I wrote them a nice note, complimenting them on the site design, navigation, organization, etc; I was even encouraging about the selection, saying that while they didn't have everything, they had more than enough to encourage at least $100 a year out of me.
Then I said, "but of course, since you only support Windows and I have a Linux box and an MP3-only portable player, I won't be giving you any business at all"...and how sorry I was about that.
A search on the word "door" through the postings so far would seem to say that nobody on all Slashdot has mentioned that the robot vacuum was the primary invention of the protagonist in Heinlein's "Door Into Summer".
Actually, that guy's "Hired Girl" product would clean any floor, switching from vacuuming to floor polishing to mopping:
"I swiped the basic prowl pattern from the 'Electric Turtles' written up in Scientific American in the late forties, lifted a memory circuit out of the brain of a guided missile (that's the nice thing about top-secret gimmicks; they don't get patented), and I took the cleaning devices and linkages out of a dozen things..."
The year of this accomplishment was about 1970, so the prediction was for about 14 years after the book's writing in 1956. The book's plot has the betrayed and ruined inventor fleeing his troubles via suspended animation that takes him to the vastly more advanced year of 2000.
Alas, the real 2003 doesn't have half the things he (fictionally) predicted for 1970.
Warping the (Fictional) Science to Make the Story
on
Ask Larry Niven
·
· Score: 1
It's obvious in a lot of science fiction that the central SF supposition (i.e. "FTL is possible") is tailored around the story, rather than the author making a major supposition and letting the story fly where it may.
This reaches a painful level in some Star Trek episodes - that dratted transporter AGAIN on the fritz because the use of it would remove the whole reason for the landing party's distress.
There was a frank discussion of this, I believe from Dr. Pournelle(?) about "The Mote in God's Eye". To the effect that you wanted a spaceship similar in personal dynamics and crew composition to a Hornblower-era naval vessel; so the particular FTL drive required long travels through normal space before it could be used, and the properties of the Langston Field meant slow, harrowing, battles of attrition, partial damage to each ship so that redundant crew were required to replace casualties, etc.
Have you ever had an SF supposition that really intrigued you as an idea worth pursuing, but couldn't think of a story to wrap around it? (At least, not unless you imposed quirks and limits on the idea that made it uninteresting to you?)
SUN more secure at the hardware level?
on
The Faded Sun
·
· Score: 2, Informative
A bunch of us OpenBSD enthusiasts at lunch the other day were listening to the guy who got most of us interested in it, explain the value of OpenBSD's recent efforts to close off executability of whole areas of memory, including the stack.
He mentioned that SUN hardware allows individual memory pages to be marked "non-executable" by the hardware, wheras Intel hardware can only select on "line in the sand" of all memory space below which the processor won't execute.
I believe he also mentioned that this is just ONE of many ways in which SUN hardware is more appropriate for really secure and reliable computing - the kind you want on those $1M servers that big corporations buy.
Anybody care to list off other reasons why SUN's hardware is more trusted for the enterprise servers? Price/Performance is nice, but it isn't everything...
The Stephenson "Hacker Tourist" would be interested in the heaviest examples of all of society's infrastructures. All your examples were about transportation infrastructures, and Stephenson did telecom.
There's still:
Power (Oil wells, Coal Mines, power plants, NUCLEAR power plants, major switch stations)
Water & Wastewater (Plants)
Food (Farms, Orchards, Ranches, various factories)...maybe these aren't as sexy technologies as telecom and transportation, but they're older, more basic, more necessary.
I've been reading about it - and even seen TV specials about this and very similar concepts for over 20 years. I recall being very excited by an Analog SF story about "ITTS Loops" more like 30 years ago.
I think I know enough about it.
You are certainly correct that it has a vast capacity. And if you know someplace with a very short digging distance (lower cost) and very, very high passenger levels (higher payback) then that's where you should start. I'd suggest somewhere around Manhattan or Paris or Tokyo.
