I agree completely, I was just making the point that it is a bit more complicated than X forwarding. As I said, I don't think this should be patentable -- specifically, I think it doesn't go into enough detail on the methods used to ameliorate latency and bandwidth issues.

I don't think this should be patentable but in my mind the implementation of OnLive is doing something a lot more complicated than X forwarding (the idea, of course, is the same).
It's working over a contended public network with a large number of hops with varying latencies, doing low latency compression on HD resolutions at reasonably low bandwidth.

As I understand it, wifi adds more unreliability (packet loss and latency) to the network path as far as they're concerned

I use LastPass which lets me store all my passwords and has an iPhone app for when I'm out and about (and if I'm more trusting, a client-side-decryption web-based system).
The only annoyance I have is when I need to set up an IMAP account (and then I just bring up the password for that account on my phone and type it in). You get all of the functionality (except the iPhone app) free.
It's a much better compromise for me... I keep non-web passwords too (like router passwords for when I need to go down to the datacentre)

While I'll wait until it's gone through peer review, I'm sure this means that news outlets will pick this up and misreport it... it's good PR for HP's research group either way!

I think you'll find he was referencing the contentious area of tabs vs. spaces (and if spaces, how many) as being a perfect way to keep a roomful of programmers busy for hours (reinforced by the second example, "should I avoid braces if I only have a single statement?".

I was using the accidental use of "newline" by the poster as an excuse for a small joke.

Call me traditional but I like to use newlines for newlines :-P/p>

They've open-sourced management code from their Infrastructure-as-a-Service cloud provider (like Amazon EC2) and their blob store (like Amazon S3), meaning you can use this release to build a public/private cloud provider. The IaaS provider code ("OpenStack Compute") is a preview and they say October is their release date.

As an open source cloud provider codebase it joins Eucalyptus, OpenNebula and a few others -- however (and since I've not used the rackspace cloud I don't know whether this is at all accurate) it should be much more stable since it's most of the code running a big cloud provider (whereas Eucalyptus and OpenNebula still tend to be rather flakey in my experience).

What's not entirely clear is how much they're holding back from the release - they say OpenStack Compute is "the same technology that underlies two of the largest and best ones out there". If they push features over to the open source project to keep parity with their own implementation then it may gain (or, in fact, may already have - I've not looked into the code much) features such as energy efficient host cluster management, shutting down VM hosts when they're not required.

There isn't really a major risk at the minute to doing this - the technology behind cloud providers is pretty simple, the cost of running a public cloud is buying and continually refreshing all the hardware and getting plenty of bandwidth and datacentre space to run workloads. Anyone with the money to spend on the hardware either probably doesn't mind spending some developer time building their own API (especially since if their customers use the API directly then having your own API is a good way to discourage customers from moving to another provider)

To maneuver you simply make components of the sail non-reflective, as I understand it... it won't give you a fast turn but as long as we don't want to dogfight in space it should be acceptable

(Hungary, I think?)

We get a much better bang for our buck by using commodity hardware - we can build & maintain a few linux SANs for the price of an 'enterprise' one. We just build software which can deal with the underlying hardware failing or the network splitting.

We do this at work - we chain-load gPXE using PXE and then use that to iSCSI boot from a Linux SAN which uses LVM COW snapshots. It's pretty good - the etherboot project rocks! We've been doing it for a while but it always gives me a kick when I type something at the commandline which wakes up a machine using IPMI & then boots it off some SAN volume

There are enough numbers. Each issuer has 1 trillion numbers and there's about a million possible issuer numbers... there's a useful description of the anatomy of credit card numbers at http://www.merriampark.com/anatomycc.htm

...or maybe the fact that he's no longer involved brings up questions about its future direction. I'm sure they took a look at reiserfs previously

Just chiming in with my experiences in VirtualBox... I couldn't get it to do anything - it just sat there at a black screen when I tried to boot a Linux image. No descriptive errors, no apparent activity.

It sounds like you want VMware Player... it's free and I find it a much more reliable virtualisation system than VirtualBox. It's effectively VMware Server without the ability to disconnect from the VM display (since they're trying to push people doing server virtualisation onto ESXi (and then onto their pricey products as a result).