Slashdot Mirror

From this overview, it is possible to conclude that DNS is a poorly specified protocol, but that would be unfair and untrue. DNS was specified loosely, on purpose. This protocol design is a fine example of what M.A. Padlipsky meant by “descriptive rather than prescriptive” in his 1984 thriller, The Elements of Networking Style (Prentice Hall). Functional interoperability and ease of implementation were the goals of the DNS protocol specification, and from the relative ease with which DNS has grown from its petri dish into a world-devouring monster, it’s clear to me that those goals were met. A stronger document set would have eliminated some of the “gotchas” that DNS implementers face, but the essential and intentional looseness of the specification has to be seen as a strength rather than a weakness.

According to this research, ads in video games are more memorable when the game includes a violent component. This makes evolutionary sense, there are good reasons to pay attention during violent incidents. If you survive violence, you should remember what happened so you'll be able to survive next time.

This ability to learn quickly in relationship with violence might be useful in Instructional Design, especially when teaching facts that are inherently boring. For example, imagine a version of Quake which requires you to shoot one specific molecule type while avoiding shooting any of the others to teach organic chemistry. While you're looking for that one molecule type, you have billboards in the game world that teach you the others, which you'll need at higher levels.

Yeah. Physical security loss means hackable. I read this old cold-boot attack paper in Communications of the ACM months ago, and it was old even then. IIRC, one can use a can of compressed air to chill the ram in a computer, remove it, and read its contents in an alternate machine environment designed to save out the contents of the ram. Then you extract security keys from the saved-out-contents, and do whatever you like. I'm pretty sure the acm paper directly referenced retrieving the keys of the microsoft disc encryption software.

Anybody remember connection machine LISP? http://en.wikipedia.org/wiki/Connection_Machine http://portal.acm.org/citation.cfm?id=319870

Well, apparently not ...

...before the Mac existed. Multiple BARCO monitors, driven by DEC GIGI terminals. You'd send drawing commands to them over 9600-baud hard lines, from a VAX 11/780 with a separate process to control each display. It was... primitive. But it was a multi-screen interactive system, at a time when such things were uncommon to say the least.

citation

Network television allows regional broadcasters to insert local advertisements, which tailors them geographically as well. It's even possible to insert advertisements into the background of sports broadcasts: Real time advertisement insertion in baseball video based on advertisement effect. So you can advertise by show and region. Customizing to the actual person is a neat idea though. How about not showing any advertisement to people who are in a bad mood? If they don't look likely to receive the ad well, then just show a generic picture?

It's too much work for applications to have to worry about errors. That's within the scope of the filesystem, because the filesystem's job is to provide reliable access to data for applications. With error-detection and error-correction, you tend to want to catch the problem as soon as possible; that usually means lower in the stack.

ACM has an article on latent disk error rates here. IIRC, Seagate's figures were about one latent error for every 10^15 bits; this figure was confirmed by Sun when they were doing the heavy lifting on ZFS. The important thing is that the latent error rate is not decreasing as disk capacities increase. ZFS specifically addresses this problem using checksum trees.

I think that bit-flips at all are unacceptable. I know that guaranteeing that they can't happen is not possible, but we're willing to spend extra money to ensure that they happen infrequently.

Playing Tetris actually gives you more brain to work with, says a new study to be published later this week.

So you're saying you had control groups of people that played other video games and Tetris showed a difference? Or a control group studying chess? I suspect the title of this article should be "Puzzles Improve Your Brain."

This, says the doctors who undertook the study, shows that focusing on a "challenging visuospatial task" like a videogame can actually alter the structure of the brain, not just increase brain activity.

So you're saying this is akin to jamming the square block in the square hole and the triangle block in the triangle hole? Or, really, any sort of two dimensional puzzles like the mazes on the back of tray mats at a restaurant? Or maybe even -- *gasp* -- any game portrayed on a 2D surface like a TV or computer screen?

The study, funded by Tetris' makers ...

I understand now.

