It now shows two lists: one for updates, and one for new software. This takes care of one of the three easy steps that I culled from discussions back in March:

1. Separate updates from new software and label them clearly. Done.
2. Leave the new stuff unchecked by default. Bzzzt! Try again!
3. When run automatically, don’t pop up a notice more than once for each piece of not-installed software. [Edit:] Done.

Unfortunately the new software is still checked by default, but one hopes that the separate list would be enough to make people stop, look, and make a conscious choice as to whether or not to install it.

I don’t know yet how it handles new software when run automatically, or whether they’ve made the ignore option apply to an entire piece of software rather than a specific installer. I’ve taken iTunes off the ignore list and set it to check daily so that I can find out. [Edit:] I haven’t seen it pop up in the last 24 hours, and according to eWeek, “Apple will now only prompt the user if there are critical security updates available.”

Apple Software Update on a system with up-to-date Safari and QuickTime, but no iTunes.

At work, I have QuickTime and Safari for development purposes. I don’t have iTunes. I don’t need it. I don’t even have speakers hooked up to the computer. But every time a new version gets released, it shows up in the Apple Software Update list, and I have to tell it to ignore it until the next time they update iTunes.

Now that Safari for Windows is out of beta, it’s doing the same with Safari*. And people are complaining. People like John Lilly, CEO of Mozilla, who sees it as an anti-competitive measure that dilutes users’ trust in software updaters.

Personally, I think there is a problem, but I hardly expected it to turn into the firestorm it has, with Asa Dotzler, c|net, digg, Techmeme, [edit] and now Slashdot, [edit 2] Daring Fireball and Wired (it just keeps going!), and dozens hundreds of commenters entering the fray.

There’s a simple solution, and it’s one of those rare cases where Microsoft gets something right in their software that Apple gets wrong.

1. Create a separate section for software that isn’t already installed, and label it clearly. It can be in the same list, as long as there’s a separation and a heading.
2. Leave the new stuff unchecked by default.
3. Added: If set to check automatically, don’t pop up a notice more than once for each piece of not-installed software.

That’s it. Done. Apple still gets to leverage their installer to make people aware of their other apps, but there’s no chance of someone accidentally installing Safari (or iTunes) by accident because they didn’t read the list too closely. Take a look at Microsoft Update and how they (currently) offer Silverlight. It’s in a list of optional software, and it’s not checked until you choose it.

That’s all this really comes down to: sensible defaults and proper labeling.

*I have to admit getting a kick out of the title, “Apple pushes Safari on Windows via iTunes updater,” because my problem is that they’re pushing iTunes on Windows via their Safari updater. It’s a matter of perspective.

Rows 4-5 test fallback behavior for objects. The idea is that if a page tries to load an external resource, but can’t—the file is missing, the server’s down, the network’s slow, the browser doesn’t have the right plugin, etc.—the page can provide alternate content. And it can be nested, so you can try, say, a video clip that falls back to an SVG image that falls back to a PNG that falls back to text. The Acid2 guide goes into more detail, with the relevant section of code being this:

<object data="data:application/x-unknown,ERROR">
  <object data="http://www.webstandards.org/404/" type="text/html">
    <object type="image/png" data="data:image/png;...">ERROR</object>
  </object>
</object>

So it tries to load the first object, which is deliberately made unloadable. Then it tries to load the second object, which calls out to a webpage which is supposed to be unavailable. Then it falls back to the third object, which is an embedded image of the eyes.

The problems have been with the second object, the one hosted outside of the test file itself.

Back in December, when Microsoft announced that their internal builds of IE8 passed Acid2, lots of people started checking it in the Firefox 3 beta, Safari, and Opera, browsers that were known to pass. And they were surprised to find it didn’t work. It turned out that the server config on www.webstandards.org had changed such that the file requested actually returned a page instead of an error, so browsers were properly loading that page instead of the eyes. It was suggested that people use an alternate copy of the test that pointed to a different external resource.