But, alas, NY to LA is not only a very, very, very long digging distance with a staggering cost, it doesn't happen to NEED 80 lanes-of-freeway of capacity to connect the two....I was even starting to google up numbers for what it does need (1.7 million tons of NY->LA cargo in 1997, at http://www.bts.gov/ntda/cfs/docs/cf97odt.xls , plus orbits finding me about 107 flights per day from NY to LA, maybe 25,000 passengers if all full...when it hit me this is silly.
A system with huge capital-construction costs will only get started, as I said above, in places close together with high traffic volumes. Since that's where the first one will go in, post on your site the travel time up and down the "Boswash Corridor", not the big jump from NY to LA. (450 miles, 40 million people), not the time for one of the last runs you'll put in.
I've forgotten the title and author of the Analog story about "ITTS loops" but I remember they were loops because VERY deep digging allowed the train to not have a high acceleration sensation as they went up to vast speeds, because they were also dropping into the earth...basically, they FELL up to speed, leveled off, then decelerated again as they rose up the other end of the tunnel.
If you can't dig that deep (and you can't), but can get people to put up with 0.25G acceleration, you can top out at 1000MPH and go 100 miles in 10 minutes - two thirds of it spent accelerating and decelerating. The full Boston-Washington trip would be just over 25 minutes. That would get you all the car, jet, and cargo traffic business for 40 million people, with one-fifth of the NY->LA dig.
I just got annoyed by NY-LA being the chosen time example at the web site.
Everybody's on about keeping the vacuum, but even before you get to that large problem, you have to dig a lot of tunnel.
A little googling from the monorail story ($124 million/mile) found a subway that ran $224 million per mile. The Channel Tunnel, surprisingly, was no more: I think it must be the savings of scale from digging a lot of it, with no stations for 30 miles, compared to urban subways with a stop every mile or two. (Total: $21B US, for 3 parallel tunnels each 30 miles long).
The proposal requires 2 parallel tunnels, so it would run about $450 million per mile.
It's obviously only of real use for distance travel: you couldn't accelerate to 350 MPH while moving about an urban centre in short hops. Connecting two cities, say 50 miles apart would run you 22.5 billion dollars. That "NY to LA" they go on about would be:
2420 miles X $450 million/mile = $1.1 trillion.
Competing system: 2 new airports the cost of Denver's: $10 billion Fleet of 44 Boeing 777's @ $230M: $10 billion
That allows departures each way every 15 minutes.
Yes, yes, it takes 4.5 hours, costs way more in energy, etc. But the 98% reduction in capital cost means the trip would be much, much cheaper.
It's a reductio ad absurdum(sp?) argument: what I've just described (an airport at each end for just ONE destination) and a dedicated fleet, is VERY, VERY expensive flying. So if the tube is so much worse, how pricey would it be?
If the tube had no operational or maintainance costs, creating the vacuum were free, and lasted forever, you'd still have to pay off the 5% interest on 1.1 trillion per year. That's 55 billion dollars. If you could manage 55 million passengers per year, (that's a couple of round trips for every inhabitant of NY and LA), it would still be $1000 per trip. Each way.
Who's going to pay that for the time reduction from 4.5 hours to 45 minutes? I guess the same crowd that fly the Concorde. It's not a large crowd. They keep a few planes flying, but they aren't going to make 55 million trips per year.
Sorry.
THEN you can start toting up the cost of putting it into a vacuum...
I'm all for it, as a concept. But we need to invent a way of making tunnels that's cheaper by at least an order of magnitude.
Well, certainly, the HURD is a low-priority now, >10 years after the Linux kernel was done.
But, 12 years ago, why did they work only on support programs of various sorts and so slowly on the central One Ring to Rule them All? It's as if a group of talented mechanics puttered away on a car project and perfected the glove compartment and brakes and reclining seats while putting in no time on the *engine*......and Torvald's project showed that one keener could have knocked off "rough consensus and running code" in a man-year.
So, OK, the better-designed and over-designed HURD would presumably have taken 3 man-years, but I don't get why there weren't two guys working on it from 1988 to 1990 and done before Mr. Torvalds.
