Being a senior IT guy at the University of Michigan, being an Ars Digita alumni, and knowing intimately how Universities work, I can answer this question:
Academic institutions LOVE to think that they are somehow different, special, gifted, unique, and dare I say it - divine.
Actually they are not that much different from other enterprises. People simply cannot comprehend that the cost of going bespoke is vast.
The MIT folk very proudly told me how they built their own system to replace the IT functions my company provides. They only managed to spend twice as much on the project as it would have cost them to by the product of my most expensive competitor and about six times what I would have charged them.
I don't think that outsourcing to India is the biggest threat to IT jobs. Outsourcing via Web Services is. Most programmers work for IT integration shops churning out bespoke widgets for clients deploying SAP or Oracle Financials, Peoplesoft and the like. A great deal of that work is repetative glue logic and gets cobbled together using string and sealing wax coding styles. Its the sort of stuff that gives Perl a bad name.
I think that over the next ten years most enterprise computing will go the way it has gone in the hotel business - outsourced commodity product. When you check into a hotel the screen in front of the assistant manager is simply a terminal to an off site central mainframe. Same goes when you step into your bank, Springfield Coop Bank probably outsources all its IT needs to one of the bank tech companies.
Payroll has been outsourced for years, why not outsource all enterprise IT systems? Payroll, Accounts Payable, Accounts Receivable, HR, each company thinks it just has to do its own thing but in practice everyone works in almost exactly the same way.
Can you explain this horrendous overhead? Most student jobs on campus are work study, meaning that the wages are subsidized by the Financial Aid office and the employer only pays 1/4. That's not even counting the grad students who'd work for free as research.
When I was at MIT the rules were that overhead was charged at a rate of 2.75 times expenses. So if I hired someone as staff and paid them $1000 I would be charged $3750 from my budget.
The rules for students were somewhat different but still pretty grasping. Basically I would be charged the amount they were actually paid plus overhead and added to that their cost of tuition, I can't remember what overhead would be on that. Tuition at MIT these days is $29K per year. So over a year a student would cost my budget something like $60K, and the student would see less than $10K of that and I would see about 15 weeks worth of work if I was lucky.
There is also overhead on external contractors but nowhere near as much.
And yes, this is a complete stinking racket. The only reason it continues is that the government allows the major research universities to do this type of padding as a means of giving them an under the table subsidy.
I have no clue where the money goes. If you look at the amount of time that the students have contact with the faculty, the amount the faculty are paid and the cost of tuition the sums don't add up. And thats before you consider places like LCS/AI which have always been self funding through government grants. Perhaps the President has a yatch somewhere like the Stanford guy had, he would have to be a lot better hiding the thing round MIT though and if the students found it...
If we pay exhorbitant license fees for second-rate crapware with first-rate marketing, we don't have any money left to pay American programmers. Or apparently, even to hire American grad students.
I am somewhat surprised that what MIT needed did not already exist as commercial off the shelf code. Their requirements are hardly very unusual, in fact since the content is not going to change much once it is put up there is not a great deal of difference between this site and any other web zine.
What this looks like to me is a boondoggle. $2 million is pretty easy to spend on software if you go bespoke. That is the main reason why most of the open source arguments you see on slashdot are bogus. If you can pay $100K for a product that is 90% complete you are one heck of a lot better off than you are paying $0 for a product that is 70% complete, maybe on a good day.
Open source is great provided it does exactly what you need if you have to do extensive programming then Gartner are completely right.
Building a system around Microsoft CMS is one heck of a lot better than mucking arround trying to make CVS do this type of thing. I don't have an issue with that part. But $2 million to customize it...
Incidentally MIT students are hideously expensive. The student may not get paid much, but the overhead charged by MIT is horrendous and the results can be 'variable' to say the least.
SSC owns the trademark, the "volunteer organisation" is made up of people who used to contribute to the SSC/linuxgazette.com site, and got angry that they migrated to a CMS, so they split and took the name. That sort of hijacking is what trademark law is for.
Exactly why should we care about this dispute?
If a bunch of folk get tweaked about using a content management system - a twenty year old technology used on the web for ten years...
Basically the group has forked and the dissident group had some links to their breakaway site. The SSC people then took out the links to stop people finding the breakaway site.
The breakaway site is also calling itself the LinuxGazette. This is probably actionable if there is a valid trademark registration. But at the end of the day it is not the name, its the audience.
If they plan to do it, they better do it quick. The only (I believe) common character of the trilogy and the Hobbit is Gandalf. Ian McKellen isn't getting any younger.
I seem to remember that Bilbo was also in the Hobbit.
As for the age of Gandalf, since he holds one of the rings of power and is thus ageless there is effectively little difference in his age between the two books.
In the real world however it aint gonna happen. Christopher Tolkien is known for being an arsehole on these kind of issues. I mean he kicked his son out of the family because he liked the idea of Peter Jackson making the movies. JRR sold the rights for LotR but not for the Silmarillion (of course since Christopher has his name on the book as well). Oh well.
One of the issues that Jackson could probably address that would help the situation would be to actually pay a decent royalty for the rights to the Tolkein familly. JRRT originally sold the rights for GBP 100,000 to meet a tax bill. Jackson almost certainly paid rather more to purchase the rights from whoever was holding them, but the Tolkeins would not have seen any of that money.
People do not go to the press and announce that they have an issue with the division of the cash, much better to complain about artistic integrity or some such bollocks. You sound so much more principled and so much less of a whinner. Of course the familly has not exactly been doing badly from sales of the books recently.
