Slashdot Mirror

My example or Logan's? Logan's is definitely flawed.

It's funny I searched for Peppers patents and didn't find anything. I did find this though Bob Moog read the second to last paragraph.

Nature Materials hasn't actually put the paper in an issue yet. When they do, there should be some decent editorial in Nature Mat itself and probably Nature's public news site. Meanwhile these articles are riffing on this press release and (apparently) some other press comments the author has made.

http://web.mit.edu/newsoffice/...

These hybrid materials could be worth exploring for use in energy applications such as batteries and solar cells, Lu says. The researchers are also interested in coating the biofilms with enzymes that catalyze the breakdown of cellulose, which could be useful for converting agricultural waste to biofuels. Other potential applications include diagnostic devices and scaffolds for tissue engineering.

The Register writes text news with the editorial style and standards of a red-top tabloid, and your reaction to the research it describes should be filtered appropriately.

What you said. Also, air temperature measurements might just be the wrong measure of climate altogether.

This is not a new idea. http://web.mit.edu/newsoffice/...

Cooling will certainly not be a problem and the geological activity is minimal (if at all present), so earthquakes and tsunamis are out of the question.

Nuclear power is frightening, since the U.S. used it to bomb Japan in 1945, and since nobody seems to know what a becquerel is, or they'd quit using it instead of roentgens. Of course becquerels are more fun, because it makes the absolute number 3.7^10 larger than if it were expressed in curies (also not a unit of radiation exposure).

Whee! http://web.mit.edu/newsoffice/...

Sharp pointy sticks are frightening! People have used them to kill other people for millions of years! Don't let anyone in Antarctica have anything to do with pointy sticks!

Cold is frightening! More people have died from freezing to death over the course of human history than have been killed in combat! Don't let anyone in Antarctica when it's cold!

Cooling will certainly not be a problem and the geological activity is minimal (if at all present), so earthquakes and tsunamis are out of the question.

Nuclear power is frightening, since the U.S. used it to bomb Japan in 1945, and since nobody seems to know what a becquerel is, or they'd quit using it instead of roentgens. Of course becquerels are more fun, because it makes the absolute number 3.7^10 larger than if it were expressed in curies (also not a unit of radiation exposure).

Whee! http://web.mit.edu/newsoffice/...

If you're just looking to pick more programming knowledge to continue teaching middle school then you do not need a MS for that. MIT's Opencourseware is a great place to start: http://ocw.mit.edu/index.htm

Hi, kudos for being interested, and not just throwing bombs.

The article you linked isn't relevant to your question, since we're interested in the question on whether gasoline lead is being measured in the ocean. As such, we only need to look at a time course for Pb in the ocean over the time period in question. (i.e., the age of leaded gasoline.) Your linked paper is concerned with much larger time scales, on the order of 10k years.

Why, because a private organization failed to centralize a decentralized currency? If you think about, this is a success of Bitcoin: it will be extremely hard to centrally regulate or control this product. The lesson here is the one the crypto-anarchist always wanted you to learn: in a digital society, only you are able to provide your own security, and only you are trust worthy.

These developments will alter completely the nature of government regulation, the ability to tax and control economic interactions, the ability to keep information secret, and will even alter the nature of trust and reputation.

-- The Crypto Anarchist Manifesto
-- Timothy C. May
-- tcmay@netcom.com

So.. you're teaching people to code by not teaching people to code?

It uses a Scratch-like interface, which is "coding" in every sense but typing. You still need to understand loops, conditionals, etc., and you have to learn how to structure a program. The only thing you don't have to know is the low-level syntax, such as correct spelling of keywords. It is a very good introduction to programming, especially for young kids that haven't learned to touch type yet.

Obligatory link

Aren't these called constraints or propagation networks?

Structure and Interpretation of Computer Programs and Higher-order Perl have got something to say about these.

But if you go back further, to pre-industrial times peoples work was slower with many more holidays - http://groups.csail.mit.edu/ma...

For most of human history 60 hours would probably not even be sufficient simply for the human need to occupy one's mind, since there weren't always the entertainment options that we have now.