The study's subjects, a group of adolescent girls, underwent MRI scans before and after a three-month Tetris practice period.

The pretty pictures wouldn't happen to be statistically erroneous now would they?

Don't get me wrong, I grew up on Tetris 2 and The New Tetris. They both still have massive replay value and really spurred me to look into polyomino based puzzles which had increased fame in the mid 1960s until everyone realized that they had little real world application (but they still show up in papers). Still, it lead me to a book by Martin Gardner who wrote Scientific American columns on Mathematical Games. If you remember those, I recommend this book. So something good came out of studying tile theory and Tetris for me but there's no evidence yet it did anything more for me than say playing Gauntlet on the NES would have.

FWIW, it isn't required that you need infinitely many line segments to define something to make the parameterization ratio an irrational number...

For example, take a right triangle. If both legs are the same length (say "1"), the length of the hypotenuse is sqrt(2) which is an irrational number w/o a repetitive pattern in the numerical representation. However, if one of the legs is length "3" and the other is length "4", the hypotenuse is of course "5" which is not an irrational number. So we have a case of the length of one line segment giving us both rational and irrational numbers.

However, as it turns out, the converse is true that the straight line is the only curve that has a rational parameterization for arc-length, but that takes a bit more math to prove it...

http://portal.acm.org/citation.cfm?id=1523523.1523896

No, you haven't. The answer to the first question is FDIC. The answer to your second and third questions is the FDA. There's no such regulatory agency for IT.

Yes! And therein lies one of the biggest questions I've had for a long while. Why *don't* we have a professional organization, complete with technical and ethical requirements, as do engineers, cpa's, lawyers, etc.?

ACM has a code of ethics, but I've never heard anyone require ACM membership as a certification issue, nor have I heard of anyone getting de-acm'd for violating the ethics... I think that would be awesome though.

I've read the paper. You can definitely reproduce GFS given enough time from the paper, since all of the synchronization and master/slave node dynamics are described very well. The actual implementation seems like it would be monstrously complex. Not something 2 or 3 guys could pull off in a weekend by any stretch. Anyone who is actually curious, here is the original paper from the acm, though you may need a membership to view it.

I did my thesis in exactly this area. My research assumed that task complexity had something to do with enjoyment, but I was wrong. Experimental evidence showed that people had more when the difficulty of the task made the outcome uncertain relative to their skill level. This jives with researcher Kevin Burns theories of enjoyment being derived from unexpected good news (the unexpected occurrence of positive information gain) - i.e. the punchline of a joke or winning against the perceived odds.

What if the work involves the plumber algorithm?

Why not wait another day before submitting the improvement? All they did now was giving the other team one day to respond, and if they succeed, I doubt they will be able to submit yet another improvement. So why not simply wait until an hour or so before the deadline, or am I missing something about the rules, e.g. any submitted improvements prolong the deadline by one day?

For the grand prize, there was a final 30-day countdown from the time the first entry that achieved greater than 10% was received, which was a month ago. So it seems like this will indeed come down to an ebay-like sniping situation in the last few hours.

I wouldn't feel too sorry for BellKor/KorBell though -- they've got many, many best paper awards at conferences and a huge degree of publicity out of the whole endeavor. In fact, in KDD 2009, they detailed most of the methods that most likely got them to the top -- i.e. they incorporated the fact that tastes and preferences drift over time. Simple, in retrospect of course. If you have an ACM subscription, you can read the 2009 paper here.

Plus, since they work for AT&T/Yahoo Research, I remember Yehuda Koren stating that the money wouldn't have gone to them anyway -- possibly a large bonus, but I think they're entitled to that anyway. So I wouldn't feel too sorry for them.