New Line could easily afford a lump sum of $10 million or so ex-gracia. I suspect if they did someting like that a lot of the resistance would suddenly disappear.
Sooner or later the Hobbit and the Silmarilyon will get turned into films. Note that the BBC managed to get hold of the rights to do radio adaptations so the issue is not exactly forgone.
I don't think that the Silmarilion would be a single film either, it would be a series. The reason the negotiations are apt to be fraught is that everyone with some brains realises that this is potentially the start of a whole new film franchise which could ultimately rival the Bond series.
either way, moore's law is dead. Kurzweil suggests that after moore's law, an exponential boost will occour, with the advent of a new technology, not simply in materials, new research all together, in accordance with the aforementioned "thillbert's law";)
Progress is not going to end, but the automatic metronome of Moore's law will no longer be the driver. The rate of progress will slow for a while then start to pick up. The Intel paper says as much.
Incidentally the point of the paper seems to be to push out the end date and fend off rivals proposing the same ideas. The tunneling effect is quite definitely the end point of traditional logic gates. The astonishing part of the paper is that the end they cite is a 16nm process (with a 5nm gate), the smallest scale currently in use is 37nm. In other words there are only four more generations to go, two generations resulting in the feature size halving which means four times the number of transistors. So if the old two year schedule were kept Moore's law comes to an end in 2011.
They also point to the fact that Intel themselves have pushed out their dates for adopting new processes and are planning for three year gaps between generations. I have suspected that Intel has been the main factor in keeping the industry to the roadmap of Moore's law for some time.
Uh, so what you are saying is that could just as well shut down the whole justice system, because the threat of jailtime for rape doesn't prevent rape? The threat of getting punished for illegal actions is highly preventive!
Folk should note the difference between this argument and its usual populist invocation, that punishments deter crime. Most studies show that criminals respond to the threat of being caught. In other words it is not the length of the prison sentence that is the most effective deterrent, it is the likelyhood of a sentence being applied.
A Clockwork Orange is actually relevant to both side's arguments. The point that Burgess was making was that thuggery was out of control because the response from politicians was opportunistic. The first response is to impose harsh punishments, dehumanising Alex in a process that was certain to turn him into a worse criminal. Then when the winds of public opinion are swayed the other way the minister suddenly sees an opportunity to present Alex as the victim.
Bounties can be effective in limited circumstances. The guy who had Saddam's sons staying as houseguests seems to have decided that he would rather live in the US with $30 million. On the other hand they have not led to the capture of Bin Laden, Al Zawahiri or any of the top Al Qaeda leaders.
The problem with bounties is that hacker groups tend to be curiously diffuse. Hackers use multiple psuedonyms. They quite often turn rival groups in if they get the opportunity to do so even without bounties. Within a group all the members are implicated. It is difficult to turn in another member without risking prosecution yourself - particlarly if the other member cuts a deal with the prosecutors.
This is assuming that De Beers doesn't push these people off a high rise first.:/
This would be a thin layer of synthetic diamond, not the mined type that deBeers has a monopoly in.
The fundamental limits are reached sooner in some technologies than others, but there is no technology that is immune from any sort of limit.
Even if there is an alternative technology the transition from silicon to a totally different substrate is something the industry has tried before and conspicuously failled at. There was a time when Galium Assenide was the bees knees, these days it is an important niche (direct band gap and all that) but nobody is building GaAs computers.
The other factor is that there seems to be a tradeoff between the point where you hit the quantum limit in a given technology and electron mobility that bites you in the a**.
I suspect that we see Moore's law start to slow before it comes to a halt.
Vega is only 25 light years away, so they'd be looking at the 70's. Of course, once they hear Disco, they'll probably decide nothing of value could have ever come from our system.
They would probably be more interested in the political situation. They would be watching the throes of the Nixon impeachment crisis in the West, China would still be in the middle of the cultural revolution, Vietnam would have ended but only just. The USSR would be in mid collapse. Latin America is run by cliques of corrupt generals who murder tens of thousands (Pinochet) or hundreds of thousands (Argentina).
There has just been a war in the middle east. The Iranians are about to kick out the Shah (a brutal thug on a par with Saddam Hussein) and the Ayatolah would appear soon after to pervert the democratic revolution the same way Lenin appeared on the scene in Russia after the Tzar was deposed.
Things don't get any better for quite a while and they get worse before they get better. The nuclear arms race accelerates, the US and the USSR are engaged in a series of proxy wars that appear likely to turn nuclear. If you look at the situation from the outside even the events of 1989 might be considered evidence of further instability rather than a good sign.
On the whole I don't think that they are going to be avoiding talking to us just because of the disco music...
I think we should get our act together globally before we start to try to join extra-terrestrial clubs. If there is anyone out there worth talking to they already know about us.
That's basically pay-to-view though. And given how resentful many Americans are of federal taxes I can see lynchings happen the day the USG tries to impose a license fee.
Yeah Americans would NEVER pay $50 a month for cable TV.
I pay enough to watch TV without having my viewing interrupted by adverts. All that Tivo will cause is a small rejig of the revenue model.
I think we will see the end of the network TV model and the low end cable model. There will be fewer channels but what is left will be more like HBO than CBS.
The low end cable TV shows that do the bare minimum to get an audience will probably have difficulty finding people who pay for them.
The network TV model of spending five million dollars to create another episode of a formula TV series like Friends will also crash. Those guys are not that funny, they were not as funny as Seinfeld when he was still going and they have not got any funnier since Jerry stopped.