Then when IE8 beta1 came out, people rushed to try it themselves. By this time the server config had been fixed, so the official copy of the test worked again. But people trying it on alternate copies ran into a problem, because it tripped a cross-site security check (IE8b1 would only load objects from the same domain) and IE8 wasn’t using the fallback content if it was blocked for security reasons. (This makes absolutely no sense. It’s like refusing to let someone board a plane if they trip a metal detector, instead of checking to make sure they didn’t miss a few coins and having them walk through again. Add to that the fact that IE is perfectly happy to load images, iframes, Flash animations, etc. from other sites, and the restriction itself seems a bit silly.)

So now Safari 3.1 is out, and has problems with exactly the same line. In my test on Windows, it displays this dithered orange band. In my test on Mac OS, it looks like this for a few seconds, but continues to display the page-loading icon. Then it finishes loading and displays the eyes, passing the test. It looks like it’s just taking a couple of seconds to check that external resource before falling back to the alternative.

I just find it interesting that all of these come down to one single piece of the test, and it’s the piece that tries to load an external resource—something that isn’t within its own control.

Oddly enough, there doesn’t seem to be much chatter from the browser community about it, at least not on sites I follow from work. There may be 25 posts on my RSS reader at home, for all I know.

I wish Apple would make the release notes easier to find. I clicked on the “more info” link in Software Update at home, but didn’t have time to really read it. I wanted to check the list at work, but there’s no menu item, it’s not visible on Apple’s website, and their search engine hasn’t indexed it yet. I had to search Google, and found it from some random person’s Twitter post. (Oh, and Apple? As long as I’m giving you advice, you’re running your site on Apache. Apache has a feature called mod_speling [sic] that will automatically correct a single-error typo when someone hits your site. I highly recommend that you look into it instead of handing out a 404 error whenever someone’s finger slips.)

User interface seems mostly the same as 3.0.

Not sure if it’s new or I just never noticed it, but the history menu has an option to reopen all windows from the previous session. It isn’t the automatic recovery offered by Firefox or Opera, but it’s the next best thing—and quite handy for cases when, for instance, Norton Antivirus has just updated itself and popped up a “will reboot in X seconds” warning, which you didn’t see because you had too many windows open. *ahem*

I believe this is the first browser released that supports embedding TrueType fonts. (IE has been able to embed fonts for years, but you had to convert them first, which may be why you don’t see too many these days.) When WebKit first added the feature last fall, I tested it out on my Les Mis page.

I really like the new developer tools (Prefs→Advanced→Show Develop menu), especially the network timeline. This, combined with YSlow on Firefox (itself an extension to Firebug), will be extremely useful for optimizing site performance.

It gets 77/100 on the Acid3 test, much better than Safari 3.0, which only scored 39/100. WebKit looks like it’s on track to be the first engine to pass again, having hit 93/100 yesterday. Oddly enough, the Acid2 regression is still present on XP (need to compare to the Mac version it displays correctly on the Mac), with an orange band covering the eyes and the border to the right of that band red instead of black.

Another odd thing: when it’s really busy, it seems to revert to a standard window frame instead of its own skin.

Who wants to bet that .Mac will be one of the first webapps to really make use of offline storage?

The mug is from the short-lived Mozilla Coffee. It seemed appropriate. Now if I can just track down some Opera Coffee, or Explorer Coffee…

*Safari 3 was included in Mac OS X 10.5 Leopard, released 2½ weeks ago. And it’s included in the 10.4.11 update for Tiger, released today. An updated version of Safari was released today for Windows, but it’s still a beta, according to Apple’s website and the license (even though the about box just says it’s Safari 3.0.4—the same version that’s in Leopard). I’d been planning to hold this until all 3 releases were out, but clearly they don’t feel that the Windows version is quite release-quality yet. So, on the premise that two out of three ain’t bad, I’m posting.