PAPER

"eyeLook:Âusing attention to facilitate mobile media consumption" http://portal.acm.org/citation.cfm?id=1095034.1095050&coll=GUIDE&dl=GUIDE&CFID=43353856&CFTOKEN=45571461

ABSTRACT:

One of the problems with mobile media devices is that they may distract users during critical everyday tasks, such as navigating the streets of a busy city. We addressed this issue in the design of eyeLook: a platform for attention sensitive mobile computing. eyeLook appliances use embedded low cost eyeCONTACT sensors (ECS) to detect when the user looks at the display. We discuss two eyeLook applications, seeTV and seeTXT, that facilitate courteous media consumption in mobile contexts by using the ECS to respond to user attention. seeTV is an attentive mobile video player that automatically pauses content when the user is not looking. seeTXT is an attentive speed reading application that flashes words on the display, advancing text only when the user is looking. By making mobile media devices sensitive to actual user attention, eyeLook allows applications to gracefully transition users between consuming media, and managing life.

PAPER

"eyeLook:Âusing attention to facilitate mobile media consumption" http://portal.acm.org/citation.cfm?id=1095034.1095050&coll=GUIDE&dl=GUIDE&CFID=43353856&CFTOKEN=45571461

ABSTRACT:

One of the problems with mobile media devices is that they may distract users during critical everyday tasks, such as navigating the streets of a busy city. We addressed this issue in the design of eyeLook: a platform for attention sensitive mobile computing. eyeLook appliances use embedded low cost eyeCONTACT sensors (ECS) to detect when the user looks at the display. We discuss two eyeLook applications, seeTV and seeTXT, that facilitate courteous media consumption in mobile contexts by using the ECS to respond to user attention. seeTV is an attentive mobile video player that automatically pauses content when the user is not looking. seeTXT is an attentive speed reading application that flashes words on the display, advancing text only when the user is looking. By making mobile media devices sensitive to actual user attention, eyeLook allows applications to gracefully transition users between consuming media, and managing life.

Searching DNA is a very subtle and complex CS problem. This was written by a good friend of mine; DASH: Localising Dynamic Programming for Order of Magnitude Faster, Accurate Sequence Alignment

This was a STOC 2009 paper so for more details than the press release, see Craig Gentry. Fully homomorphic encryption using ideal lattices. In STOC '09: Proceedings of the 41st annual ACM symposium on theory of computing, pp. 169--178. 2009. It's online here for those with access to that.

The abstract for Gentry's article can be found at: http://doi.acm.org/10.1145/1536414.1536440

[Definition: Using or following a link for any purpose is called traversal.] Even though some kinds of link can associate arbitrary numbers of resources, traversal always involves a pair of resources (or portions of them); [Definition: the source from which traversal is begun is the starting resource] and [Definition: the destination is the ending resource]. Note that the term "resource" used in this fashion may at times apply to a resource portion, not a whole resource.
[Definition: Information about how to traverse a pair of resources, including the direction of traversal and possibly application behavior information as well, is called an arc]. If two arcs in a link specify the same pair of resources, but they switch places as starting and ending resources, then the link is multidirectional, which is not the same as merely "going back" after traversing a link.

My links all involve dojo.xhrget, handleAs: json. I use lots of callbacks.

My next reincarnation, should I be happy enough to find someone willing and able to bankroll my journy, would be into the world of LAML (Lisp As a Markup Language). For those with ACM Portal access, and for the rest of us:

ABSTRACT
Functional programming fits well with the use of descriptive markup in HTML and XML. There is also a good fit between S-expressions in Lisp and the XML data set. These similarities are exploited in LAML which is a software package for Scheme. LAML supports exact mirrors of the three variants of XHTML 1.0, SVG 1.0, and a number of more specialized XML languages. The mirrors are all synthesized from document type definitions (DTDs). Each element in a mirror is represented by a named function in Scheme. The mirror functions validate the XML document while it is generated. The validation is based on finite state automata automatically derived from the DTD.

The LAML Hello World:

(load (string-append laml-dir "laml.scm"))
(laml-style "simple-xhtml1.0-transitional-validating")
(write-html â(TM)(raw prolog)

(html 'xmlns "http://www.w3.org/1999/xhtml"
(head (title "Hello World"))
(body (p "Hello" (a â(TM)href "http://www.w3c.org/" "W3C")))))

(end-laml)

Maybe the NVIDIA writer should have written that NVIDIA invented the General Purpose GPU ? From the wiki it seems like they might have been pioneers there.

