saljam

It was only Jack Copeland today. I had to miss that one as well.

The other public (free) lecture is the Penrose lecture on Monday at 20:00. You still have to register online I think.

There's a conference in his honour this weekend in Manchester, where he did much of his work. On the list of speakers are Hoare, Knuth, Penrose, Cerf, Kasparov, Fred Brooks, Rodny Brooks, Shamir, and many others. It's quite exciting to get so many luminaries in the same event.

http://www.turing100.manchester.ac.uk/

Edit: It's also quite expensive!

I don't see why CS can't “take a problem and produce a suitable solution.” Understanding the essential underlying theory (complexity theory, etc.) is essential to solve these problems.

Furthermore, not all Computer Science degrees are purely theoretical course where all you do is esoteric logics and category theory. If anything, I think there's too little theory in the UK's CS courses these days.

I find it odd they call themselves “Wahhabis.” For starters, that term isn't something a “Wahhabi” would call himself. I've never been able to trace when it was first used. However, it's often used by western scholars to refer to “the Saudi guys” when classifying Muslims.

Does anyone know more about this group?

Plan 9 has an extra permission for files; "append." When set, this means programs are only allowed to append text to the end of the file. Both simplifying logging (just open the file and write) and making it more secure (no arbitrary writes.)

I wonder if Lennart and Kay would accept a solution that simple...

For what it's worth, that presentation was by William Kahan. “The Father of Floating Point” and the guy behind the IEEE 754 standard. He also has a Turing award.

How timely! Just this morning I learnt a few anecdotal factoids about Turing's stay in Manchester, and his quarrels with the "old engineering guard" (Williams and Kilburn.) This comes from one of the professors who just retired from the University -- where Turing worked for a few year on the Manchester SSEM, also known as the Baby. It was the first stored program computer - effectively the first Von Neumann computer (http://en.wikipedia.org/wiki/Manchester_Baby http://www.computer50.org/)

The Baby was built by Freddie Williams and Tom Kilburn. They were both engineers who believed they were building a machine on which to do real science. At the time the only way to program the machine was by flipping switches and hacking away in binary machine code. Alan Turing was hired to come to Manchester and provide a sort of higher-level language so that scientists can use the machine, a sort of an assembler if you will.

However, it turned out that Turing didn't mind all this bit twiddling at all! Instead of doing what the rest of the engineers were expecting, he ended up writing all sorts of AI and maths programs to work on problems he thought were interesting. Problems, which the "old guard" thought of as mere games. They were trying to do science, while this guy is just playing! Oh how the world changed... This lead to a great deal of friction between him and the engineers, the remains of which can still be seen today in Manchester. For example, none of the CS buildings were named after him -- the Alan Turing building was completed in 2007 and became the home of Mathematics and Theoretical Physics, but not CS.

Some have suggested that in addition to the above, his sexuality didn't practically help when dealing with the alpha-male 1950s engineering types he was dealing with, but let's not go that way...

Simplicity was my main point. The fact that it was faster just followed. As a matter of fact I haven't rebooted this machine in about 3 weeks.

/run has been in Debian unstable for quite a while now actually, at least 3-4 months. /var/run is just a symlink. I think this is related to the systemd stuff.

Personally, the first thing I do when I install a Linux is write a simple /etc/rc script which mounts filesystems and launches daemons. No /etc/init.d/* start stop stuff, no /run. The result is almost always simpler and faster. My thinkpad takes 3 seconds from bootloader to X11.

That's not what I was proposing. The whole point is that you select who you trust and how much you trust them for any given subject. A random reviewer like the one on Amazon wouldn't have much influence on my rating unless I "trust" them personally, or some one I highly trust in turn highly trusts them.

I'm saying trust too much now.

Or, this could lead to the birth of a decentralised web-of-trust review and recommendation system. Imagine if something like this works: When a person reads a book, he rates it according to his taste. He would also rate how much he "trusts" the opinion of any of his peers. Then, for any book, he could ask the question "How good do my peers think this book is?" or "How good does the faculty of University of Manchester School of Computer Science collectively think this book is?" This then propagates the ratings of this group, and the ratings of the people they "trust", to come up with an answer. The amount of trust can act as some sort of a coefficient at each step.

Now, instead of saying "a group of well-read professionals, picked and chosen by Random House, thought this books was good enough," you can say "a group of people, whom I trust to be well-read professionals, or to have good taste, or to be accurate, or to be knowledgeable in their field, thought this books was good."

The technology is there, and frankly I won't be surprised if someone already wrote something like this.