The thing I find fascinating about the US network TV model is how they spend so much money to achieve total mediocrity. For example what is it with summer re-runs? You have a sparse resource, network time. So you pay one guy $2 million an episode you will show repeatedly, why not get a second string guy in who will do a cheap episode for a mere $10K or so and have some variety? The thing about summer re-runs is that nobody is waiting for the new season of the X-Files, it never stopped.
Another wierd network TV choice is that they all run a carbon copy of the old Carlson late night TV show which is itself a copy of Bob Hope's radio act. Same presenter every night with the same formula.
What is somewhat more depressing is that in the 2000 election the late night TV shows were the only place where anyone asked any of the candidates a challenging question. Thats because the campaign managers have to work to get their candidate on those shows, with news shows they will pick the most favorable interviewer. It is somewhat sad that they even avoid Larry King's softball questions these days because he is considered to aggressive.
They can all go bust as far as I am concerned. So long as I can still see Norm Abrams (on the Porter Cable channel), Jon Stewart, Jessie James, pretty much all the rest of the junk can go to the hell it came from.
Recognizing advertising and falling prey to it are two different things. I know the italian job was a big ad for the BMW mini, but I also know that car has nothing going for it that other cheaper cars don't, except the "look", which I find to be inferior to the original mini, which once (one or two years) came in an AWD version at least,
You are missing the point. The movie was not a mere product placement for the mini, the whole point of remaking the film was the new mini. The original mini was a slow seller until the original Michael Canne version of the Italian job came out. Then it was suddenly an overnight hit, the cool thing to be seen in in swinging London.
Without the car there is no movie. A remake of the Italian job with any other car would have been a flop. Its not like Bond who can drive a Bently or an Aston Martin, but lets face it everyone knew that it was pure product placement when he was driving the BMWs. The film had to return to the Astons because the product placement was damaging the credibility of the film.
If a President lands in Iraq and nobody notices does he gain votes?
Only if you count the military vote....
In my experience there is nobody as cynical as an enlisted infantryman posted to a war zone. I very much doubt that the GOP will find as many enlisted men voting for them as did in 2000.
Real science decidely does have a place in our decision-making processes in any field. But by definition that means the hypotheses and data must be verifiable by multiple, perhaps antithetical parties. That's science, not this mumbo-jumbo the parent was talking about.
Popper's falsification principle will identify pseudo-science 100% of the time. The problem is that it gives a fals positive for real science most of the time if you use the original strict criteria. Popper himself admitted as such (see the OUP volume on Scientific revolutions where Popper responds to Kuhn). Popper originally developed his falsification principle as a reaction to the logical positivists their verification principle clearly did not work for induction. It is somewhat ironic that the two best known 'members' of the Vienna circle were never actually members. Wittgenstein refused to join and Schlick blackballed Popper because he denounced Wittgenstein's antics. Popper then used his falsification principle to mount an attack against the freudian psueuds and Marxism.
A test that does work is intent, is the theory intended to be subject to falsification? are the modifications of the theory to match observations made in good faith? Unfortunately in the process the test is no longer sharply objective.
One of the depressing aspects of US academia (which is spreading to the rest of the world) is the confusion of technique with method. When I was at MIT there were lots of folk who spent their time endlessly analysing the results of an intrinsically flawed experiment to five decimal places.
The big problem with the social sciences is physics envy. People want to be able to do the same type of science they think the physicists do, comming up with firm, immutable results.
The point I was making in the original piece is that the data is simply too thin and too unreliable to allow that type of analysis. There is no measurement in social science that is not contaminated by multiple effects.
The use of science in politics has been contaminated by the emergence of endless 'crank tanks' dedicated to providing the result their sponsors pay for. We have already seen the laughably named 'Independent' institute being paid by Microsoft to tell us that network effects do not exist. Until recently these institutes have mostly been right wing outfits funded by Richard Mellon Scaife.
Now that George Sorros is committed to spending equally large sums to attack the right one wonders what the outcome might be. Fortunately Sorros is one of Popper's students and he seems to have a personal commitment to the ideas Popper advanced. Still one has to wonder if the result might not end up being the left turning to the same junk science and propaganda that Scaife has pushed the right into.
or maybe Bush did not actually go to Iraq. Like the 1960s moon mission, the Air Force just flew Bush and these gullible reporters around in circles before landing at a fake base in an allied country somewhere. The reporters were only on the ground for three hours. How would they know where they actually landed without cell phones or GPS?
I don't believe that happened, but I would not be surprised if a conspiracy theory of exactly that type started to spread through the Internet.
If a President lands in Iraq and nobody notices does he gain votes?
But then wouldn't it lay blame directly on those "clueless parents" instead of on the "Video Game Industry" when their child does something moronic, like shoot at passing cars?
Probably for the same reason that two 15 year olds shoot up a school and folk appear on slashdot within a nanosecond explaining how gun control would not possibly have prevented the event.
Very little political debate in the US seems to ever be influenced by science, it is mostly predjudice and emotion.
Mind you, things are not necessarily that much better in science. Remember that AIDS/Polio vaccine connection that came up about ten years ago. Instead of checking a pretty strong prima facie case the establishment tried to quiet the issue - litterally in this case with a threat of libel proceedings. Thats not science.
Last year the stories had finaly percolated back to Nigeria and suddenly people were refusing the vaccine. Bad news when polio is inches from being erradicated. Betcha wondering why it took so long for the vaccine to make its way back to the people who were used as guinea pigs for testing, oh, well guess not.