I think everybody understands that this is a bit like the Panama Canal -- virtually all 6 million cubic yards of earth excavated by the French; then the Americans came in when the project was in collapse and finished up the last 10% of the work, and reaped nearly 100% of the reward.
Can anybody explain in words of one syllable why the crucial closing of the loop, the final cherry on top that made the whole thing work, the kernel itself, eluded the GNU programmers? If they were such amazingly great coders, why was Mr. Torvalds able to do what they didn't? You'd think they'd have done that *first*...
I've volunteered for my local UUG for some years, and work at a *large* employer that predictably went all-Microsoft (except on big servers) and would distill my advice down to this:
1) Advocate Linux servers at work, stressing, when challenged, that, if nothing else, their mere presence will have a good effect on the cost of the MS contract. In the old days, they spoke of the "million dollar Amdahl mug". That is, drinking from it while negotiating your IBM mainframe contract let the IBM guys know you were shopping around for clone mainframes, that you weren't locked in. Hopefully, we can speak of the "million dollar Tux doll" some day...
2) Push Linux desktops in places where every cent counts: friends, family, non-profits you volunteer at, your church, and your school. (Oh, Lord, just give us the schools and you can keep your Fortune 500.) And you can best do this (or, really, ONLY do this) by *volunteering* to install and admin the machines yourself.
That last one takes commitment, but that's what the man's asking for.
Pushing the Linux desktop at work is wasted effort; but if we can build some exposure and mindshare in the non-business world, it's a foot in the door. The guy who wrote about "word of mouth" was dead on. Every single Linux desktop that you get into play will cause 10 people to see it and hear about it.
If this thing can record HDTV, it'll have a market. Indeed, if it can just record the now-existing digital cable and satellite signals (several million customers now, I think) at their 450-ish lines of resolution, I'd buy one as soon as it drops below about $900.
I got a "Super-VHS" recorder, had an awful time with it, finally chalked it to experience and got the best one on the market. It's pretty decent, but still doesn't actually get the whole 450 lines, and tends to add some noise and mutate the colours a touch.
D-VHS sounds like it would be able to reproduce the satellite/dig-cable signal perfectly, since it will actually do over 4X that bandwidth...and as HDTV signals become more common, you'll be able to time-shift and save movies, unless they carry out the threat to embed signals the VCR will obey, to not copy.
Which they'll eventually find will screw both this tape format, and HDTV itself.
If your goal is to have some effect, you'll go farther by not worrying about the exact truth and say whatever has the most punch.
RMS screed is *bad* because it runs on and on and on - the more words, the more dilute their effect. Pick ONE point.
AT HOME: "Sorry, I can't afford Word at home, it's $300, please 'save as text' or 'save as HTML', or just cut & paste the text.
AT WORK, MESSAGE FROM OUTSIDE: "Sorry, Corporate Policy prohibits opening Word files since we got hit by a virus from one." (plus the 'save as' stuff). This one's great, they'll never send another.DOC to the whole corp.
AT WORK, MESSAGE INSIDE: (ok, here you're nailed. They KNOW you have Word, and that there's no corp. policy)
"Why are you attaching a file? Just put it on the shared disk, and save storage space".
(OK, that doesn't stop Word use, but that battle's lost at most corps anyway. It does stop the stupid practice of attaching at all, when there's a shared disk.)
PS: Don't you especially hate it when you open the.DOC and find it was plain text with no formatting anyway?
What's absolutely crucial to Microsoft is that the news be delayed UNTIL THE FIX IS ISSUED for P.R. reasons. Now note that EVERY news story has "a fix is available" in the lead paragraph. In half of them, it is in the opening sentence.
This gives a second meaning to the term "damage control".
My rough guess is that non-commercial organizations - academia, informal hacker clubs, private individuals, etc - would have released this at least a week or two ago. Really, isn't a month with the hole wide open and nobody even aware of the vulnerability more than long enough?
But by holding back until Microsoft was prepared to handle the media, preferably writing the press release itself - a commercial firm will be owed a big favour by Microsoft.