Apple: Finally pre-ordered Mac OS X Leopard, removing the temptation to run out to an Apple store or Fry’s this weekend (though I’ve been meaning to put some more RAM in the Windows box). Saved a few bucks by ordering from Amazon ($10 off the family pack, would’ve been $20 off the standard box), and picked the free shipping so that I won’t be tempted to install it until there’ve been a few days’ worth of bug reports.

Meanwhile, I’m wondering when Safari 3 comes out for Windows and Tiger. Tonight at 6:00? Monday? I’m looking forward to this putting some of the new CSS3 capabilities into the hands of potentially 5% of the web audience.

Opera: Speaking of web browsers, Opera 9.5 beta came out yesterday. In addition to lots of work on rendering & site compatibility (as seen through the last few weeks’ worth of alpha releases), they’ve launched a new service called Opera Link. It’s primarily a bookmarks sync service, plus a web-accessible interface. So you can automatically sync multiple copies of Opera—including Opera Mini—and also be able to access those bookmarks from Firefox, IE, or a computer where you’re a guest (friend, computer lab, cafe, etc.). I think the biggest impact here is going to be syncing between the desktop and phone, like Safari on the desktop and the iPhone.

On the other hand, imagine adding a bookmarklet or Firefox extension to more easily update from—or even fully sync with—other browsers. Or better yet, a way to synchronize Opera Link with, say, del.icio.us, which can integrate fully with both Firefox (via an extension) and Flock.

Spam: I’m astonished that, with the amount of comment spam that hits this blog (many thanks to Bad Behavior and Spam Karma for helping stem the tide!), I’ve only netted 7 comment spammers for Project Honeypot since they started tracking comment spam 6 months ago. I guess the software is smart enough to only hit the real forms?

Wordpress: Just released version 2.3.1 with a bunch of bugfixes and (of course) a security fix. Updated.

A bit closer to home, this site is currently seeing 64.6% IE, 26.2% Firefox, 4.4% Safari, 1.2% Opera (which probably includes both the desktop and mini versions). Splitting IE into versions, we’ve got 35.9% IE6 and 28% IE7. We’re already at the point where IE6 users are a minority (albeit the largest one), and more than 50% of visitors are using something more modern.

I’m looking forward to the next 2 milestones: IE7 overtaking IE6, and Firefox overtaking IE6. Come to think of it, I’d really like to get rid of IE6. Its time has passed, and the web will be better off without it, just as it’s better off without Netscape 4.

del.icio.us helps somewhat, especially since I discovered I can add it as a search in both Firefox and Opera, but web apps have a certain amount of delay that doesn’t work for the most frequently-accessed sites. And I don’t want to add yet another web app, I want to sync the bookmarks in each browser.

Most of the solutions I’ve found (.Mac, Google Browser Sync, Opera 9.5, various Firefox extensions) are geared toward syncing two or more copies of the same browser on different computers. What I want is to bookmark a site in Firefox on one computer, and have it show up in Safari on another.

Any suggestions?

Increased consumer choice, of course, is a good thing. The most immediate benefit, though, is that Windows-based web developers (the majority) who haven’t been willing to buy a Mac to test their sites in Safari will be able to do full testing on all four major rendering engines: Trident (IE), Gecko (Mozilla/Firefox/etc.), Webkit (Safari) and Presto (Opera).

Also, there’s some really cool stuff available in recent versions of WebKit that will be great to have available for a wider audience.

Interesting thought: this may be the first browser released since Opera expanded to Linux in ~2000 that is available in the same version on Windows and Mac, but not Linux. Even when Internet Explorer was available for the Mac, it used a different engine than the Windows version did.

I wonder what impact this will have on the development of Swift. Its main claim to fame was porting WebKit to Windows, and it’s been months since their last release.

I also wonder what the status is on re-merging the KHTML and WebKit forks. It’s gotten to the point that Konquerer is only an approximation of Safari, making testing on Linux a little harder than it used to be.