So finally the science establishment gets panicky and does the tests that should have been done when the controversy started. They checked the remaining vials of vaccine from the tests to see if they could identify HIV DNA or money DNA. Turns out that the monkeys used to incubate the vaccine were a type that do not have a HIV strain and there was no HIV virus detected.
So the establishment got it right all along? Well not really, why didn't they insist on doing the test when the story first broke? The only logical reason to resist would be if you feared the result.
if you're going to whack someone, first hide your phone in a restaurant a couple miles away....then you can "prove" you weren't at the crime scene.
A while ago, before 9/11 I was sitting in a bizare meeting with a bunch of wireless execs who were breathlessly telling us how great their new location finder service was going to be. They could send you adverts targetted at people in a particular location.
I was rather unpopular when I asked if the customers would buy a product if the chief benefit was going to be to enable a new kind of spam. "Perhaps they don't get the choice"
I was even more unpopular when I pointed out that the regulators in Europe would blast this type of thing on privacy grounds. "Oh the regulators tend to be more sensible than the general public".
I pointed out that my cousin, one of those regulators has survived two assasination attempts and may have an opinion about a technology that gives away his position. In Europe privacy is not something that you muck arround with.
Today the risk of this type of scheme would be obvious even to a US legislator. Now right to life will be able to stalk doctors who provide abortions by telephone, Saddam loyalists will be able to stalk senior Republicans and Al Qaeda will be able to stalk everyone.
So they are finally working out socially acceptable ways to package up the same technology. Was it really necessary to have the dotCOM bust before some folk got a clue?
What's with the fucking Republican moderators? Go fuck yourself, moderators.
The GOP sent out an alert asking people to post responses. They have a group of activists who are paid to surf the Web at minimum wage and report any political speech by dissidents.
Someone should tell Rove to be a bit more carefull, not all the people he has working there are Republicans:-)
No reason to wait... This kind of thing has happened many times before. You send 'em a "comply or desist" letter, which should weed out all those infringing due to ignorance. So far, all GPL infringment cases have been settled out of court, which is why everybody's saying it's "untested in court". In reality, no lawyer in their right mind would actually try to fight the GPL. It's not like this issue's never come up before.
Well Boies is certainly wrong but he is not insane, just exceptionally greedy and prepared to make an argument he must know is shit. I was really pleased to hear that Boies is defending Conrad Black in his embezzlement case.
The source code availability issue is somewhat odd, it is from a pre-Internet age. There is no difficulty in getting hold of Linux source. The issue to argue over would be any modifications of Linux. If Linux is so good it should not be necessary to modify the kernel to run embedded. It is not clear that there isa modification here, a layered app is not a modifocation.
The other weak point if someone were to bring a case is standing. The only party with the right to bring a suit would be the copyright holder or holders. That is why RMS is keen on people signing over the copyrights of free software to the Richard Stallman Ego Enhancement Foundation, so that future generations can enjoy LiGNUx as it was meant to be known.
Eventually some people who complained a lot got their vote counted and their right to vote restored.
But Harris and Bush succeeded in keeping over a hundred thousand voters off the rolls, most of whom were illegally and illegitimately denied their right to vote. The fact that the errors were corrected in a small number of the cases in the report changes nothing. Those two witnesses were called to show the lengtrhs that Harris and Bush went to deny their rights.
I took a look at Workindevs posting history. A lot of posts on electronic voting, all trying to convince us that there is no need to worry dears, those nice republicans are 1000% trustworthy and none of them would ever tell a lie, the WMD will eventually be found, the tax cuts are not almost exclusively for the rich and Nixon was not a crook.
Also, Dean is a Bush clone. He even went to a more prestigious prep school than Bush. Dean's grandmother invited Bush grandmother to her wedding. Dean and Bush even went to Yale at the same time. They are both Old Money Bluebloods, and that is who they are going to look after and have empathy with.
Oh come on, Dean got a medical degree. I don't think you get onto those courses through the 'positive discrimination' that Yale practices to favor 'legacies'. I certainly don't think you can get a medical degree and practice as a doctor if you don't have the brains and work at it.
Dean got into full time politics in by accident. He was the leuitenant governor, a part time job in Vermont and became governor when the sitting governor suddenly died in office. He was a successful governor for over a decade.
The odd thing about Texas politics is that the Lieutenant Governor actually has more power than the governor. It is a legacy from a time when they had a corruption issue with the then incumbent.
Kucinich OTOH is a card carrying union member from a working class background.
Which is great. But that is not enough for a platform.
Kucinich has the luxury of knowing that he is not going to get nominated. So he does not have to worry about advancing politically costly ideas with a marginal return.
This claim is a pretty good example of somebody trying to find evidence to support his preconceived conclusions, and it is thoroughly debunked in the dissenting statement.
Ah so a dissenting statement by a group of Republicans proves that there was no electoral fraud!
Diebold say that their machines are completely reliable. So I guess that settles the matter, NO NEED TO WORRY DEARS.
That is it, sleep tight, the GOP tells us that they do not rig the polls, so we should just believe them. If anyone claims anything different THEY MUST BE A GODLESS COMMUNIST so anything they say can be ignored.
BTW the dissent disputes the findings by rejecting all the testimony and facts found as 'anecdotal' they provide no contradictory claims, nor do they dispute the evidence in detail, they dispute the claim that the returning officer and governor had responsibility, which is odd when you look who was responsible for compiling the list at the center of the claims and for distributing the wrong and illegal advice to the returning officers.
