I really, really hate what Gigabyte does with their BIOSes, considering their BIOS backed itself up on the end on some of my disks, changed the OS-visible size of the disk using Host Protected Area (HPA), squashing the mdraid metadata that was happily living there.

By the time I understood what was happening, I had had 3 of my 6 RAID disks screwed, as I had swapped the disks around ignorantly thinking it was some controller error.

That feature was not advertised, and that version of the BIOS had a bug where this feature didn't properly detect which disks it could accomplish this on (it only looked for NTFS/VFAT partitions, natch) and could not be disabled. While I can understand the purpose and usefulness of the feature, releasing with such a bug has made me swear off Gigabyte.

For the reference, it was a GA-P35-DS3, with BIOS F12.

Marlinspike's approach, implemented in a Firefox extension presented at DefCon '11, is to do away with the notion of CAs altogether in SSL, replacing it with a distributed network that reports on the certificate they see. Basically, if the certificate you see agrees with the rest of the network, then you're not being spoofed.

He had previously explained the properties a replacement to the CA system had to demonstrate in order to be viable

Moxie Marlinspike, the author of Convergence mentioned in TFA, addressed that very problem in a post. Long story short: a DNSSEC system would worsen the rigidity and centralization of the current CA system.

People are confusing this master key that breaks HDCP, saying it can help decrypt Blu-Ray discs. That's not the case: Blu-Ray is encrypted with AACS, which has a similar concept of device keys derived by a master key. AACS has a mechanism of revoking compromised device keys. Getting the AACS master key would bypass that mechanism, and would be great news.

This key isn't the AACS master key This is an HDCP key, which would allow one to create a "unauthorized" device that can connect to HDCP-encrypted HDMI and succesfully decrypt the HD stream.

HDCP has been known to be nearly broken since 2001, in that obtaining the device keys of 40-50 devices is enough to calculate the master key.

Huh, WDE has to give away their HDTVs? Where do I sign up?

To clear things up a little, from TFA:

That sounds cool, but the cynic in me believes that won't happen for a variety of reasons ("we don't have any unsold products! The last ones went on eBay for 1$ each!")

I'd like to preemptively buy a "Free Assange" wikileaks t-shirt. It doesn't exist yet, but I figure it's only a matter of time before it's necessary.

I suggest that from now on, articles about far-out piracy number thrown out by special-interest should include a link to Wolfire's excellent analysis of video game piracy. Choice quote:

It does not, however, say that Gibson approved it.

Considering the distances involved, I found it funny that the sentence implied simultaneity. Voyager 2 is about 92 AU out (according to WP), which is 12 light-hours and 45 light-minutes. So if they send the signal in the morning, the memory will be reset in the afternoon, and they can hope for clean signals the day after.

By and large, government just means worthless expense, and pointless obstruction.

So how's that broadband market going for you? How about health care?

(and that's just off the top of my head.)

this post written from Europe.

I've always envied their ability to maintain control over their brand and use it for appropriately portioned good and evil ;)

From what I've understood from their past posts and various interviews, they owe a lot to their business manager, Robert Khoo.

They like explaining how they actually sold their brand and rights away before Khoo came aboard. Luckily, the buyer disappeared into bankruptcy and nobody else has since claimed that ownership. I'm not too sure of the details and current veracity of this :)

picture it: In the living room, the Killer (SCO) is lying on the ground, dead. The Hero (Novell) can finally relax and embrace his Girl (the OSS community). But no, shocking twist! The Killer rises behind them, brandishes a knife, and...

the hardware support for h.264 is really just a programmable DSP in most case

Yeah, no. I work as an embedded SW developer on the U8500 chip. Which codecs we have to support is factored in very early in the architectural phase, specifically because we then decide what HW acceleration is cost and power-effective. Efficient decoding from a power perspective is vital in the mobile market, of course. A DSP is better than a general-purpose CPU but much worse than dedicated HW. Also, you can't hold the expected performance (1080p at 60fps!) in our frequencies without HW.

My company originally cornered the MPEG2 market by being the first to provide a hw decoding chip. Newer codecs are... more complicated.

I'm probably already saying too much. Just a second, there are a couple of guys in black suits at the door, let me s:JKSDG*E^N#NB-- NO CARRIER

"A multitude of browsers will make the web more standardised and easier to browse".

Hah, that made me laugh. As much as it pains me to say it, Microsoft Windows standardized the desktop, and Internet Explorer the Web. Sure, the quality of that 'standard' was terrible, but at least it was a standard.

Every week I see cool new features demonstrated. But they're all tied to disclaimers such as Demo works best in Safari 4.x and pretty well in Firefox 3.5. and use css properties like "-webkit-text-stroke". That is the opposite of a standard.

I hope browser diversity will pull IE kicking and screaming into HTML5, but I certainly don't expect standardization!