It keeps getting suggested that pre-industrial people worked longer hours but the evidence just doesn't back that up. http://groups.csail.mit.edu/ma...

Many argue that our pre-industrial ancestors actually worked less - http://groups.csail.mit.edu/ma...

If you like your work (or money) enough that you want to spend your whole waking life doing it then all kudos to you. For many of us though money isn't everything, and doing anything for 40 (+) hours a week, week in and week out, just isn't our ideal life style choice.

You aren't going back enough.

Before the industrial revolution, "according to Oxford Professor James E. Thorold Rogers, the medieval workday was not more than eight hours".

"Detailed accounts of artisans' workdays are available. Knoop and jones' figures for the fourteenth century work out to a yearly average of 9 hours (exclusive of meals and breaktimes)[3]. Brown, Colwin and Taylor's figures for masons suggest an average workday of 8.6 hours[4]. "

You mean like this?

There actually is a very good use case for visual based programming, it's harder to stuff something up once you're abstract enough that it's just a bunch of blocks tied together. A lot of industrial stuff is coded like this. Visual function block diagrams tied together, hell Schneider's latest acquisition use a Microsoft Visio based tool for "programming" controllers. Even more interesting is the standards for safety systems has less stringent requirements on certification and verification if you use a function block or ladder logic based editor instead of just writing text.

Just like books with pictures, a text free programming environment has it's uses. I say this as someone who has written precisely one Android app, and I don't have a clue how to code in Java and know very little about the Android API: App-inventor

Do you think women are stupid? They can't figure out what they like or don't like?

No. There must be some explanation for why there is a "leaky pipeline" for women. A certain percentage of women enter, say, CS majors in college. A smaller percentage graduate with CS degrees, and the attrition is higher for women than for men. A certain percentage of women with CS degrees get jobs in the tech industry. *Of the recent graduates with CS degrees*, a lower percentage of women get jobs than men. Of the people who enter the tech industry, a lower percentage of women than men stay for 10 year. A lower percentage of women than men get promoted. A lower percentage of women than men start their own tech companies. And so on.

There are a number of possibilities here. One is that women are just bad at tech, or inferior to men or something. This seems unlikely because women perform well in other professions such as law and medicine, and there is a huge amount of scientific evidence saying there are no differences in IQ, etc., between the genders. Plus, that argument was used for a couple of thousand years to keep women from competing with men in the workplace so it has a lot of baggage, and people are justifiably hostile to the suggestion. So let's call that settled -- no one here is arguing in favor of that proposition (unless Lawrence Summers

is posting as AC).

Another possibility is that women are subject to systemic bias that makes it hard for them to succeed in certain careers. This was the conclusion of the MIT Gender Equity Project. This is uncomfortable for many people to contemplate. You, for example, do not seem like you possess overtly misogynist views and you probably do not see those views in your male colleagues. If men are not opposed to women in IT, then what could be the problem? Well, read the MIT study. A combination of unconscious factors can indeed add up to institutional bias.

There is also a third possibility that we ought to keep in mind. That is the possibility that efforts to get more women into IT are doing more harm than good by coaxing women into a career they're not really committed to, and then find they don't like and easily drop out of. I do not believe this is the case because the MIT study and similar studies adequately explain the phenomena we see. However, it should not be unthinkable to consider that we may be trying too hard to get women into IT, and the question of how to get them into the field is somewhat independent of how to help them succeed once they get there.

Or that without preferential treatment they will go elsewhere?

It's an empirical fact that women leave IT at a higher rate than men, and the causes for their departure are well documented: the incompatibility of an IT career with primary child-rearing responsibilities is a major cause, as is lack of advancement and opportunity. So without some change in workplace conditions, or "preferential treatment" as you put it, women demonstrably do leave IT and go elsewhere at a higher rate than men.

I would add that efforts to address the attrition of women from IT do not have to be "preferential" to women in the sense that men can't benefit from them. A single father faces a lot of the same challenges as a single mother, for example. Men can benefit from mentoring and career coaching, which is one way to help everyone (including women) learn how to achieve high job satisfaction and high productivity.