The other pillar of the dissent report is a regression analysis by a partisan statistician. As anyone in political science knows you can prove anything with regression analysis. Recently two reports on the subject of gun ownership were exposed as frauds, regression analysis does not show that gun ownership was rare in the US at the time the constitution was written, socre one for the NRA. Regression analysis does not show more guns less crime either score one for their opponents.
Regression analysis is a dangerous tool in political science because the number of factors influencing the data is large and the data available is small. Virtually any desired result can be obtained if you tweak your model.
The dissent is like one of those IHR rants that tries to claim that the Holocaust never happened. If you dismiss all the evidence you collect as anecdotal and play games with statistics you can claim black is white - which is exactly what the GOP did here.
Again, Bull. Read the USCCR report before you look like an ass. The report did not find voter fraud, it found votor "disenfranchisement".
You call it voter disenfranchisement, I call it electoral fraud.
Harris and Bush willfully disobeyed a rack of court orders that required voters with the right to vote to be allowed to register. They also selected ChoicePoint to scrub the voter lists despite the fact they were the most expensive by 25 times the cost of the cheapest bid.
I am not stupid enough to expect the Bush Administration to allow a report to be printed that concluded that GW was elected by means of a fraud concocted by his brother. However the facts found by the report validate the claim that the ballot was rigged in a massively partisan manner.
Do you have any documentation for this? I've never heard of that before, if its true that sure sounds like a smoking gun to me.
The allegation that Jeb Bush and Harris fixed the poll was confirmed by the commission. The guy who was unearthing this was Pallast. The US press would not publish him but the UK guardian, the BBC and the international press did. The US press only reported the story in July 2001 after the commission confirmed the evidence and Bush had been installed.
See the link in my sig for the first chapter of Pallast's book which documents the evidence at length.
You won't see the US media looking into this story or the Diebold story, they were bought off long ago. What is particularly sickening is the fact that these press barons like Maxwell and now Black that wield so much power so frequently turn out to be outright crooks. Black just got caught with his hands in the till, diverting $25 million from Hollinger without the knowledge of the board. He also diverted another $200 million with their approval.
Murodoch does not appear to be much different, he just installed his son to run Sky TV in the UK.
Slashdot Mirror
Actually they are not that much different from other enterprises. People simply cannot comprehend that the cost of going bespoke is vast.
The MIT folk very proudly told me how they built their own system to replace the IT functions my company provides. They only managed to spend twice as much on the project as it would have cost them to by the product of my most expensive competitor and about six times what I would have charged them.
I don't think that outsourcing to India is the biggest threat to IT jobs. Outsourcing via Web Services is. Most programmers work for IT integration shops churning out bespoke widgets for clients deploying SAP or Oracle Financials, Peoplesoft and the like. A great deal of that work is repetative glue logic and gets cobbled together using string and sealing wax coding styles. Its the sort of stuff that gives Perl a bad name.
I think that over the next ten years most enterprise computing will go the way it has gone in the hotel business - outsourced commodity product. When you check into a hotel the screen in front of the assistant manager is simply a terminal to an off site central mainframe. Same goes when you step into your bank, Springfield Coop Bank probably outsources all its IT needs to one of the bank tech companies.
Payroll has been outsourced for years, why not outsource all enterprise IT systems? Payroll, Accounts Payable, Accounts Receivable, HR, each company thinks it just has to do its own thing but in practice everyone works in almost exactly the same way.
When I was at MIT the rules were that overhead was charged at a rate of 2.75 times expenses. So if I hired someone as staff and paid them $1000 I would be charged $3750 from my budget.
The rules for students were somewhat different but still pretty grasping. Basically I would be charged the amount they were actually paid plus overhead and added to that their cost of tuition, I can't remember what overhead would be on that. Tuition at MIT these days is $29K per year. So over a year a student would cost my budget something like $60K, and the student would see less than $10K of that and I would see about 15 weeks worth of work if I was lucky.
There is also overhead on external contractors but nowhere near as much.
And yes, this is a complete stinking racket. The only reason it continues is that the government allows the major research universities to do this type of padding as a means of giving them an under the table subsidy.
I have no clue where the money goes. If you look at the amount of time that the students have contact with the faculty, the amount the faculty are paid and the cost of tuition the sums don't add up. And thats before you consider places like LCS/AI which have always been self funding through government grants. Perhaps the President has a yatch somewhere like the Stanford guy had, he would have to be a lot better hiding the thing round MIT though and if the students found it...
I am somewhat surprised that what MIT needed did not already exist as commercial off the shelf code. Their requirements are hardly very unusual, in fact since the content is not going to change much once it is put up there is not a great deal of difference between this site and any other web zine.
What this looks like to me is a boondoggle. $2 million is pretty easy to spend on software if you go bespoke. That is the main reason why most of the open source arguments you see on slashdot are bogus. If you can pay $100K for a product that is 90% complete you are one heck of a lot better off than you are paying $0 for a product that is 70% complete, maybe on a good day.
Open source is great provided it does exactly what you need if you have to do extensive programming then Gartner are completely right.
Building a system around Microsoft CMS is one heck of a lot better than mucking arround trying to make CVS do this type of thing. I don't have an issue with that part. But $2 million to customize it...
Incidentally MIT students are hideously expensive. The student may not get paid much, but the overhead charged by MIT is horrendous and the results can be 'variable' to say the least.
Exactly why should we care about this dispute?
If a bunch of folk get tweaked about using a content management system - a twenty year old technology used on the web for ten years...
Basically the group has forked and the dissident group had some links to their breakaway site. The SSC people then took out the links to stop people finding the breakaway site.
