Two friends of mine, one (green) in Durham (until recently in Glasgow), the other (yellow) in Cheshire, heard ice cream vans around 5pm today, when it was raining across most of the UK. The synchronicity was interesting; I find the different reactions say something - I don't know what, I'm not a sociologist. But it made me laugh, at least, and I hope that the posters will forgive me!
Luca Badoer (the driver who most rues the FIA's tendency to abbreviate driver names to the first three characters of the surname, so Lewis Hamilton is labelled HAM, Rubens Barichello is labelled BAR, etc) is now confirmed to be Ferrari's second driver at Spa this weekend.
So - you didn't do very well at the circuit the drivers seem to find a dull place to drive. So we'll put you in the car for what is quite possibly the best drivers' circuit of the season. I'm sure you'll do better than 17/20 there, Luca. Good luck!
I happened to be looking for the old Mosaic web browser today (yes, I need to get a life...) and it seems to be 15 years to the day since its README was updated - 5475 days ago:
The linux-static binary is dated Jul 18 1996 - a mere 13 years ago. Interestingly, apart from DNS resolution, it still works out-of-the-box on Ubuntu 9.04:
I heard about Asda offering this.
Now it seems they've stopped, but 3 other places are offering free delivery - when collecting in store.
Have I missed something here?
I have a machine which contains 4 identical 146Gb internal disks. These are generally detected as sda, sdb, sdc and sdd, as one would expect.
I have been ignoring sdc and sdd, instead using sda and sdb to create a mirror, using md, so that /dev/sda1 and /dev/sdb1 are a mirror, /dev/md1, which contains the root filesystem. Similarly, sda2/sdb2 make md2 for swap.
All very well and good.
Today, I reinstalled the machine (RHEL5.3, using KickStart) and left it to its own devices. I came back, and found that it had booted into the *old* installation - it had installed onto sda/sdb, as requested, but then rebooted with Grub calling the kernel with args "root=/dev/md1"
My hypothesis is that the kernel saw the old md1 device first (for whatever reason - presumably unpredictable on x86 hardware?) and booted that. Maybe enough reboots would eventually get my newly-installed OS; after 2 reboots it didn't manage it, and I would prefer to have some ability to predict what OS a machine will load after a reboot!
Is there a scenario for booting from a *specific* /dev/md1? At this stage, LABEL=/ does not apply, and a bunch of SATA disks (this is a Sun X4450 FWIW) all look the same and apparently have no serial numbers by the time it gets through to the Operating System.
Before md declares that md1=sda1+sdb1, is it possible to define what sda and sdb are?
A recent study of Linux kernel (not userspace) contributions makes for an interesting read.
Greg KH, one of the three coauthors (and a Novell/SuSE developer), pointedly mentioned not so long ago, that Canonical don't give much back to the Linux kernel that they so enjoy the popularity of. I suspect that they give more to projects such as Gnome, than RedHat or IBM do - they have a different focus.
It is interesting to read different reports on these findings.
The Register wrote Linux: More contributors, more code - A lot more than just Linus Torvalds, choosing to emphasise that Torvalds' own contributions have dropped out of some of the statistics, though it is acknowledged that Torvalds' role these days is more that of a coordinator ("benevolant dictator") than code writer.
Computerworld went for Who Writes Linux? - Big Business, putting their emphasis on the fact "Linux isn't written by lonely nerds hiding out in their parents' basements. It's written by people working for major companies."
In other news - most Cheddar Cheese is not made in the Cheddar Gorge; there is only one company in Cheddar making Cheddar Cheese.
I would be very surprised if anybody involved in IT seriously held the view that most (or any?) Linux kernel development work was done in "parents' basements" in the first place.
It is interesting to note how many non-commercial contributions there are, though; 18.2% of changes are from contributors who actively stated that they are not doing the work on behalf of any corporation, and a further 7.6% were "unknown".
After that, RedHat claimed 12.3%, IBM and Novell 7.6% each, Intel 5.3%, "Consultant" 2.5%, and Oracle 2.4%.
That's on the 2.6.11-2.6.30 timescale; from 2.6.24-2.6.30, we see an increase of individual contributors. "None" goes from 18.2% to 21.1%, RedHat drop to 12.0%, IBM and Novell drop to 6.3% and 6.1% respectively; Intel grow to 6.0%, "Unknown" drop to 4.2%, and Oracle grow to 3.1%. The report notes that better research has helped to reduce the amount of "unknown" contributions.
It is interesting to see how much RedHat and Novell (the two major financial beneficiaries of Linux) contribute to the kernel.
There has been controversy over Canonical's (Ubuntu's) lack of contribution, but they are far more focussed on the user experience; it would be strange if they put more Linux Kernel code back than the Debian distribution upon which Ubuntu is based.
It would be interesting to know how many of the "None" and "Unknown" are working on behalf of a particular distribution. Presumably many of them are working on Debian or similar less "official" distributions. Others will be working for the benefit of the community, to scratch their own itch, or to improve their skills and/or CV.
THE UNIX TIME-SHARING SYSTEM  PRESENTS SEVERAL INTERESTING
CAPABILITIES AS AN ARPA NETWORK MINI-HOST. IT OFFERS POWERFUL
COMPILERS, AN EDITOR BASED ON QED, A VERSATILE DOCUMENT
PREPARATION SYSTEM, AND AN EFFICIENT FILE SYSTEM FEATURING
SOPHISTICATED ACCESS CONTROL, MOUNTABLE AND DE-MOUNTABLE
VOLUMES, AND A UNIFIED TREATMENT OF PERIPHERALS AS SPECIAL FILES.
Alan Turing was the greatest computer scientist ever born in Britain. He laid the foundations of computing, helped break the Nazi Enigma code and told us how to tell whether a machine could think.
He was also gay. He was prosecuted for being gay, chemically castrated as a 'cure', and took his own life, aged 41.
The British Government should apologize to Alan Turing for his treatment and recognize that his work created much of the world we live in and saved us from Nazi Germany. And an apology would recognize the tragic consequences of prejudice that ended this man's life and career.
I've signed the petition. For years I've had a firewall called "turing", named after this unspoken genius. One can't help but wonder to what extent his security clearance, his sexuality, and his resulting suicide have lead to the lack of any official acknowledgement regarding his contribution to our success in WWII. The 30-yr-rule is long gone, so it is long overdue that we acknowledge the contribution that Turing's work made to the war effort, and that we acknowledge that the way that he was treated by our government lead to his suicide.
It is suggested that Apple's "bitten-apple" logo (and maybe even their name) is a nod towards Alan Turing.
We should be proud of our achievements, and repent of our failures.
We owe an apology to Alan Turing for the way that he was treated. I hope that an Alan Turing born in the UK today would be better received.
This does irritate me with packaging; this deodorant claims to be "NEW improved formula" at the top, whilst claiming to be the "Original" at the bottom of the packaging.
It is either new or original. It cannot be both, unless this is the first iteration. In this case, it is clearly not. Not that "New improved formula" would make sense for a new-and-original product anyway.
Similarly with "New and Improved" - it's either something new, or it's something old that has been improved. It can't be both.
aka "why don't people think like I think?" or "my petty grievances should matter to everyone else". Ah, this is a blog - those two assumptions are already taken for granted!
McLaren have posted a series of 4 5-minute YouTube posts on Turning MP4-24 into a grand prix winner. A fairly humble set of articles, going over how they have addressed the situation of their initial poor performance for the first half of the season, back up to finally winning their first (possibly only?) race of 2009.
Erm, I think that what Biella is saying is that Richard Stallman had a good idea and implemented it well. I could be wrong, though - there are a lot of very long words in her article!