Sure it's just like your other computers. minus all the silicone and transistors.

Let me just repeat this, there is *no* semiconductor in the box, and *no* transistors on the chip.

And yes this is all out in the open and can be read up in Nature, MIT mirror

Expressing methodology in a computer language forces it to be unambiguous and computationally effective. The task of formulating a method as a computer-executable program and debugging that program is a powerful exercise in the learning process. The programmer expresses his/her poorly understood or sloppily formulated idea in a precise way, so that it becomes clear what is poorly understood or sloppily formulated. Also, once formalized procedurally, a mathematical idea
becomes a tool that can be used directly to compute results.

Gerald Jay Sussman

http://web.mit.edu/lipoff/www/CoSI/CoSI_files/SussmanAbstract.html

Would coding in visual languages like Scratch qualifies? Everybody should learn how to solve problems and do tasks in a formal way, and see how that solution runs by itself, without their intervention, free will, or common sense. Doing it wriitting text or manipulating diagrams is independent of the core question.

"Honey Encryption" to Bamboozle Attackers with Fake Secrets

Tom Simonite writes at MIT Technology Review that security researcher Ari Juels says that trickery is the missing component from the cryptography protecting sensitive data and proposes a new encryption system with a devious streak. It gives encrypted data an additional layer of protection by serving up fake data in response to every incorrect guess of the password or encryption key. If the attacker does eventually guess correctly, the real data should be lost amongst the crowd of spoof data. The new approach could be valuable given how frequently large encrypted stashes of sensitive data fall into the hands of criminals. Some 150 million usernames and passwords were taken from Adobe servers in October 2013, for example. If an attacker uses software to make 10,000 attempts to decrypt a credit card number, for example, they would get back 10,000 different fake credit card numbers. "Each decryption is going to look plausible," says Juels. "The attacker has no way to distinguish a priori which is correct." Juels previously worked with Ron Rivest, the "R" in RSA, to develop a system called Honey Words to protect password databases by also stuffing them with false passwords. Juels says that by now enough password dumps have leaked online to make it possible to create fakes that accurately mimic collections of real passwords and is currently working on creating the fake password vault generator needed for Honey Encryption to be used to protect password managers. This generator will draw on data from a small collection of leaked password manager vaults, several large collections of leaked passwords, and a model of real-world password use built into a powerful password cracker. "Honeywords and honey-encryption represent some of the first steps toward the principled use of decoys, a time-honored and increasingly important defense in a world of frequent, sophisticated, and damaging security breaches."

For those with an open mind, consider auditing this MIT course on climate http://web.mit.edu/newsoffice/...

A project specifically about finding undefined behavior is STACK. It didn't find any problems on the two projects I tried it on, but one of those is rather small and the other is pretty mature, so maybe most of the undefined behavior has been fixed already.

Just setting the warning levels a bit higher ("-Wall -Wextra" in GCC; despite the name "-Wall" doesn't actually enable all warnings) will already help a lot in spotting potentially dangerous constructs.

Also useful is Clang's analyzer mode ("clang --analyze"), maybe not so much for undefined behavior, but it did find a few cases of wrong pointer use (such as a potential null pointer dereference) in code I tried it on.

For C++ there is also Cppcheck, which is good at finding potential problems related to classes, for example data members that are not initialized by a constructor: initialization of class types is enforced, but C++ does allow data members of primitive types to be uninitialized after construction, for some reason.

Though the code behaves differently with, and without optimisation, and does not work on the new compiler whereas it did on the old,
this does not mean it is a bug in the compiler.

GCC, Clang, acc, armcc, icc, msvc, open64, pathcc, suncc, ti, windriver, xlc all do varying optimisations that vary across version, and
that rely on exact compliance with the C standard. If your code is violating this standard, it risks breaking on upgrade.

http://developers.slashdot.org/story/13/10/29/2150211/how-your-compiler-can-compromise-application-security
http://pdos.csail.mit.edu/~xi/papers/stack-sosp13.pdf
Click on the PDF, and scroll to page 4 for a nice table of optimisations vs compiler and optimisation level.