It is GOOD to be owed a big favour by Microsoft when you have a profit-making IT business to run...
Pass some laws making it an offense to be egregiously insecure, on the grounds that you have made yourself part of the problem, a menace to others on the public network.
If you're wide open to becoming a siteful of zombies to be used in DDOS, it's like leaving a gun unsecured - on your front lawn.
Far from costing budget money, the fines levied will be a revenue source. And the fear of the fines and the shame of the criminal charge will spur pointy-haired bosses into Getting Serious about security in a way that some tiny tax break never will.
It's a question of SQUARE footage
on
This is IT?
·
· Score: 1
Hey, the/. crowd that knows Metcalfe's Law by heart is missing another simple square law in the comments so far: square footage.
Perhaps as many as 5% of downtown commuters today come by walking or biking, because they actually live in downtown apartments.
If this thing triples the radius outward from downtown that is "walkable", then it increases the number that can come without a car by NINE fold, no? That's a simplistic computation, but the LONG term effect will be to make more apartment buildings go up further away from downtown because you'll be able to "walk to work".
Oh, and the other "square footage" issue only a few have alluded to is that these will probably be allowed in public transit (in my Calgary hometown, bikes are only allowed on the trains, and only off-peak hours), thus increasing public transit usage, and will surely be allowed into your cubicle/office, (doubles as your coatrack), so the "stolen bike" issue will be reduced.
It'll certainly succeed in "medical need" and niche roles - my Dad who needs a hip replacement will get one for sure - he doesn't need a wheelchair but any walk over 200 yards is torture. And we'll see what else it does after that!
You aren't really specific about whether you most want to help in places in need (in which, yes, you'll surely have to pick some other contribution than I.T.) or just go abroad with the talents you've got.
Many commentators say the best bang for the 1st world's development bucks will come from helping nations that are hurting but are also developing - to help tip the balance towards upward progress.
In which case, I think you don't even have to travel far. Learn Spanish and do some travelling south of the Rio Grande and I think you'll find Mexico and other latin american countries are in endless need of teachers and NGO office volunteers that know their way around a computer.
With Linux, customers "end up being in the operating systems business," managing software updates and security patches while making sure the multitude of software packages don't conflict with each other," Miller said. "That's the job of a software vendor like Microsoft."
==============
In short, the REAL headline for this story is:
"Microsoft admits is was THEIR job, not customers' job, to apply all those IIS security patches, and thus admits responsibility for all damage from the Code Red and Nimda worms."
Microsoft's compensation cheques are presumably coming out next week.
A lot of this argument sounds like VHS vs Beta 20 years back. You can go with the popular one, or the slightly-better one. (Which is *VASTLY* better in the opinion of people who generally admit to being picky, picky, picky.)
On the other hand, if you admit your listening tends to be mostly casual (background while you work or party) and it's quite rare that you sit still in an otherwise quiet room, intently listening with your eyes closed to great speakers and equipment, then for God's sake, go with "popular". Nothing's more embarrassing than somebody carefully buying the best caviar, then eating it with mustard on Wonderbread.
I have 3000 MP3 files ripped at 128 from my own collection, the library, etc, used for background music, working at the computer, (noisy) parties, and my portable MP3 player that would only get a half hour if I used a high bit-rate. They're fine for those purposes.
If I want a better listening experience, it has to come from a quiet living room and my full stereo - in which case I have or can get the original CD!
Wasn't there a whole other variable, as well? The material of the coin?
I recall that guineas weren't just an extra shilling compared to a pound, they were GOLD - and professionals & gentlemen took payment in gold, by preference.
And aren't you missing a unit? Sherlock Holmes stories keep mentioning a "sovereign" and I think that was a gold coin, too...but how many shillings was it worth?
The Romance of the Slide Rule is very much alive today, the love affair is just transferred to the modern version of the tool.
Slashdotters and Tom's Hardware afficionados who explore every capacity and foible of their machines are following the same urge as slipstick experts who practiced use of the LL/0 scale (which did an e^(-x) operation in one move) just so they were prepared should it come up one day.
