Slashdot Mirror

Count me in. Building 20 was the beloved eyesore that was a part of the Institute's lore. Everybody would always be sure to tell the incoming freshmen that the buildings were saved from being razed because they were deemed historical landmarks (because that's where RADAR was invented).

Along comes Bill Gates (who never made a single academic contribution to computer science) waving around some money, and now it's suddenly OK to raze these buildings. And he gets a building named after him, not to mention all the Microsoft software that gets crammed down students' throats now. Brilliant.

I do wonder where the ROTC students are "doing their thing" now. When I was an undergrad at MIT, there was a movement to kick ROTC off campus because of their discriminatory policies toward gay students (which contravened MIT's anti-discrimination policies). But I never personally had it in for ROTC, and other nearby colleges (e.g., Harvard) had to send their ROTC students to MIT because they were unable or unwilling to accommodate ROTC themselves.

I do think it's cool that LCS and the AI Lab got consolidated into CSAIL, but these buildings are eyesores. Even the "Pei toilet" looks better. (For those who don't know, one of the buildings on campus was designed by I.M. Pei, and is covered by what appear to be bathroom tiles on the outside. IIRC, that's the building where the Media Lab is housed.)

http://web.mit.edu/gsc/www/programs/ring/bezel.shtml

My old office window looked out on the Stata Center and I frequently eat lunch there. I have always considered it a monstrous waste of money. A good building should blend form and function. The Stata Center was cleary designed with a particular form in mind, and the engineers were forced to make function follow. A nice building with few problems and more efficient use of space could have been built for much less money (like the new Brain and Cognitive Science building across the street). The money saved could have been put to better use as financial aid for students. Or for the upkeep of MIT's other building. MIT tends to sink ridiculous amounts of money into new construction projects, but seems unwilling to spend the money to properly care for the older buildings already constructed.

MIT graduate here.

I was around when they first unveiled the building and no body liked it from the start. The administration thought it would be a neat idea to put this ugly red metal installation art piece in the grass in front of Stata. Some creative people over at EC decided to turn it into a giant swing. The administration got angry and took it down right away. I tried to find pictures of this hack, but I couldn't find any. However, in a related incident, some more MIT hackers did this to the MIT sign sitting outside of Stata. I think that says it all.

The building has always had problems. During the first year, the fire alarm would randomly go off and everyone would have to evacuate. This was especially bad because the first floor housed a handful of classrooms (almost everyone had at least one class in 32-123, which held around 300 people, regardless of their major).

I held several UROPs (undergraduate research) in Stata and I can attest that the open work environment doesn't work. I usually ended up sitting around a bunch of people that weren't even in a related group so they became huge distractions. They would talk to each other a lot and brainstorm, but I was left trying to concentrate on my work. In the end, I just set up the software on my laptop and worked from my dorm room.

The floor layouts are definitely confusing. I always got lost when I had to find a professor's office for the first time. More importantly, I'd get lost trying to find a bathroom on a particular floor. Not cool.

Ironically, there is a huge water filtration system present just outside that harvests the tons of rainwater that we get and uses it in the toilets and stuff. I'm surprised that hasn't broken yet (maybe it has and I just don't know it yet).

And the only reason why MIT made such an odd looking building is for tourism. Tons and tons of people visit MIT every day for tours. They may be visiting MIT explicitly or they may just be visiting Boston and decided to take the trolley tour (which starts in Kendall Square, i.e. 2 blocks from the Stata Center) and they ALL take the same pictures. They pose in front of Building 7 or in Lobby 7 (77 Mass Ave.), they'll pose in front of the Great Dome in Killian Court and they all pose in front of the Stata Center (either the steps to the third floor or the, now reconstructed, amphitheater). I mean, without a few interesting sights, the tourists would get bored. While I agree that this sort of tourism doesn't necessarily generate MIT revenue, but it does generate attention and enough attention can be used to turn into money.

Check this guy out: http://www.csail.mit.edu/biographies/PI/bioprint.php?PeopleID=479

He's like a mix between Steve Ballmer and Phil Harrison.

The complex, which houses a Who's Who of Computing including Tim Berners-Lee and Richard Stallman.

Why do you want to design a system to 'protect people who do something new'? Surely the system should be run to promote progress in science and the useful arts, as specified in the US Constitution?