_All_ modern compilers do this as part of optimisation.

GCC 4.2.1 for example, with -o0 (least optimisation) will eliminate if(p+100p)

This doesn't on first glance seem insane code to check if a buffer will overflow if you put some data into it. However the C standard says that an overflowed
  pointer is undefined, and this means the compiler is free to assume that it never occurs, and it can safely omit the result of the test.

One of the comments points to DPF, which uses dynamic code generation to demultiplex packets. This is a very promising and surprisingly old idea. A dynamically generated classifier/filter could replace the entire network input path, and interface well with Van Jacobson's net channels. In addition to providing superior performance, it would afford far greater flexibility and modularity of code.

This is probably because they want the signature checker to fit in the CD boot loader. For historical reasons, bootable CDs imitate a floppy during the initial boot process, and contain an image of a 1.44MB floppy with a FAT file system. When you boot an PC-type x86 machine from CD, that simulated floppy (the file "floppy54.fs" for OpenBSD) is read by the BIOS and a file from it is executed.

This process is so retro that the initial program loaded is executed in 16-bit X86 mode.

how many backdoors can you hide in something that is open source?

Quite a few, if you're clever (although of course you only need one). Code that introduces a vulnerability can be very subtle -- so subtle that even if someone discovers it, they are likely to think it is a bug rather than something that was placed there deliberately.

You may not have noticed, but there isn't a single set of laws governing how to drive in the US. While the various states generally agree on stuff like stop at stop signs and red lights, less important details often differ. This is one of them. The actual universal law is that if you're going slower than the "normal" speed of traffic, you stay in the right lane. A quick google turns up this summary of various state "keep right" laws: http://www.mit.edu/~jfc/right.html

MIT OpenCourseWare is up to around 2200 courses... let alone the 20+ they've done through MITx. MIT has spent tens of millions of dollars giving free education material to the world.

Disclaimer: I work for MIT OpenCourseWare and still get annoyed that we have a ton of people who don't know about us cranking away at free course materials for the world for more than a decade! (MIT OpenCourseWare was announced in 2001.)

Cost has nothing to do with it (although going to school for free is a sweet deal). If I truly wish to accomplish something I'll find a way, regardless of cost.

Then good news! And WTF are you whining about?

FTFA: "The four-week online course, aimed at technical professionals and executives... The course will be offered at $495."

Register here.

Worth investigating: an Indian charity worked with together with Stanford and MIT D-Lab to develop <$50 above-knee prosthetics, and has been building 60,000+ per year for the last few years... Again, that's part cost, not chargable cost, but still...

Here's an article about them from 2011: http://www.smh.com.au/world/indians-work-miracles-on-a-shoestring-20110617-1g7yl.html (mirrored at http://jaipurfoot.org/media/media_reports/pdf/shoestring.pdf)

Except it is not a clearly labeled button, it is a button which has a function far beyond its label.

Strictly speaking, this isn't so. 'Like' is a well-defined word. That Facebook grants extra access to people who like certain things doesn't change the meaning of the word.

Are we ready for Tonika yet?

If you can accept that there may be more physical dimensions than our standard 3 + Time and that there could be a 4th physical dimension then it most certainly could be perpendicular to the orbit of the satellite.

https://groups.csail.mit.edu/mac/users/rfrankel/fourd/FourDArt.html

MIT Prof Peter Hagelstein, one of the rare true believers in battery-type cold fusion is teaching his cold fusion seminar again. Just about everyone else in academia does not believe him. Peter has done brilliant work in other subjects such as Xray lasers, so MIT tolerates him.

You might also like to have a look at this paper on using HMMs to convert a (continuous) chromatographic signal into (discrete) base pairs "calls" during DNA sequencing: Link. The problem seems similar to the one you are working on, in many respects.

Hell, I gave my daughter Scratch and they used it to make the "kitty cat" dance.

The *entry level* salaries for Google and Apple engineers in Silicon Valley is $105K. That's over fifty bucks an hour assuming a 40-hour work week.

