jcl

Silverlight is Microsoft's answer to Flash: a browser plug-in that supports 'rich' Web applications. The only real-world applications of it that I can remember encountering are the NBC Olympics site (which almost worked sometimes) and a few Microsoft sites. (As I said above, Silverlight is in a sense an evolution of the Java plug-in. One of the improvements is that it's failed in the market immediately instead of slowly withering away over several years as Java applets did.)

It's a very long step. The underlying platform is somewhere between six and fifteen years old. [Edit: At least six if you only count .NET, twelve if you include .NET's immediate ancestors, J++ and COM, fifteen if you include Java. Silverlight is a 'better Java plug-in' the same way .NET (like J++ before it) is a 'better Java'.]

Lazy answer: look over these two threads on proggit. (If you'd like something a bit richer than text, Stanford and MIT have their introductory courses online.)

Avant changelog.

Stop listening to those people. Dreamweaver may well be limited but there's nothing wrong with "Web 1.0". It is incredibly difficult to produce a "Web 2.0" site that works at all, let alone works well, and impossible to produce one that works well across all platforms. Probably best to start with the basics: HTML 4.01 or HTML 5, CSS 2.1, and JavaScript 1.5. On the server... it seems like no two people in the world can agree on a Web stack and I don't see why you and I would be exceptions.

Zoos. Apparently no one bothered to photograph them before capturing them.

It's in Firefox 3.1. AFAIK WebKit is faster than Gecko.

Perhaps. TraceMonkey appears to slightly faster than V8 at the moment. (And, interestingly enough, SquirrelFish seems to be very close.)

XP or Vista. It's possible that it'll support Win9x down the line but I wouldn't hold my breath. Chrome spawns a process for every tab and I don't think Win9x was really designed to accommodate that kind of behavior. (It's not ideal for NT either, but only for performance reasons.)

I doubt it. Even if we could we'd need more information; there are thousands upon thousands of ARM µcs.

Indeed. The breakage was expected, the exact form it took wasn't. I was a bit surprised when mount told me that I had hda3 mounted as / when hda3 didn't exist

IINM all ATA HDDs, parallel or serial, that are supported by libata (the newish unified ATA driver) are mapped as /dev/sd?. Edit: Yeah, I rolled a kernel with IDE disabled and libata enabled and everything moved over to sd?. I think. It was hard to tell with everything broken.

$ echo $LS_COLORS or more readably $ echo $LS_COLORS | sed -e 's/:/\n/g' Directories that writable by people other than owner and group. As for why you'd have them, I have no idea. I didn't recognize the color scheme because I've never (AFAICR) encountered an other-writable directory (that didn't have the sticky-bit set, which changes the colors but is otherwise irrelevant to this discussion). Edit: "chmod o-w $DIR"

Symlinks? If the LS_COLORS environment variable is set, look for... hmm, maybe 94;42. Edit: Debian apparently displays other-writable directories as blue-on-green (ow=34;42) by default. Your blue might just be lighter than mine. (94;42 is light blue-on-green.)

I normally have thirty or forty tabs open in sessions lasting weeks (I only kill my browsers when I reboot for system updates) and I've rarely if ever seen Firefox 3 break 300 MiB.

If your distro packages smartmontools, that would be the place to start.

Godwin's law.

Formatting hda{1,2,3,5,6} shouldn't affect the partition map. Formatting hda4 might eliminate hda{5,6} but I would expect that hda4 would then be unusable because it would be typed as an extended partition in the MBR but the extended boot records in the logical partitions would be missing. The brute force solution would be to change the MBR partition table manually.

That won't delete the partitions. Nuking hda4 might.

Would you like some ridicule and hatred? This fall's MacBook refresh roulette has left me bitter and I'm willing to take it out on strangers.

The embed element isn't in XHTML. Edit: And that document probably needs an XML declaration to be conformant in all contexts: <?xml version="1.0" encoding="ISO-8859-1"?>

Thread neutralized.

It looks like the DPS doesn't believe that repair services are covered after all: I should have known better than to give a journalist the benefit of the doubt.

Heh. Firefox (iceweasel), SeaMonkey (iceape), Opera, Safari, IE, Epiphany, NetFront, links, lynx and will probably reinstall Konqueror eventually. Committed to variety.

Firefox and Opera are both fine. Safari, too. IE sucks but is improving rapidly.