Maybe it does promote progress to have patents on software, but it's not a foregone conclusion; study some of the arguments (there may also be a good site arguing in favour of swpats, but I don't know of one) and decide what works best in the public interest, rather than just assuming that any measure in favour of 'inventors' is going to help the public.

He does not own nor can he own a trademark unless it is registered. Secondly, you cannot own a trademark to words that are too simple, such as ABC Computer. I think SimpleDog would come into this category of simple names. It's just too common. There was kid named Mike Rowe who was a software developer. He registered the domains MikeRoweSoft.com and MikeRoweSoft.net. Microsoft threatened and sued and lost. They finally did what they should have in the first place. They bought the names from this highschool kid.

There was also a man named McDonald who was also a software developer. You guessed it. He registered McDonalds.com and .net before McDonald's Hamburgers did. McDonald's sued and lost and finally bought the name from Mr. McDDonald.

Here's an article: http://msl1.mit.edu/furdlog/?m=20040306
The U.S. 6th Circuit Court of Appeals in Cincinnati ruled that Michelle Grosse did not violate the law when she used the *name of Lucas Nursery and Landscaping Inc* in her domain name for a Web site she created to complain about the Canton, MI nursery.

1. Install the necessary fonts (http://www.mozilla.org/projects/mathml/fonts/)
2. Apply symbol font fix (http://silas.psfc.mit.edu/tth/symfontconfig.html

Both require non equilibrium plasmas to work as advertised and that just does not work (The ions collide with electrons far more often than they fuse). In fact unless they can find a massive flaw in our current understanding of plasma physic thermodynamics neither can break even. Well the Bussard one defiantly, since its constant state.

And Bussard had responded directly to that issue:

Ions spend less than 1/1000 of their lifetime in the dense, high energy but low cross-section core region, and the ratio of Coulomb energy exchange cross-section to fusion cross-section is much less than this, thus thermalization (Maxwellianization) can not occur during a single pass of ions through the core. While some up- and down- scattering does occur in such a single pass, this is so small that edge region collisionality (where the ions are dense and "cold") anneals this out at each pass through the system, thus avoiding buildup of energy spreading in the ion population (Ref. 14).

In layman's terms, the Polywell design fuses ions faster than they maxwellianize, thanks to the ratio of time in core to time in edge. The full high level paper from Bussard can be found here [askmar.com].

You only need to maintain the non-maxwellian distribution long enough for the ions to fuse before they maxwellianize. Thermalization in the outer edge dominates the coulomb interactions from the core more than the collisions dominate the fusion rates. Those are the conditions that allow fusion to occur faster than maxwellianization. No magic, no violation of physics, just a beneficial design that Rider and Nevins both overlooked in their assumptions.

This view is the general consensus of held by physicist, not just my view.
And it's a very good thing that science isn't a democracy. There are many researchers who do not agree with the consensus. Some from MIT and University of Wisconsin-Madison.

Software is pure mathematical expression. As such, I can understand copyrighting specific implementations of ideas in software, but not the patenting of algorithms. The latter is a serious attack on intellectual freedom; Its privatizing math!

Mechanical devices are pure physics applications. For that matter, electric circuits and chemicals such as pharmaceutical products are all grounded in pure physics. Should a patent for a mechanical devices, electric circuit, or drug be invalid because it "privatizes physics?" Should a patent for a new type of vehicle be invalid because "it's all just physics?"

Pure mathematics (such as an algorithm) or pure physics are not patentable. But use of mathematics or physics for a useful application may be patentable under U.S. law.

The argument that software should not be patentable because software boils down to just mathematics has the backing of expert authority. It makes some sense from a mathematical perspective: if an algorithm f_1 is unpatentable, and adding a small algorithm to f_n+1 to f_n still creates an unpatentable algorithm, then all algorithms that have unpatentable algorithms at their very base are unpatentable by induction.

But public policy (such as patent law) is not bounded by theoretical mathematical models, even nice, clean ones supported by very intelligent people. At some point under U.S. law, unpatentable mathematics can turn into useful, patentable inventions. Where is the precise boundary? That's like asking: "how many grains of sand does it take to make a heap?"

