. If not, I’ll use inline-block instead. Which perhaps I should be doing anyway, except for the annoying fact that Firefox 2 doesn’t support inline-block and Firefox 3, which does, is still in beta.

I’ve reported the Caio Hack issue to Microsoft using their “Report a Webpage Problem” tool. The form emphasizes that you shouldn’t send anything that could identify you, so instead of reporting the problem on one of my own sites, I sent the page describing the hack. This probably means I reported it in the wrong way. 😕

It looks like Activities isn’t actually context-sensitive yet, since it’s offering to show me a map even when I’ve selected random prose instead of an address.

Having messed with it more than I probably should over the last 24 hours, I’ve come to a decision: During beta 1, any rendering problem I encounter in IE8b1 that works the way I want it to in Gecko, Opera, Safari and IE7, I’m going to assume is a bug in beta 1. I’ll try to narrow them down & report them when I have a chance, but I won’t actually change my sites’ code (except for retargeting IE-specific workarounds) until at least beta 2.

Opera.Firefox.I’ve been using the Opera 9.5 previews across the board since September, and the Firefox 3 beta 2 on my secondary work computer for the past month, and I just can’t bring myself to go back. The full-history search available in both browsers has got to be the most useful new feature I’ve seen in a browser since inline spell-check.

Really, the only things holding me back from jumping up to Firefox 3 on my main computers at home and at work were Firebug and some of the HTML validator extensions. Firebug is complicated enough that I didn’t want to rely on the Nightly Tester Tools to disable the compatibility checks. Then I found out that there’s a Firebug beta that does work with Firefox 3. That was enough. Last night I took the plunge.

IE7Meanwhile, things look good on the ditch-IE6 front. After last month’s false alarm due to a local maximum, it looks like IE7 has solidly overtaken IE6 on this site! For the first 13½ days of January, Internet Explorer accounted for 62.5% of total hits. IE7 was 33.5%, and IE6 was only 28.4%. Even better, that’s barely over 1 percentage point from Firefox’s 27.2%!

Most likely, a lot of people got new computers for Christmas. New Windows boxes would mostly be Vista, and would ship with IE7. Another factor might be techies visiting their relatives and helping clean up/update their computers. They might have taken the opportunity to install IE7 or Firefox.

Internet Explorer.Okay, this will mean nothing to most people out there, but to web developers, particularly those who use standards-based design to maximize compatibility with different browsers, this is monumental.

An internal build of Internet Explorer 8 has passed Acid2.

The Acid2 test was released in April 2005 to test a number of pieces of the HTML and CSS standards that, at the time, no modern browser handled according to spec. The purpose of the test was to prod browser developers into improving their products, and to do so consistently, so that developers would have more tools available for cross-browser sites.

At the time, Microsoft dismissed its its importance entirely. Even though they were working on rendering improvements for IE7, they stated that Acid2 was not one of their goals. Meanwhile Opera and Firefox were both in the wrong phase of their development cycles to make sweeping changes, so Safari jumped on it and became the first browser to pass. (Every once in a while I see someone say Opera was the first, and I have to wonder where they were.) Opera followed with version 9, and the Firefox 3 betas pass it as well.

With Gecko (Firefox), WebKit (Safari), Opera and IE accounting for the four biggest web browsers and the most popular minor browsers (Flock, Camino, Shiira, etc., plus IE shells like Maxthon), this shows unprecedented convergence among clients. It will be much easier to develop a cross-browser website that runs on IE8, Firefox 3, Opera 9+ and Safari 3+.

There are, of course, many aspects of the specs that aren’t covered by Acid2. And there are emerging standards like HTML5 and CSS3. And there are plenty of other bugs, quirks, and extensions among various browsers (IE’s bizarre concept of having layout, for instance, trips up all kinds of weird issues). And then there’s waiting for IE8 to be released, and moving people up from IE7, not to mention all the people we still have to move up from IE6. Full benefit is probably at least 3 or 4 years away. *sigh*

(via WaSP Buzz)

Internet Explorer.I know global statistics still show IE7 only taking up 25%–35% of overall Internet Explorer usage, but stats on this site show a slightly different story (usually skewed toward the crowd more likely to install/upgrade a browser). For the first three days of December, I’m seeing more IE7 users than IE6.

Not by a lot. IE7 has 32.7% and IE6 has 30.3% of the total. And I expect it’ll level out or even reverse as stats from a regular work week filter in. But still, something has finally surpassed that moldering, zombified, shambling heap of a web browser.

Next step: getting Firefox’s numbers (currently 26.8%, also above the global levels) over IE6.

Come on, let’s put a stake in this relic. It’s done.

Update (Thursday): And now Microsoft is finally starting to talk about IE8…even if it is just to say they’ve picked out a name. Whee. 😐

As for the stats, the gap has closed somewhat in the last 2 days, with IE7 at 31.6% and IE6 at 31.2%. This is definitely looking like a home/office split. I’m going to have to write a script sometime to do a daily breakdown of browser versions and see if this actually fits.

Update (Saturday): Yes, IE6 has caught up. 32.2% to 31.1%. *sigh* It turns out I was just seeing a local maximum. 🙁