I still don't think it's a valid claim. Try reading Myer & Sutherland'sOn the Design of Display Processors, or Levinthal and Porter's Chap - a SIMD graphics processor. There was also the Ti graphics chips eg (34010). It happens all the time. IIRC many of the SGI machines were done with programmable hardware but I guess that wasn't exposed to the end user.

You should check out this reference: (Apologies, you need access to the ACM Library to read the article) http://portal.acm.org/citation.cfm?id=299161. Robert Glass discusses this exact issue. The article offers some references to research done using alternative approaches to inspections.

Non-paywall link: http://faculty.fortlewis.edu/ADAMS_E/CLASSES/CS370SWENGRII/WebWinter04/NOTES/Glassinspectionarticle.pdf

You should check out this reference: (Apologies, you need access to the ACM Library to read the article) http://portal.acm.org/citation.cfm?id=299161. Robert Glass discusses this exact issue. The article offers some references to research done using alternative approaches to inspections.

Funny you mention that, I saw an article yesterday which claims to have created an encryption scheme where encrypted data can be modified, written into, queried, and anything "that can be eciently expressed as a circuit" by a person without the decryption key.

If I'm reading the paper correctly, it would mean google could host data, and without having access to the data itself, could still permit user lookups and modifications. Of course that doesn't allay concerns of 3rd party reliability, the encryption scheme is inefficient, and this method may not be robust enough to support the complexity of an sql query, but who knows if it wouldn't be possible in the future.

In any event, you should be pointing to here or (better yet!) here if you want people to actually get to see good ol' EWD's (in)famous letter.

But you should probably be looking at this instead.

In any event, you should be pointing to here or (better yet!) here if you want people to actually get to see good ol' EWD's (in)famous letter.

But you should probably be looking at this instead.

So if you want to know why teachers like Mike would cringe when they saw a GOTO, it is because it works fine IF you know what you are doing and how the code will be processed. But with so many learning code the VS way, with everything drag and drop, for many of those GOTO will simply blow up in their faces. Better just to not let them know it exists in the first place.

See, this argument gets applied to all kinds of bad language features. Really, I think that sentence should read "it works fine IF the code is being maintained by one programmer who never has to deal with anyone else, doesn't mind working for at the same place until he dies and will pass the family trade on to his kids so someone will know what the hell is going on after he's dead." I did an internship a few years ago at a company where the billing software was still essentially all written in QBasic. They were paying two contractors $70/hr each for about 2 full months of work to make minor changes to the software because it was so difficult to maintain. GOTO had its day on the sun, but it's over for a good reason.

Here's a good place to start: http://portal.acm.org/citation.cfm?id=1283933 John Backus' 1977 Turing Award talk.

This might sound like a pun, since I am posting a link to an article that needs a subscription to be read, but it's not it is in the Communications of the ACM, talking about the problems with peer-review committees in the Systems area: Program Committee Overload in Systems (this should be accessible to anyone in a computer science department network. But the moral of the paper is that because of the increasing number of people in the field versus the number of people willing to do peer-review in conferences and journals, less relevant science gets published. They put very good arguments criticizing a lot the way in which science tends to get published nowadays. If you can access it, I think it's a good read about this particular subject.

I'm surprised after reading the spec and blog that a bunch of this seems to be re-inventing Objective-C and bits of concurrent object oriented C. http://portal.acm.org/citation.cfm?id=157352.157356

Not only that, but extensions to a language don't really foster supporting a new paradigm. http://www.codinghorror.com/blog/archives/000169.html

Scientific Journals cannot be compared to TV documentaries. I'm not familiar with many other fields, but the IEEE Spectrum and ACM publish journals that are widely used as technical resources in engineering. Journals are not primarily a form of entertainment.