Public policy ultimately has to deal with a world that is not easily modeled into discrete sets. There are better arguments against software patents then the argument that the algorithms are just mathematics.

And under your system stations won't try to overpower each other?

Such a thing as trying to overpower the competition would lead to an arms race none of those trying could win. It's rather easy to add power to a radio station up to a point. With a number of different stations broadcasting on the same frequency all anybody listening would hear is a bunch of garbage, interfering signals. Since there would be no clear signal advertisers would not pay to advertise. However by shifting the frequency a little there would be no interference, with today's technology radio stations can use much closer frequencies than was possible in 1934. The only reason to limit radio, and tv stations and require licenses is to limit competition. And of course big media like Clear Channel wants to limit competition.

Imagine if newspaper publishers and printers had to have a license, those granted licenses could have kept competition out. It being easy and cheap to print is responsible for creating thousands of newspapers and magazines. Heck, with DTP software anyone can create and publish a broadsheet. About 20 years ago a writers group I was in put together several short stories we had written and ran off copies, well Barnes and Noble ran them off, to hand out. Some publish broadsheets of only a few pages, or maybe only one or two pages, they can then sale small ads and hand out the broadsheet.

Now imagine someone starting up a small special interest radio station. A person with an interest in calypso, reggae, music could start up a station and sell ads to local businesses that sale the music, to a Cajun restaurant, or to a local band that plays the music. Another person could startup a talk radio station that is about model railroads and sale ads to hobby shoppes.

As mentioned a few comments below, affect/emotion detection can be useful for people with autism. Rana El Kaliouby has been working on a system that does exactly this in the Affective Computing group at the Media Lab. My understanding is that they have built it into an appliance which is robust to poor lighting situations, which is pretty impressive. There is also a great deal of interest in this in the human-robot interaction and human-computer interaction communities in general, so that robots and computer systems can respond more naturally to people in social situations.

I would have expected a more on topic explanation (nerds, matters, etc.) from a low ID-er, such as yourself, instead of the tired old psychobabble used by most armchair shrinks. but I guess you bought your ID on EBay too, eh?

(mods: apologies for the vitriolic remarks. I will not hold it against you for cracking the karma-whip.)

The original designers of all that equipment have either retired or died.

Not completely - I work as a NASA subcontractor, and I work with a few people who were around for the tail end of Apollo (granted, most are looking to retire soon - but they are still very sharp). But the real problem is information rot. Think about it - all the designs and reports from the 1950's and 1960's are written in paper. Fourty year old paper and photographs. Even in the best of storage conditions, these things degrade. I've been shown original documentation from wind tunnel studies in the Apollo era, and you can't glean meaningful data anymore. The Schlieren photograps are so washed out, you can't discern the shock structures anymore. Printed plots are faded. So much data is lost. Not all of it though. A lot of it got scanned a number of years ago, and posted online. In fact, much of it is public, on the NASA Technical Reports Server.

Why bother to design a lander that runs off of sunlight and generates its own oxygen from waste products when it's going to be launched by people who can't tell the difference between yards and meters?

Please, now. Read this report from IEEE Spectrum. It was as much an organizational problem as a units one. FTA:

Even if what ruined the Mars Climate Orbiter mission can be overcome, it should not be forgotten. The analogies with the Challenger disaster are illuminating, as several direct participants in the flight have independently told Spectrum.

In that situation, managers chose to cling to assumptions of "goodness" even as engineers insisted the situation had strayed too far into untested conditions, too far "away from goodness." The engineers were challenged to "prove it ISN'T safe," when every dictum of sound flight safety teaches that safety is a quality that must be established--and reestablished under new conditions--by sound analysis of all hazards. "Take off your engineering hat and put on your management hat" was the advice given to one wavering worker, who eventually went along with the launch decision.

Similarly, various versions of the trajectory debate in the final days of the flight indicate that in the face of uncertainty, decision-makers clung to the assumption of goodness; assertions of trajectory trouble had to be proved rigorously. Just the opposite attitude should have ruled the debate.

