pieter

From the mailing list:

"We believe that the intention of secure boot is to protect against malicious use or modification of pre-boot code, before the ExitBootServices UEFI service is invoked. Currently, this call is performed by the boot loader, before the kernel is executed.

Therefore, we will only be requiring authentication of boot loader binaries. Ubuntu will not require signed kernel images or kernel modules."

That's completely different from what Fedora is doing (signing all kernels and modules). I hope for them Microsoft agrees with their interpretation and won't revoke their signed binaries. I'm not sure what advantage they would get from a signed boot loader, if you can run any arbitrary kernel from within the loader.

I don't understand the sample budgets -- is that cost per hour? day? week? month? Why don't they specify something as basic as that somewhere?

I think this is a better design than the old one. The old one was nice if you were traveling, but really, how many people did that?

This one fits much nicer in a cupboard next to your AppleTV, and won't take up 2 or 3 sockets in an extension cord. It also makes the wiring easier, since you don't have to make sure your optical audio cable and ethernet cable can actually reach the power sockets.

That would explain why the Ethernet ports are still 100mbit.

ITER won't produce any stable power at all, so they most likely mean something like "A single run of ITER will consume about half a gram of hydrogen while having a peak output of 500MW (for a few seconds).

we're not talking about copyright law here, but patent law.

how could this possibly be faster than a direct connection of the client to your origin server?

You don't have to look in the comments, it says so right in the article:

     The proposal is to change the default to 'upstream'

This device world be a pretty cool server if you could put your own OS on it -- kind of like a much faster Raspberry Pi with built in wif, flash storage and PSU

This post seems to completely miss the point of selectors and how you can match classnames and use pseudo selectors. This isn't a bug in jQuery at all, but just the way css selectors work.

Especially telling is this phrase:

    Here are some examples of invalid selectors in jQuery because they contain invalid characters:
    $( "#person.name" ).addClass( "highlight" ); // FAIL

This is of course a perfectly valid selector, selecting a node that has both the id person and the class name. Saying that the selector is invalid is simply wrong.

Anyone got a mirror of the contents? It requires me to log into Google before I can read the post.

The other option would be directly selling them themselves, instead of being at the mercy of a few existing retailers.

I'm not sure that would go smoother, but at least you wouldn't have to go to each retailer and sign up for their lists or having to check their sites which one will get a new batch first.

(EDIT: I'm not saying I would think that's a good idea, but I think that's what the GP meant)

64-bit small instances are perhaps the best new announcement. We now have two different builds and AMI's for all our different node types, just so that we can use a small instance when necessary and a large instance otherwise. We can now drop half of our images and use 64-bit everywhere.

The medium instances are also very welcome, since previously the gap between small and large was quite big (4x the memory for 4x the price).

It's not a bigger risk than people loading jquery from google, or including google maps or analytics or kissmetric or twitter or Facebook code.

This really sounds too enterprisey. It's unclear to me what the benefit is of using SWF, and lost my interest when it started listing business definitions.

Hey Jérôme, your site's blog really needs some padding on the left when viewing in a small window.

Not true, The Pirate Bay is filtered on Ziggo as of today and will be filtered on XS4All tomorrow. They have appealed the ruling, but have to filter in the meantime.

I like the similarity, but you have to make it more honest:

You're a writer, but don't have a good typewriter. Apple is willing to lend you one for free. The only issue is that if you then want to sell your book, you have to sell it through Apple's store. If you don't sell a single copy, Apple won't charge you for the use of the typewriter. If you sell a lot of copies, Apple gets a cut, both for the book being in their store (publicity) and the use of their typewriter. You're still allowed to use different typewriters, and you can even sell books written on a different typewriter in Apple's store, though they'll still take a cut.

Doesn't sound terribly unreasonable to me

I don't think you understand -- they're only restricting what you can do with the output of the program, not your content itself. You're free to copy/paste it into any other tool and distribute the output of that tool.

Not really; most OSes now enable ipv6 privacy extensions by default, which means a client will connect with a different IP every hour or so.

They probably chose the 6th because it's on a wednesday; they chose June 8th last year, which was also on a wednesday.

Good news, but looks like the site is down currently.

In summary: Facebook, Google, Yahoo and Bing will permanently turn on IPv6 on June 6. Some ISP's also joined, but it's not yet clear what they do at that date -- it may be as little as allowing users to opt-in for IPv6 testing in restricted areas.

Physically it should be possible; you have 32GB micro SD-cards, 32 of which take up about 5.3cm^3. That would be about the same as 1.5x5.0x0.7cm, which seems like something that would fit that knife.

>This silliness is explained by the fact that NASA employees want to work on this project, even if it's going to cost more and deliver less than other teams working on it.

Not at all. It's explained by the fact that congress keeps putting more money in the space launch system than NASA itself requests; moving money to the project that NASA would like to use for its commercial programs.

It's not NASA being silly here – it's congress wanting to impress the world with its own space system.

We tried Sinatra-synchrony, but had some problems with it with latest sinatra and ruby. Eventually I just used eventmachine-synchrony together with Rack::Fiberpool, which works just fine too.

The point with the cable is that extension cables aren't allowed by the USB standard, so Apple created a cable that would only work with their specific device for which they guarantee that it works. I've tried the extension cable with other devices (it works just fine if you insert the plug at an angle), and have had mixed results.

I think Apple did exactly the right thing here and did it to avoid other problems ("I plugged in my iPad through the extension cable to my macbook and now it won't charge").

I don't really care anymore about the design of Google Reader. The only thing I use it for is for syncing my read items between my different reader apps (like Reeder for iPad). As long as they keep supporting that I'm fine, but this seems like a service anyone could implement in case they stop the API access.

I understood it as meaning that you can use it for at least 3 hours, but if they have 'time over' you're allowed to use more. Not sure though.

Interesting idea, but the pricing is just much too complex. By having to choose between monthly/weekly, morning/day/evening and then number of hours there are perhaps 50 different plans available? I'm just too confused to make a choice.

"never" is also "a matter of weeks or less"