This isn't the only Sun censorship going on. Tim Bray (of XML fame, now Sun's Director of Web Tech) had a very insightful post on his 'ongoing' blog comparing Sun's strengths and weaknesses with Oracle's. It was up for all of a day before the lawyers stepped in and made him take it down.

It was all in vain, of course -- caches and copies will beat redactions every time. Here's one copy:

Us and Them

Interesting stuff!

It is a pity that people talk about virtual reality and related fields without even understanding the basics - but that is the consequence of media hype surrounding this field, together with people calling non-immersive, often even non-interactive applications "virtual reality". Computer games, SecondLife, QuicktimeVR are not VR, period - you cannot really achieve meaningful feeling of presence there. Of course, it sounds and sells better if you stick a gee-whizz sticker on the box ...

I love when people decide their definition of a phrase is the right one and that the rest of the world is wrong.

Word!

http://www-vrl.umich.edu/intro/index.html#NonImmersive
http://portal.acm.org/citation.cfm?id=1029964
http://www.fiu.edu/~mizrachs/VR.html
http://www.agocg.ac.uk/reports/virtual/37/chapter2.htm
http://www2.parc.com/istl/groups/uir/publications/items/UIR-1993-07-Robertson-Computer-NonImmersive.pdf

If you're serious about the topic, someone above mentioned Wittgenstein. The Saphir-Worf hypothesis is basic reading for linguistics. Here is a paper called "Notation as a Tool of Thought" written by a guy called Kenneth Iverson that discusses the effect that computer languages have on expression of thought.

Blithering about Kant being the first Python programmer and other such vacant nonsense may be entertaining in a limited way, but there are serious and fascinating issues in the study of linguistics, including those dealing with artificial language.

Encryption systems with this ability are called homomorphic. Systems where this can be achieved efficiently include ElGamal (in which it is possible to compute an encryption of the product of two plaintexts given only two ciphertexts) and the Paillier system (in which it is possible to compute an encryption of the sum of the two plaintexts given only the ciphertexts).

As you say, this property is pretty useful for applications such as secret ballots (when combined with distributed secure computation protocols which are mostly derived from this).

The difficulty with using this approach to database operations is that it is likely to involve transmitting an awfully large set of data which sort-of negates the point again as others have said about the papers cited in the original question.

It looks like this article is implementing the system that Jeff Han made about a few years ago and famously presented at TED. I'm glad to see this DIY article out there since it is getting lots of people interested in physical hacking, but I wish it would have referenced what came before. Here's the UIST paper:

Han, J. Y. 2005. Low-cost multi-touch sensing through frustrated total internal reflection. In Proceedings of the 18th Annual ACM Symposium on User interface Software and Technology (Seattle, WA, USA, October 23 - 26, 2005). UIST '05. ACM, New York, NY, 115-118. DOI= http://doi.acm.org/10.1145/1095034.1095054

Fair point -- I never thought of that. However, there are some other links to this:

• Inside a Google Data Center, including a video
• Notes from someone who says they were at the conference (also aggregated at the ACM

Maybe this is legit...

I can't hear you sonny, type louder!

So either I went to the wrong university, and consistently the wrong employers

Employers aren't there to teach you these things, and a lot don't care so long as you crank out code that mostly works (and, let's face it, it's not always easy to tell for them, and the concepts of "formal correctness" and "readability" and "maintainability" are often not even on their radar, unless a developer brings that up). And as for university - it may well be. If they had an OOD design course and never mentioned it, or at least described it in a formal way, then they wasted your time.

Of course, it had somehow become an established norm that "object-oriented design" course in the uni is basically just applied Java programming; from what I've seen, the best you can expect from a typical graduate when it comes to OOD theory is to be able to recite "encapsulation, inheritance, polymorphism" when asked what OO even is. It's especially ironic as only one of those three is actually a required ingredient, and even that is wrongly named. The very notion of MI gets people educated that way thoroughly confused when they first meet it, and you can forget about multimethods...

or it's one of those self-evident principles that just didn't have a name before Barbara turned up.