Other complaints about JPL go more directly to its existing style. One of Spectrum's chief sources for this story blamed that style on "JPL's process of 'cowboy' programming, and their insistence on using 30-year-old trajectory code that can neither be run, seen, or verified by anyone or anythin g external to JPL." He went on: "Sure, someone at Lockheed made a small error. If JPL did real software configuration and control, the error never would have gotten by the door." Other sources commented that this problem was particularly severe within the JPL navigation team, rather than being a JPL-wide complaint.

Some MIT hackers did just that. It's beeping instead of transmitting, but ya know =)

http://hacks.mit.edu/Hacks/by_year/2007/sputnik/

You're right, but for all of the wrong reasons. Carbon nanotube does kind of make sense if the manufacturing processes can be sorted out. But doing so amounts to coming up with systems that can somehow play tetris with free carbon -- a difficult problem at best. That said, the problem of carbon nanotube construction becomes trivially solvable when one takes into account that time itself is best expressed in terms of cubic heterodynes -- an understanding of which makes trivial the construction of materials far superior to carbon nanotube. Additional insight can be found here and, surprisingly here.

For kids, Alice has the staying power that Squeek and LOGO don't.

You know, I think it's unreasonable that patents can so greatly reduce people's freedom to create things, for fear that some of it may infringe upon some fairly trivial patent... Obvious or not, it places an unreasonable burden on developers, to use what they've learned except for those things they've learned about which are patented.

Yep. That's why the League for Programming Freedom says that "nobody should be able to dictate what kinds of programs you can write."

May I recommend Turtle Geometry to you?

AFAICT it's harder in some parts than SICP, and its all LOGO based.

I mean, how can you not to love a book that teaches how to simulate arbitrary curved surfaces and later simulate the curvature of space-time with just a turtle?

Scratch from the MIT Media Lab is everything I got out of Logo when I was kid except more fun. Kids can still learn recursion, sprite manipulation, even some coler stuff I don't remember from Mr. Keating's 6th grade symposia on the wonders of "mt". I have sent this link on to anyone that haskids and a computer. Awesome fun, and what the hell--some learning to boot.

I recently introduced my kids to programming with Scratch http://scratch.mit.edu/ rather than LOGO. It does include everything that LOGO does, but it has a lot of benefits that make the feedback from programming more immediate and accessible, and the web site is great for sharing ideas, sprites, etc.

Logo was good, but the language landscape is so vast now there are better languages for almost every task to which logo can be put.

It's interesting that most comments here are along the lines of "I saw LOGO when I was 7" through "I played with LOGO for an hour in middle school." I tinkered with a couple of LOGO variants as a kid, but my real LOGO experience ended up being... in college.

In my school's Computer Science department, the class that weeded out (or at least delayed) the majority of students was our Discrete Structures course. The theoretical part of the course focused on typical discrete logic, discrete math, sets, predicate calculus, etc topics. But the unusual part was that the professor was determined to break us out of the C++ mold that the introductory programming courses began. Therefore, he picked LOGO as the language for the course. Sure, interpreted LOGO wasn't the most blindingly-fast choice, but the list-based nature of the class made it very much a "LISP Light" that we could quickly work with for solving problems. Surprisingly, it was extremely flexible for the kinds of logic problems we were working with. By the end of the year, I really had to rethink my initial concept of "oh no, turtle graphics." Plus, we got exposed to a bunch of quite interesting offshoots, such as StarLogo, a massively-parallel-turtle variation of LOGO.

If you've never had to write a parser for an arbitrary boolean arithmetic expression in LOGO, then you've never really lived... (Er...)

MIT's OCW website helps. It has lecture notes, problem sets and solutions. Unfortunately, 18.01 (single variable calculus) is the lowest course on there that I know of.

If you're looking for multi-variable calculus (18.02) or differential equations (18.03), OCW has those pages too. Any course that starts with 18 is a math course (at MIT), though some will start with a 6 (the course number for EECS).

Another idea is to pick up an AP Calc book for practice. The College Board website even has a large PDF with an overview of AP Calc, including a handful of sample questions.

Good luck and most of all, practice.

Take a look at http://ocw.mit.edu/OcwWeb/web/home/home/index.htm

Along the same lines one set of tools I've had some good experience with has been http://ocw.mit.edu/OcwWeb/Mathematics/18-03Spring-2006/Tools/ , a set of parameter tweaking applets for simple differential equations. You can really get a good feel for what's going on by screwing around with parameters in realtime, and in my experience a good feel for what is going on is one of the most essential parts of actually making mathematics work for you.