I must be almost the same age as "timothy" who started this article - it was slide rules for us up to about grade 12, when the richer kids were all getting calculators that did more than 4 functions...then in University, calculators were suddenly essential (but I carried the Rule to exams in case of battery failure).
What I remember is that most geeky kids were old enough to *learn* to use a slide rule at the start of their teens...right around the time you get coordinated & generally grown-up enough to be allowed to touch precision tools. And slide rules, with their fine machinining, high cost, and smooth movement (metal sucked; bamboo rules...mine still work perfectly after 25 years in the drawer) were clearly *TOOLS*, not toys.
A slide rule was a grown-up possession, a minor Rite of Passage.
I don't think slide rules will ever be forgotten as long as dad's, grandad's and great-granddad's get found in attics. But what worries me is my Dad also taught me dozens of MENTAL calculation tricks that nobody needs any more.
The easiest one was squaring number that ends in five. Chop off the five. Multiply the rest by one greater than itself; tack 25 on the end. (i.e. 35 squared: 3*4=12, the answer is 1225.) The hardest was memorizing ten anti-logs to three places; you can do remarkable estimation-level calcs with that one. (I've long forgotten them.)
I've been working in municipal GIS for several years. The very first thing every consultant told us, was also the first thing you hear from every organization that's run big GIS projects: the technology is easy; the data is hard.
GIS just involves a lot of data - and a lot of problems cross-referencing separate but related datasets. In our case, its the (say) water, sewer, parks, roads and legal maps (and 22 others) that are all drafted separately but must register on top of one another correctly to create synergy.
I've gone from one elegant-but-buggy GIS product (VISION*) to just doing "GIS" by connecting my CAD maps (Microstation) to databases, to a second run at GIS with the Microsoft of the GIS world (ESRI). And they were right; the technology was never my big problem. Staying on top of the ever-mutating dataset is 98% of the work.
I see GIS having big effects in the government and industrial world but almost none at the consumer level, because you have to be a big organization to manage enough *data* to need a GIS to understand it all.
There will be consumer products and services - your PDA will show you the direction to the nearest Thai restaurant - but the only technology you'll need to get this from some remote GIS engine will be a browser.
Oh, and PS, that bodes ill for Open Source GIS software outside the academic world, because big organizations have a positive fondness for the "Microsofts" of any industry they buy from, and an aversion to "unsupported" products. Hate to be the messenger, but its just a corporate-culture fact of life.
I don't have the Japan-knowledge creds of others saying this, but even the most cursory knowledge of the island argues against it.
Skip that they are a "feudal" culture of loyalty to The Company, with a near-permanent ruling party, etc... that stuff can change. They've made bigger changes. (Playing catch-up to make them...hardly "clicks down the road".)
No, let's look at deeply-held cultural traits that aren't about to go away. They are genetically and culturally homogenous. Differences from the mainstream are not well-liked. Foreigners are not popular.
By the way, in this, they are NOT like England, which is an often-conquered mish-mash of natives, North Europeans, Norman French, and a Roman gene or twelve. (And the English language reflects it.)
And "The Future" is surely one of increasing diversity and inter-mingling. The plummeting cost of travel and the advantages of trade demand it.
If Japan even joins that future, much less makes it, they'll have to radically change deeply-held attitudes to do so.
Slashdot Mirror
At Sun's site
u nf lash.20031117.3.html
http://www.sun.com/smi/Press/sunflash/2003-11/s
The article says "Java Desktop" and "interoperable with Linux, Solaris, and Windows".
So, whoa -- China is going ALL-JAVA ?
I wrote them a nice note, complimenting them on the site design, navigation, organization, etc; I was even encouraging about the selection, saying that while they didn't have everything, they had more than enough to encourage at least $100 a year out of me.
Then I said, "but of course, since you only support Windows and I have a Linux box and an MP3-only portable player, I won't be giving you any business at all"...and how sorry I was about that.
Best I could do....