I said software engineers, not software engineers at google. So you can knock about 10 grand off right there. And until you provide a citation about how much Google pays its employees, we're going with the state average. $95k a year comes out to $45.67 an hour. This is actually more than San Fransciso lists for the profession -- $40.66. We'll go with the more generous figure here.

So you're making $45.67 an hour. Woo! Big time money now. But Uncle Sam just showed up, and he wants his cut. Your biweekly was $1,826.80. Now it's $1,376.72 if you take a single deduction and are single. That's $688.36 per week net. As it turns out, taxes in California are a bitch.

MIT has created a Living Wage Calculator. I linked it directly to San Francisco for you.

They estimate that you need to net $1,929 to be above poverty. You're making about a third more than that. Coincidentally, most financial experts will tell you that having about 25-35% of your income as discretionary is the ideal case: Less and you can't really save any money for retirement, etc.

So as you can see, $95k might seem a princely sum to you, but it's not really. Especially when to get it you're working 80 hour weeks so your net per hour is about $8.60.

So no, if we're gonna mod people up or down on the basis of factual statements, you're going to -1 land, bud.

So everyone who doesn't donate hundreds of hours of work for free is selfish?

In my opinion, yes. Everyone should make time in their life to improve the world in some way.

Just out of curiosity, what do YOU donate?

1. I spend several hours a week teaching Scratch programming to 3rd-6th graders.
2. I volunteer as a math tutor at my son's elementary school for two hours per week.
3. I am a member of the "Ten Gallon Club" at the Red Cross blood center (80 one pint donations).
4. I have written several free educational apps for iPads and Android Tablets, and plan to write many more.
5. My wife and I funded a scholarship for two Naxi girls to attend a university.

Pardon me for nitpicking a bit, but incandescents are not "continuous spectrum". Generally speaking, they are more continuous than fluorescents and LEDs, but continuous they are not.

MIT society of physics students: "one can observe a continuous spectrum by looking at an incandescent light bulb."

The Hour of Code was teaching the outdated, sequential type of programming

Sequential code may be inadequate for advanced programming, but it certainly isn't "outdated". As a professional programmer, 90% of my code is purely sequential. Even parallel code has sequential blocks, and parallel programming skills can only be built on a solid foundation of basic understanding of sequential processing. Arithmetic skills are not enough to do calculus, but that doesn't mean arithmetic is "outdated".

I teach Scratch to 3rd-6th graders in an after school program. It is an event driven language, so they have to deal with parallelism and race conditions at some point. But they don't learn that until they have a solid foundation first.

To elaborate, just with "official" tools you have:

On Windows you have PowerShell installed by default since Win7 and JScript/VBScript since Win98. With .NET framework you also get C# and VB.Net compilers. As free downloads, you have C/C++ compiler as a part of Windows SDK and all of above in Visual Studio Express.

On Linux, in all main stream distros you get at least one of Perl/Python, often - GCC and sometimes Ruby, not to mention Bash and Awk. Repositories have a ton of other compilers and libraries with bindings available on your fingertips.

And even though I'm not familiar with OSX, I do know that Xcode is just a download away.

And then you have browsers, with things like http://jsfiddle.net/ , http://ideone.com/ , http://repl.it/ or http://scratch.mit.edu/projects/editor/

And then you have all the other languages and IDEs, free and proprietary all over the web...

If anything, we have overabundance of languages, but lack of interest due to all the content to consume chipping away the need to create.

I guess you're talking about http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-22.html#%25_sec_3.3.4 "A Simulator for Digital Circuits":

This system typifies a kind of program called an event-driven simulation, in which actions (``events'') trigger further events that happen at a later time, which in turn trigger more events, and so so [sic].

I suppose the modern version uses events more for user interaction, and also gets rid of lots of idiotic, stupid parentheses.

There is a nice book about composing complex systems from more primitive ones: Structure and interpretation of Computer Programs

As a scientist I always test the null hypothesis to quantify usefulness of my research. They did a bunch of work, but is it any better than a simple randomized selection of text?