And our good friends at MIT - http://ocw.mit.edu/OcwWeb/web/courses/courses/index.htm#Mathematics

http://ocw.mit.edu/

What a total waste when a far smaller investment in deep geothermal technological development will yield a far more stable, reliable and efficient energy source. See 384 page MIT study here http://web.mit.edu/ceepr/www/mit%20geothermal%20study.html.

What did they teach it?

It's at the Dar al-Kutub al-Misriyyah, which looks like this.

/. FTW!

looks like it's time for this old jem http://web.mit.edu/dzych/www/FunFiles/COMP.OS.Beer Maby it's time to update it.

Wow, what a fascinating point. Did you copy/paste that straight from SCIgen?

Actually, They would!

http://people.csail.mit.edu/rahimi/helmet/

How about a cyborg/Wifi joke instead?

If your kid likes Squeak, you might want to take a look at the Scratch project. It's somewhat similar, released by MIT, and there are a ton of projects you can tear apart and modify released Creative Commons - Sharealike Attribution on the site.

Only Windows/Mac unfortunately, but there is a Linux version planned apparently.

Think of these XOs as a very cheap way of distributing all the textbooks needed to take a child through the K-12 years.

After contemplating the cost of an XO vs. the cost of printing and shipping several hundred pounds of textbooks (and dry bookshelves to hold them all), it might be worthwhile to look at what can happen after K-12, through initiatives like MIT's OpenCourseWare.

Amen to that!

And with stuff like http://freetechbooks.com/freetechbooks for instance and the http://ocw.mit.edu/MIT openCourseware it really is beside the point whether you get credit or a "real-life" experience, whoever derives whatever value from this resource do so for their specific reasons and their specific conditions and circumstances and therefore this sharing of valuable life tools need to be commended

http://ocw.mit.edu/OcwWeb/web/courses/courses/index.htm

MIT World already does this for many important events at MIT (including, but not limited to, course lectures). Plus, the quality is better than Youtube videos.

I have the good fortune to attend an excellent law school -- one of Harvard, Yale, or Stanford -- from which literally anyone in the class who wants to get a job that pays market rate can, and in the market of his choice. I agree that the calculation can be substantially different for lower-ranked law schools -- but it's also true that each year, probably ten thousand students or more gain entry to this law firm market tier. Granted, the cost of living is higher in New York and other primary markets, but nothing stops a law graduate from working in, say, Cleveland and making close to the same salary -- we're talking around $120-140k instead of $190k. The hours are also substantially less in those markets. It's really good money.

Since you mention MIT engineers, this site quotes a starting salary of low sixties for people with no work experience. I don't know if this is up to date or if it represents engineers generally, but my understanding is that engineers' salaries peak, for the very best of the profession, at or less than where big firm lawyers' salaries start. (Obviously this does not account for entrepreneurs like Larry Page and Sergey Brin -- but then, they were PhD students, and enjoyed a degree of success so rare as to be insubstantial from an ex ante point of view. Similarly, it does not account for engineers who move into management, since the skill sets are nearly disjoint and the engineering degree is probably a poor way to pursue that career path.)

You're also quoting 1999 salaries. Law firm starting salaries shot upwards during the dot-com bubble and never came back down. In the past two years alone, they climbed from $155k (including bonus) to $190k. It's a different profession now than it was even a decade ago: longer hours, yes, but also a lot more money.

Finally, your debt calculation needs serious work, at least in my case. First, I received about half of my law school education for free, because top law schools give out a fair amount of grant money and my family isn't all that wealthy. What remained, I covered with government loans. There was a loophole in the government loan laws that let you lock in low interest rates on Stafford loans by consolidating. I did so, and all $40k or so of my loans are locked in at an interest rate that is below 2% when you account for the reductions for on-time payments and automatic online payments. I am also going to drag repayment out over 30 years. Using a conservative rate of return of about 8% for an investment in an S&P 500 index fund, accounting for the way inflation (say, 2%) nearly perfectly counters the interest, the amount I will actually need to pay in 2007 dollars is $16,211.17, less than the price of a new Honda Civic.