BTW, this isn't limited to books. You can substitute "book" with academic paper (and you get peer review!), films, music, or pretty much anything provided by a social-web recommendation service,

Even though it's one of my favourites, modern high performance hardware with branch prediction, out-of-order execution, and multiple issue generally makes this trick rather unnecessary. Combine that with the cleverer compilers we have these days and you get pretty much the same performance out of the simple copy.

C is full of surprises, and generally my favourite trick is trick du jour. While it's not strictly for C, Hacker's Delight[1] is my favourite collection of bit twiddling tricks. A highly recommended read if only for the intellectual value.

[1] http://www.hackersdelight.org/

I've tried vbox when I used a mac and couldn't get past the user interface. The gui severely limits what I can easily do with it, and it's just plain irritating sometimes.

Now I'm back on linux and kvm/qemu, and I'm much happier. Why are people using vbox over kvm?

Probably, c.f. bind and union mounts .

Symlinks are bad. They break the natural semantics of a hierarchical tree-like filesystem, and turn it into a messy half-assed graph.

What happens if you move a symlink? Will it still point to the same object? That depends on whether it's relative or absolute. What if you move the directory containing it? Do you have to recursively check for the correctness of links every time you operate on a directory? Sounds unreasonable to me... What if it links to a location which isn't mounted? Whats the definition of '..'? If you cd into a linked directory and do an 'ls ..', should that list the parent of the target or the link? How would you implement that?

The point is, symlinks are a messy kluge which probably wasn't thought out very well. You can look at what Rob Pike has to say about it himself on http://cm.bell-labs.com/sys/doc/lexnames.html

South Africa never got past its apartheid stage until a global boycotting movement forced it to. This isn't just a matter of "politics". This is everyone's responsibility, including scientists.

Start a war?! I think you've got your history mixed up. Palestine was colonized by the British Empire and then given away to the Zionists. Since then Palestinians have had their homes and land brutally confiscated for the incoming Israeli settlers.

http://en.wikipedia.org/wiki/Balfour_Declaration_of_1917

This disappoints me. CERN is one of the most important representatives of the European scientific community. Instead of boycotting the apartheid state which openly practices racial cleansing on a daily basis, CERN embraces it. A truly sad world we live in today if that's what our brightest minds consent.

Something tells me this won't go well on this forum, but I had to say it.

1) It's also the first part of some of the most popular given names. http://en.wikipedia.org/wiki/Abdullah_(name) The word for slave simply doesn't have the same meaning as in English.

2) "Slave" (عبد) means black just as much as "nigger" means the same in English. Or "chink" means Chinese. It's a pejorative term. Anyone decent enough would simply say "black" (اسود) or "tanned" (اسمر).

I'm a native Arabic speaker, and I agree with you on point 10. If you've learned formal Arabic properly - i.e. attended school - then even texts from the pre-islamic era should be accessible. (Mostly poems, see http://en.wikipedia.org/wiki/Muallaqat)

However, there is a definite decline in the number of people who can fluently and eloquently use MLA. Keeping this (sad) trend in mind, you will definitely find native speakers who have trouble understanding not only the old texts, but even the more silver-tongued of the modern ones. This isn't unique to Arabic. Perhaps a parallel can be found in the constant confusion between the possessives and the abbreviated verbs with English pronouns even by native English speakers.

I'm not claiming that formal and eloquent Arabic is going to die any time soon, however the rise of movements such as Masry (http://arz.wikipedia.org/wiki/%D9%88%D9%8A%D9%83%D9%8A%D8%A8...) is a sure tell sign that maybe there aren't as many people speaking it as we thought there are.

And I believe something should be done about this.

This is pretty amazing. Very well done!

It would be fantastic if it had a Gmaps-style zoom feature. A ZUI of a sort.

http://en.wikipedia.org/wiki/Zooming_user_interface

(I added this comment to the HN corkboard above but it got removed very soon afterwards!)

Edit: This was already suggested in http://news.ycombinator.com/item?id=1963075

I use the plan9 from userspace[1] version a lot on both Linux and OS X, almost exclusively actually. Read the man page[2] or even the paper[3]. I recommend giving it some time to get used to, it pays off.

Acme-SAC is built on top of Inferno; the virtual directory structure you're seeing is Inferno's namespace. I find it helps if you know a bit about that first.

[1] http://swtch.com/plan9port/

[2] http://plan9.bell-labs.com/magic/man2html?man=acme&sect=...

[3] http://plan9.bell-labs.com/sys/doc/acme/acme.html