It was formulated by her in 1987. Unless your 8 years were mostly in Simula or Smalltalk, it did have a name by the time you've started working with OO. Definitely so if you've been doing Java or .NET.

As for self-evidence... I sort of wish it was, and it really is very simple conceptually, but the fact that so many people still get it wrong over and over again shows that, apparently, it's not all that self-evident for your typical coder.

From the Fscking Patent:

One problem existing in the art is that there are no systems and methods to bridge the gap between the programming paradigm used for object-oriented systems and the programming paradigm used for relational systems.

O RLY? They honestly want us to believe that they invented O/R mapping? Then what is this ACM paper from 1996?

Object-relational mapping by Scott Amber

Either somebody didn't do their homework and their patent is going to fall under a weight of prior art, or they're just plain patent trolls. Given that they waited until 2009 (9 years after the patent was issued!), I'm leaning toward the latter.

We published a paper that did evolutionary algorithms in the browser some time ago: Browser-based distributed evolutionary computation: performance and scaling behavior. In the same conference, there was another paper: Unwitting distributed genetic programming via asynchronous JavaScript and XML

We published a paper that did evolutionary algorithms in the browser some time ago: Browser-based distributed evolutionary computation: performance and scaling behavior. In the same conference, there was another paper: Unwitting distributed genetic programming via asynchronous JavaScript and XML

The original release of C#, which grew out of COOL, was indeed mostly a "let's copy it!" response to Java. But C# 2.0 and above is different. The design and implementation .NET generics came out of Microsoft Research Cambridge team headed by Don Syme, which included Andrew Kennedy. That same Don Syme is now heading the F# team (did you know Visual Studio 2010 will include Visual F#, by the way?), another longstanding MSR project. C# 3.0 and LINQ in general was strongly influenced by Haskell, specifically through Erik Meijer (worked on Haskell with Simon Peyton-Jones, and later designed VB9) and via C-omega and X# projects, both also of MSR - you should read Erik's paper "Confessions of a Used Programming Language Salesman: Getting the masses hooked on Haskell" about his role in LINQ and VB development, and its roots in his Haskell work (it used to be here, but it's down at the moment).

So you're very, very wrong. In fact, as time goes, more and more Microsoft Research ideas and even implementations find its place in .NET platform and the languages.

Are you sure? I found this paper about doing a fast Fourier transform with probabilistic CMOS hardware, with errors making it right through to the output -- but the errors were so small that, for many applications, they are irrelevant. And they were able to get by with 82% less power than a conventional approach with the same precision.

So they are taking a general-purpose CPU and using is as GPU?

No. As others have pointed out, there are GPU functions on the chip. More interestingly, there is a boatload more vector stuff that's not just a bolt-on. This is beyond AVX and includes gather/scatter, mask registers and 16-element vectors.

See the Larrabee paper for the details. It's an interesting architecture, for sure.

And to answer my own question (hate doing it on /. but somebody has to set the record straight).

http://scholar.google.com/scholar?hl=en&q=Senthil%20Cheetancheri&um=1&ie=UTF-8&sa=N&tab=ws

There is no published work on his so called groundbreaking research:

http://scholar.google.com/scholar?hl=en&lr=&q=Senthil%20Cheetancheri&um=1&ie=UTF-8&sa=N&tab=ws

I would expect much higher story verification standards by Cmdr Taco and NetworkWorld

Another related actually high profile published work on the subject:
http://portal.acm.org/citation.cfm?id=1159961

The ACM doesn't count?

There is one. Meet the ACM.

However, while you have to have a license to call yourself a professional engineer, no such license is required to call yourself a computer scientist.

But, to be fair, computer science isn't really about software development, although they often go hand in hand.

Read this interview with the original CTO for details on the brilliant designs for mesh networking, low power, and especially the display - there are fantastic advances here, the questions are whether this was the right place to do them...