Yes, I still spend three years in law school, but these are damn nice years, during which I probably work less than 30-35 hours per week and spend the rest of the time drinking/playing video games with friends, traveling to nearby state parks, and generally enjoying an awesome life. Also, I could not have gotten a high-paying job out of law school. Working long hours in a high cost-of-living city at a job with potential for advancement, I'd have been lucky to gross $40-50k per year. After all, I have no work experience and no marketable job skills. By comparison, last summer alone I worked 17 weeks at a law firm and grossed $52,700 -- more than the total cost of this year's tuition.

The fact is that I have an awesome deal. I'm not saying this to be smug, but to make the point that yes, it is very much in the economic interest of America's smartest college graduates to choose law over engineering or even medicine. Firms are throwing money at me in quantities that are obscene for a 24-year-old. It shouldn't be at all surprising if there is substantial brain drain in other disciplines with this as the alternative. For me, it was not an academic concern: I come from a family without a lot of money, and I was a math majo

I've watched the videos that go with the SICP book, and learned quite a bit from them.

I'm sure I would have gotten more out of actually taking the class, but the videos alone were still helpful.

Free sharing of knowledge will only help create more and better engineers and scientists. MIT does something similar as well- at least outlines, and sometimes full lecture notes and videos are available at http://ocw.mit.edu/ for almost all their courses.

> Lisp could be cool, but I hate the fact that it's not standardized

So you haven't met the Common Lisp standard or the Scheme standard (respectively the union and the intersection of preceeding Lisps)?

Conceptually, most of the functional programming languages aren't that different. (There is a slight divide between the graph reduction languages like Haskall and it's derivatives and the direct reduction languages like Scheme). Much like procedural languages, once you've learnt the concepts, it's easy to move between them.

Get yourself a copy of "Structure and Interpretation of Computer Programs" :

http://mitpress.mit.edu/sicp/

read it, do *all* of the exercises and you *will* be a better programmer and you *will* understand functional programming.

Next...

http://people.csail.mit.edu/rahimi/helmet/

WTF are you smoking? There was no professional moving company involved. Students and alumni moved the cannon.
They just happened to have put in a modicum of effort to lend credence to their social engineering, while
working in a few subtle jokes. http://hacks.mit.edu/Hacks/by_year/2006/mitcannon/

For instance, the "moving company"'s name is "How & Ser." & is a ligature of et, latin for and. Read aloud,
How et Ser becomes Howitzer.

The first statement is false. For tax year 2007, the personal exemption is $3400 and the standard deduction is $5350 for single filers. Thus, at least the first $8750 of income is effectively exempt from taxation. Your other two statements are correct, but in my personal experience it's unusual for TA/RA stipends to approach that amount.

MIT stipend is "science and engineering doctoral students who have passed qualifying exams to a salary of $2121 per month and master's students to $1939 per month." ( http://www-tech.mit.edu/V126/N8/8stipends.html ) That's about $9,000-$10,000 per semester which goes over the personal exemption on a semester alone. Then, there are 3 semesters per year which a graduate student can be employed.

In fact, stipends *are* exempt from Social Security taxes, provided that one works fewer than 20 hours a week, and only while regular classes are in session.

You're right on that. But, federal, state and city taxes still have to be paid and the non-immigrant student is not expected to use social security in the future so it's fair I suppose. On the other hand, filing that non-resident alien form loses a few tax benefits - the one thing that pops to mind is the President Bush's tax refund wasn't available for non-resident tax filers.

I realize this wasn't your point, but you should have a look at kerberized ftp and ftpd.

I sympathize with your indignation. I don't see why people can't just recognize what you are trying to do and suggest a better way (if there is one). Ok, so chroot is not the equivalent of BSD "jails" or Solaris "containers." Is there an equivalent for linux? Shouldn't there be? It seems like these are important questions. Calling someone incompetent for wanting this functionality and not finding it is just an attempt to distract from deficiencies in linux.

Your high ID gives a hint on the origin of your cluelessness, young one.

You have much to learn.

To summarise: when referring to MIT, "hacks" are pranks, often played on rival universities, usually Caltech or Harvard.

However, he may not have laughed at this one: http://hacks.mit.edu/Hacks/by_year/2004/stata_kiosks/