(via Asa Dotzler)

No doubt there’s a 500-comment Slashdot discussion already.

Update: Slashdot’s all over it, and Opera Watch has a thread going as well.

Update 2: I’ve posted my thoughts on the implications for Opera. There’s an update at CSS3.info, where they have previews of upcoming CSS features available in Safari 3.

Update 3: I’ve updated the Alternative Browser Alliance to reflect Safari’s new status. This also solves a nagging doubt I’ve had as to whether the default browser on Mac OS should really be considered “alternative.” On Windows, it definitely is.

Update 4: The Webkit team and Web Standards Project have weighed in. The Windows version of WebKit should be available later today, which will be nice for following progress on issues as it moves from beta toward final version. It turns out there’s a regression and at least the Windows version no longer renders the Acid2 test correctly.

Update 5: The author of Swift says that Swift isn’t going away [edit: the blog has since vanished], and points out that “Swift renders more like a Windows Application, both in the GUI and in WebKit. Safari, looks just like OS X, similar to iTunes 6 and below.” Ever since Apple started porting apps to Windows, I’ve found something odd: A common complaint about third-party Mac software is that it doesn’t look and feel native (one of the big reasons we have Camino as well as Firefox), yet when Apple ports their own apps to Windows, it makes them look exactly the same as they do on Mac OS instead of making them work like native apps. I mentioned this to Katie yesterday and she suggested it might be a case of turnabout being fair play.

There’s a tip floating around to speed up the Safari web browser by changing a hidden setting, “page load delay.” There are testimonials by people who are really impressed with how much faster Safari is after making this change. Only one problem: The setting doesn’t exist anymore in current versions of Safari (1.3 or later), so changing it has no effect.

The author of the shareware tool in question responded, saying that he honestly had no idea that the setting had been removed, and offering a refund to anyone who wanted their money back. And there are a couple of other optimizations it can make.

There are some things that the human mind just isn’t good at measuring objectively, and perception of time depends very much on circumstance. “Time flies when you’re having fun” and “A watched pot never boils” have been known for ages.

While there is an “end of an era” feeling to this, it’s kind of like losing the last veteran of World War I. It’s of more historical significance than anything else. When Microsoft released IE5/Mac, it was hailed as the most standards-compliant web browser available. But Microsoft abandoned it years ago.

Fortunately, not only is Safari a worthy successor, but there are other options as well. What’s great about the web browser field these days is that the major players are constantly improving their offerings and working toward greater compatibility. And soon any website that wants to cater to Mac users will no longer be able to fall back on “Just use IE!” They’ll have to test in Safari, and of course the easiest way to build a website that works in IE/Win, Safari, and Firefox (the two defaults and the major alternative) is to start with standards-based code in the first place—which improves compatibility with even more browsers. Users get more choices, and websites get more users. Everyone wins.

So who’s next? Well, Opera 9 beta 1 is very close—there’s a pair of red squares that should be black, but that’s it. Neither IE7 nor Firefox 1.5 will have much in the way of Acid2-related fixes, though the trunk builds of Firefox show improvement, so 2.0 has a chance 3.0 might make it will pass (since 2.0 will use the same engine as 1.5).

I realized that the detail page pulls out Mac OS X, which makes up 2.8%…but MSIE doesn’t say whether it’s running on Classic or OS X. Fortunately IE 5.2 is OS X-only, so we can add in that 0.6%, leaving us with an estimated 3.4% on OS X and 1% on Classic.

So, to the extent that these stats are reliable…

• Nearly one fourth of Mac users visiting this site are still running an obsolete version of the OS.
• 65% of Mac OS X users are using Safari, with only 20% on Internet Explorer.

Anything more detailed is going to require going through the logs myself or writing my own stats script, so I have no idea how the remaining 15% breaks down.

*All of hyperborea.org.