The breakaway site is also calling itself the LinuxGazette. This is probably actionable if there is a valid trademark registration. But at the end of the day it is not the name, its the audience.
I seem to remember that Bilbo was also in the Hobbit.
As for the age of Gandalf, since he holds one of the rings of power and is thus ageless there is effectively little difference in his age between the two books.
One of the issues that Jackson could probably address that would help the situation would be to actually pay a decent royalty for the rights to the Tolkein familly. JRRT originally sold the rights for GBP 100,000 to meet a tax bill. Jackson almost certainly paid rather more to purchase the rights from whoever was holding them, but the Tolkeins would not have seen any of that money.
People do not go to the press and announce that they have an issue with the division of the cash, much better to complain about artistic integrity or some such bollocks. You sound so much more principled and so much less of a whinner. Of course the familly has not exactly been doing badly from sales of the books recently.
New Line could easily afford a lump sum of $10 million or so ex-gracia. I suspect if they did someting like that a lot of the resistance would suddenly disappear.
Sooner or later the Hobbit and the Silmarilyon will get turned into films. Note that the BBC managed to get hold of the rights to do radio adaptations so the issue is not exactly forgone.
I don't think that the Silmarilion would be a single film either, it would be a series. The reason the negotiations are apt to be fraught is that everyone with some brains realises that this is potentially the start of a whole new film franchise which could ultimately rival the Bond series.
Progress is not going to end, but the automatic metronome of Moore's law will no longer be the driver. The rate of progress will slow for a while then start to pick up. The Intel paper says as much.
Incidentally the point of the paper seems to be to push out the end date and fend off rivals proposing the same ideas. The tunneling effect is quite definitely the end point of traditional logic gates. The astonishing part of the paper is that the end they cite is a 16nm process (with a 5nm gate), the smallest scale currently in use is 37nm. In other words there are only four more generations to go, two generations resulting in the feature size halving which means four times the number of transistors. So if the old two year schedule were kept Moore's law comes to an end in 2011.
They also point to the fact that Intel themselves have pushed out their dates for adopting new processes and are planning for three year gaps between generations. I have suspected that Intel has been the main factor in keeping the industry to the roadmap of Moore's law for some time.
Folk should note the difference between this argument and its usual populist invocation, that punishments deter crime. Most studies show that criminals respond to the threat of being caught. In other words it is not the length of the prison sentence that is the most effective deterrent, it is the likelyhood of a sentence being applied.
A Clockwork Orange is actually relevant to both side's arguments. The point that Burgess was making was that thuggery was out of control because the response from politicians was opportunistic. The first response is to impose harsh punishments, dehumanising Alex in a process that was certain to turn him into a worse criminal. Then when the winds of public opinion are swayed the other way the minister suddenly sees an opportunity to present Alex as the victim.
Bounties can be effective in limited circumstances. The guy who had Saddam's sons staying as houseguests seems to have decided that he would rather live in the US with $30 million. On the other hand they have not led to the capture of Bin Laden, Al Zawahiri or any of the top Al Qaeda leaders.
The problem with bounties is that hacker groups tend to be curiously diffuse. Hackers use multiple psuedonyms. They quite often turn rival groups in if they get the opportunity to do so even without bounties. Within a group all the members are implicated. It is difficult to turn in another member without risking prosecution yourself - particlarly if the other member cuts a deal with the prosecutors.
This would be a thin layer of synthetic diamond, not the mined type that deBeers has a monopoly in.
The fundamental limits are reached sooner in some technologies than others, but there is no technology that is immune from any sort of limit.
Even if there is an alternative technology the transition from silicon to a totally different substrate is something the industry has tried before and conspicuously failled at. There was a time when Galium Assenide was the bees knees, these days it is an important niche (direct band gap and all that) but nobody is building GaAs computers.
The other factor is that there seems to be a tradeoff between the point where you hit the quantum limit in a given technology and electron mobility that bites you in the a**.
I suspect that we see Moore's law start to slow before it comes to a halt.
They would probably be more interested in the political situation. They would be watching the throes of the Nixon impeachment crisis in the West, China would still be in the middle of the cultural revolution, Vietnam would have ended but only just. The USSR would be in mid collapse. Latin America is run by cliques of corrupt generals who murder tens of thousands (Pinochet) or hundreds of thousands (Argentina).
There has just been a war in the middle east. The Iranians are about to kick out the Shah (a brutal thug on a par with Saddam Hussein) and the Ayatolah would appear soon after to pervert the democratic revolution the same way Lenin appeared on the scene in Russia after the Tzar was deposed.
Things don't get any better for quite a while and they get worse before they get better. The nuclear arms race accelerates, the US and the USSR are engaged in a series of proxy wars that appear likely to turn nuclear. If you look at the situation from the outside even the events of 1989 might be considered evidence of further instability rather than a good sign.
On the whole I don't think that they are going to be avoiding talking to us just because of the disco music...
I think we should get our act together globally before we start to try to join extra-terrestrial clubs. If there is anyone out there worth talking to they already know about us.
Yeah Americans would NEVER pay $50 a month for cable TV.
I pay enough to watch TV without having my viewing interrupted by adverts. All that Tivo will cause is a small rejig of the revenue model.
I think we will see the end of the network TV model and the low end cable model. There will be fewer channels but what is left will be more like HBO than CBS.
The low end cable TV shows that do the bare minimum to get an audience will probably have difficulty finding people who pay for them.
The network TV model of spending five million dollars to create another episode of a formula TV series like Friends will also crash. Those guys are not that funny, they were not as funny as Seinfeld when he was still going and they have not got any funnier since Jerry stopped.