As a quick test of the null hypothesis, below I have selected a random bible verse and inserted into the middle a random statement from SICP after the nearest to center semicolon, comma, period, and or or:

God is faithful, and he will not let you be tempted beyond your ability, this takes two arguments, a symbol and a list, but with the temptation he will also provide the way of escape, that you may be able to endure it.

And we have seen and, evaluating this combination involves three subproblems, testify that the Father has sent his Son to be the Savior of the world.

Jesus answered him, “If anyone loves me, he will keep my word, However, if we allow mutators on list structure, sharing becomes significant, and my Father will love him, and we will come to him and make our home with him.”

Then Jesus told his disciples, “If anyone would come after me the machine repeatedly executes a controller loop, changing the contents of the registers, until some termination condition is satisfied, let him deny himself and take up his cross and follow me.”

Verses from: Random Bible Verse. I scrolled around the TOC with my eyes closed, clicked a link, then repeated the process waggling my mouse erratically to select sentences from SICP. YMMV.

Software has been eating the free lunch Moore was providing before it got to the users; the sad reality is that the typical end-user hasn't seen much in the way of performance improvements - in some cases, common tasks are even slower now than 10 years ago.

This point of view is common, even though its odd disparity with reality make it seem almost anachronistic. Software isn't bloating anywhere near as much as expectations are.

Oh, sure, it's true that much software is slower than its predecessor. Windows 7 is considerably slower, given the same hardware, than Windows XP which is a dog compared to Windows 95, on the same hardware. But the truth is that we aren't running on the same hardware, and our expectations have risen dramatically. But in actual fact, most implementations of compilers and algorithms show consistent improvements in speed. More recent compilers are considerably faster than older ones. Newer compression software is faster (often by orders of magnitude!) than earlier versions. Software processes such as voice recognition, facial pattern matching, lossy compression algorithms for video and audio, and far too many other things to name have all improved consistently over time. For a good example of this type of improvement, take a look at the recent work on "faster than fast" Fourier Transforms as an easy reference.

So why does it seem that software gets slower and slower? I remember when my Dell Inspiron 600m was a slick, fast machine. I was amazed at all the power in this little package! And yet, even running the original install of Windows XP, I can't watch Hulu on it - it simply doesn't have the power to run full screen, full motion, compressed video in real time. I was stunned at how long (a full minute?) the old copy of Open Office took to load, even though I remember running it on the same machine! (With my i7 laptop with SSD and 8 GB of RAM, OpenOffice loads in about 2 seconds)

Expectations are what changed more than the software.

Hey, I've worked with quite a few developers that couldn't write coherent sentences or spell. Not all of them were bad at coding, however.

I bet they knew the alphabet, and could at least read the keywords. I volunteer for an after school program that teaches Scratch to 3rd and 4th graders. At that age (8 and 9) they are ready to understand programming. In kindergarten they are not. You can't keep the kids interested unless their programs involve graphics and animation. To do that, they need to understand distance, angles, and rotations.

I think (I was never in one) that the first tier universities allow even undergrads to interact with the world experts and do research under their direction (see http://web.mit.edu/urop/). This is a non-scalable function, which MOOCs can't do.

This is the reason MITx is such a good idea for MIT - it doesn't eat into their customer base, but that of lesser universities.

http://star.mit.edu/cluster/

The rest of it is easily scriptable. I have some ebs based AMIs that on bootup, connects to a central server,
registers itself (ticks up a text file, and adds itself to /etc/hosts).

If you combine starcluster for generic cluster management with the existing Amazon provided tools
http://blog.roozbehk.com/post/35277172460/installing-amazon-ec2-tools)
this is really only a days worth of scripting and testing.

There are also several public AMIs on Ec2 that are oriented towards scientific computing.
http://www.google.com/search?q=ec2%20ami%20scientific

This is my day job stuff.

I have used MIT's starcluster In the past for something very similar to this workflow. It provides a very user friendly interface for EC2 spot interfaces for almost the exact workflow you're looking for. They provide AMI's you can customize and a relatively well documented set of commands to easily launch spot instances.