The thing I find fascinating about the US network TV model is how they spend so much money to achieve total mediocrity. For example what is it with summer re-runs? You have a sparse resource, network time. So you pay one guy $2 million an episode you will show repeatedly, why not get a second string guy in who will do a cheap episode for a mere $10K or so and have some variety? The thing about summer re-runs is that nobody is waiting for the new season of the X-Files, it never stopped.
Another wierd network TV choice is that they all run a carbon copy of the old Carlson late night TV show which is itself a copy of Bob Hope's radio act. Same presenter every night with the same formula.
What is somewhat more depressing is that in the 2000 election the late night TV shows were the only place where anyone asked any of the candidates a challenging question. Thats because the campaign managers have to work to get their candidate on those shows, with news shows they will pick the most favorable interviewer. It is somewhat sad that they even avoid Larry King's softball questions these days because he is considered to aggressive.
They can all go bust as far as I am concerned. So long as I can still see Norm Abrams (on the Porter Cable channel), Jon Stewart, Jessie James, pretty much all the rest of the junk can go to the hell it came from.
You are missing the point. The movie was not a mere product placement for the mini, the whole point of remaking the film was the new mini. The original mini was a slow seller until the original Michael Canne version of the Italian job came out. Then it was suddenly an overnight hit, the cool thing to be seen in in swinging London.
Without the car there is no movie. A remake of the Italian job with any other car would have been a flop. Its not like Bond who can drive a Bently or an Aston Martin, but lets face it everyone knew that it was pure product placement when he was driving the BMWs. The film had to return to the Astons because the product placement was damaging the credibility of the film.
Only if you count the military vote
In my experience there is nobody as cynical as an enlisted infantryman posted to a war zone. I very much doubt that the GOP will find as many enlisted men voting for them as did in 2000.
Popper's falsification principle will identify pseudo-science 100% of the time. The problem is that it gives a fals positive for real science most of the time if you use the original strict criteria. Popper himself admitted as such (see the OUP volume on Scientific revolutions where Popper responds to Kuhn). Popper originally developed his falsification principle as a reaction to the logical positivists their verification principle clearly did not work for induction. It is somewhat ironic that the two best known 'members' of the Vienna circle were never actually members. Wittgenstein refused to join and Schlick blackballed Popper because he denounced Wittgenstein's antics. Popper then used his falsification principle to mount an attack against the freudian psueuds and Marxism.
A test that does work is intent, is the theory intended to be subject to falsification? are the modifications of the theory to match observations made in good faith? Unfortunately in the process the test is no longer sharply objective.
One of the depressing aspects of US academia (which is spreading to the rest of the world) is the confusion of technique with method. When I was at MIT there were lots of folk who spent their time endlessly analysing the results of an intrinsically flawed experiment to five decimal places.
The big problem with the social sciences is physics envy. People want to be able to do the same type of science they think the physicists do, comming up with firm, immutable results.
The point I was making in the original piece is that the data is simply too thin and too unreliable to allow that type of analysis. There is no measurement in social science that is not contaminated by multiple effects.
The use of science in politics has been contaminated by the emergence of endless 'crank tanks' dedicated to providing the result their sponsors pay for. We have already seen the laughably named 'Independent' institute being paid by Microsoft to tell us that network effects do not exist. Until recently these institutes have mostly been right wing outfits funded by Richard Mellon Scaife.
Now that George Sorros is committed to spending equally large sums to attack the right one wonders what the outcome might be. Fortunately Sorros is one of Popper's students and he seems to have a personal commitment to the ideas Popper advanced. Still one has to wonder if the result might not end up being the left turning to the same junk science and propaganda that Scaife has pushed the right into.
I don't believe that happened, but I would not be surprised if a conspiracy theory of exactly that type started to spread through the Internet.
If a President lands in Iraq and nobody notices does he gain votes?
Probably for the same reason that two 15 year olds shoot up a school and folk appear on slashdot within a nanosecond explaining how gun control would not possibly have prevented the event.
Very little political debate in the US seems to ever be influenced by science, it is mostly predjudice and emotion.
Mind you, things are not necessarily that much better in science. Remember that AIDS/Polio vaccine connection that came up about ten years ago. Instead of checking a pretty strong prima facie case the establishment tried to quiet the issue - litterally in this case with a threat of libel proceedings. Thats not science.
Last year the stories had finaly percolated back to Nigeria and suddenly people were refusing the vaccine. Bad news when polio is inches from being erradicated. Betcha wondering why it took so long for the vaccine to make its way back to the people who were used as guinea pigs for testing, oh, well guess not.
So finally the science establishment gets panicky and does the tests that should have been done when the controversy started. They checked the remaining vials of vaccine from the tests to see if they could identify HIV DNA or money DNA. Turns out that the monkeys used to incubate the vaccine were a type that do not have a HIV strain and there was no HIV virus detected.
So the establishment got it right all along? Well not really, why didn't they insist on doing the test when the story first broke? The only logical reason to resist would be if you feared the result.
A while ago, before 9/11 I was sitting in a bizare meeting with a bunch of wireless execs who were breathlessly telling us how great their new location finder service was going to be. They could send you adverts targetted at people in a particular location.
I was rather unpopular when I asked if the customers would buy a product if the chief benefit was going to be to enable a new kind of spam. "Perhaps they don't get the choice"
I was even more unpopular when I pointed out that the regulators in Europe would blast this type of thing on privacy grounds. "Oh the regulators tend to be more sensible than the general public".
I pointed out that my cousin, one of those regulators has survived two assasination attempts and may have an opinion about a technology that gives away his position. In Europe privacy is not something that you muck arround with.
Today the risk of this type of scheme would be obvious even to a US legislator. Now right to life will be able to stalk doctors who provide abortions by telephone, Saddam loyalists will be able to stalk senior Republicans and Al Qaeda will be able to stalk everyone.
So they are finally working out socially acceptable ways to package up the same technology. Was it really necessary to have the dotCOM bust before some folk got a clue?
The GOP sent out an alert asking people to post responses. They have a group of activists who are paid to surf the Web at minimum wage and report any political speech by dissidents.
Someone should tell Rove to be a bit more carefull, not all the people he has working there are Republicans :-)
Well Boies is certainly wrong but he is not insane, just exceptionally greedy and prepared to make an argument he must know is shit. I was really pleased to hear that Boies is defending Conrad Black in his embezzlement case.
The source code availability issue is somewhat odd, it is from a pre-Internet age. There is no difficulty in getting hold of Linux source. The issue to argue over would be any modifications of Linux. If Linux is so good it should not be necessary to modify the kernel to run embedded. It is not clear that there isa modification here, a layered app is not a modifocation.
The other weak point if someone were to bring a case is standing. The only party with the right to bring a suit would be the copyright holder or holders. That is why RMS is keen on people signing over the copyrights of free software to the Richard Stallman Ego Enhancement Foundation, so that future generations can enjoy LiGNUx as it was meant to be known.
But Harris and Bush succeeded in keeping over a hundred thousand voters off the rolls, most of whom were illegally and illegitimately denied their right to vote. The fact that the errors were corrected in a small number of the cases in the report changes nothing. Those two witnesses were called to show the lengtrhs that Harris and Bush went to deny their rights.
I took a look at Workindevs posting history. A lot of posts on electronic voting, all trying to convince us that there is no need to worry dears, those nice republicans are 1000% trustworthy and none of them would ever tell a lie, the WMD will eventually be found, the tax cuts are not almost exclusively for the rich and Nixon was not a crook.
Oh come on, Dean got a medical degree. I don't think you get onto those courses through the 'positive discrimination' that Yale practices to favor 'legacies'. I certainly don't think you can get a medical degree and practice as a doctor if you don't have the brains and work at it.
Dean got into full time politics in by accident. He was the leuitenant governor, a part time job in Vermont and became governor when the sitting governor suddenly died in office. He was a successful governor for over a decade.
The odd thing about Texas politics is that the Lieutenant Governor actually has more power than the governor. It is a legacy from a time when they had a corruption issue with the then incumbent.
Kucinich OTOH is a card carrying union member from a working class background.
Which is great. But that is not enough for a platform.
Kucinich has the luxury of knowing that he is not going to get nominated. So he does not have to worry about advancing politically costly ideas with a marginal return.
Ah so a dissenting statement by a group of Republicans proves that there was no electoral fraud!
Diebold say that their machines are completely reliable. So I guess that settles the matter, NO NEED TO WORRY DEARS.
That is it, sleep tight, the GOP tells us that they do not rig the polls, so we should just believe them. If anyone claims anything different THEY MUST BE A GODLESS COMMUNIST so anything they say can be ignored.
BTW the dissent disputes the findings by rejecting all the testimony and facts found as 'anecdotal' they provide no contradictory claims, nor do they dispute the evidence in detail, they dispute the claim that the returning officer and governor had responsibility, which is odd when you look who was responsible for compiling the list at the center of the claims and for distributing the wrong and illegal advice to the returning officers.
The other pillar of the dissent report is a regression analysis by a partisan statistician. As anyone in political science knows you can prove anything with regression analysis. Recently two reports on the subject of gun ownership were exposed as frauds, regression analysis does not show that gun ownership was rare in the US at the time the constitution was written, socre one for the NRA. Regression analysis does not show more guns less crime either score one for their opponents.
Regression analysis is a dangerous tool in political science because the number of factors influencing the data is large and the data available is small. Virtually any desired result can be obtained if you tweak your model.
The dissent is like one of those IHR rants that tries to claim that the Holocaust never happened. If you dismiss all the evidence you collect as anecdotal and play games with statistics you can claim black is white - which is exactly what the GOP did here.
You call it voter disenfranchisement, I call it electoral fraud.
Harris and Bush willfully disobeyed a rack of court orders that required voters with the right to vote to be allowed to register. They also selected ChoicePoint to scrub the voter lists despite the fact they were the most expensive by 25 times the cost of the cheapest bid.
I am not stupid enough to expect the Bush Administration to allow a report to be printed that concluded that GW was elected by means of a fraud concocted by his brother. However the facts found by the report validate the claim that the ballot was rigged in a massively partisan manner.
This is clearly a worthwhile contribution to world culture.
The allegation that Jeb Bush and Harris fixed the poll was confirmed by the commission. The guy who was unearthing this was Pallast. The US press would not publish him but the UK guardian, the BBC and the international press did. The US press only reported the story in July 2001 after the commission confirmed the evidence and Bush had been installed.
See the link in my sig for the first chapter of Pallast's book which documents the evidence at length.
You won't see the US media looking into this story or the Diebold story, they were bought off long ago. What is particularly sickening is the fact that these press barons like Maxwell and now Black that wield so much power so frequently turn out to be outright crooks. Black just got caught with his hands in the till, diverting $25 million from Hollinger without the knowledge of the board. He also diverted another $200 million with their approval.
Murodoch does not appear to be much different, he just installed his son to run Sky TV in the UK.