A search on the word "door" through the postings so far would seem to say that nobody on all Slashdot has mentioned that the robot vacuum was the primary invention of the protagonist in Heinlein's "Door Into Summer".
Actually, that guy's "Hired Girl" product would clean any floor, switching from vacuuming to floor polishing to mopping:
"I swiped the basic prowl pattern from the 'Electric Turtles' written up in Scientific American in the late forties, lifted a memory circuit out of the brain of a guided missile (that's the nice thing about top-secret gimmicks; they don't get patented), and I took the cleaning devices and linkages out of a dozen things..."
The year of this accomplishment was about 1970, so the prediction was for about 14 years after the book's writing in 1956. The book's plot has the betrayed and ruined inventor fleeing his troubles via suspended animation that takes him to the vastly more advanced year of 2000.
Alas, the real 2003 doesn't have half the things he (fictionally) predicted for 1970.
It's obvious in a lot of science fiction that the central SF supposition (i.e. "FTL is possible") is tailored around the story, rather than the author making a major supposition and letting the story fly where it may.
This reaches a painful level in some Star Trek episodes - that dratted transporter AGAIN on the fritz because the use of it would remove the whole reason for the landing party's distress.
There was a frank discussion of this, I believe from Dr. Pournelle(?) about "The Mote in God's Eye". To the effect that you wanted a spaceship similar in personal dynamics and crew composition to a Hornblower-era naval vessel; so the particular FTL drive required long travels through normal space before it could be used, and the properties of the Langston Field meant slow, harrowing, battles of attrition, partial damage to each ship so that redundant crew were required to replace casualties, etc.
Have you ever had an SF supposition that really intrigued you as an idea worth pursuing, but couldn't think of a story to wrap around it? (At least, not unless you imposed quirks and limits on the idea that made it uninteresting to you?)
A bunch of us OpenBSD enthusiasts at lunch the other day were listening to the guy who got most of us interested in it, explain the value of OpenBSD's recent efforts to close off executability of whole areas of memory, including the stack.
He mentioned that SUN hardware allows individual memory pages to be marked "non-executable" by the hardware, wheras Intel hardware can only select on "line in the sand" of all memory space below which the processor won't execute.
I believe he also mentioned that this is just ONE of many ways in which SUN hardware is more appropriate for really secure and reliable computing - the kind you want on those $1M servers that big corporations buy.
Anybody care to list off other reasons why SUN's hardware is more trusted for the enterprise servers? Price/Performance is nice, but it isn't everything...
"Walking away from $404K" ?? The poor MIT accountants were probably already counting on it.
They're just walking the halls in a daze today, muttering, "404 not found..."
The Stephenson "Hacker Tourist" would be interested in the heaviest examples of all of society's infrastructures. All your examples were about transportation infrastructures, and Stephenson did telecom.
...maybe these aren't as sexy technologies as telecom and transportation, but they're older, more basic, more necessary.
There's still:
Power (Oil wells, Coal Mines, power plants, NUCLEAR power plants, major switch stations)
Water & Wastewater (Plants)
Food (Farms, Orchards, Ranches, various factories)
You know, it just hit me that nobody seems to be talking about the cost. The ET3 web site just says the cost will be "low".
1) What cost do you predict per mile of tunnelling?
2) What cost do you predict for the final, air-tight system, per mile?
It would involve the assumption of some new tunnelling technique to come in much below the chunnel cost of $200+ million per mile.
As to what airtight tunnel costs, there's no good precedent, but if you're studying the matter, presumably you have a number in mind...
I've been reading about it - and even seen TV specials about this and very similar concepts for over 20 years. I recall being very excited by an Analog SF story about "ITTS Loops" more like 30 years ago.
...I was even starting to google up numbers for what it does need (1.7 million tons of NY->LA cargo in 1997, at http://www.bts.gov/ntda/cfs/docs/cf97odt.xls ,
I think I know enough about it.
You are certainly correct that it has a vast capacity. And if you know someplace with a very short digging distance (lower cost) and very, very high passenger levels (higher payback) then that's where you should start. I'd suggest somewhere around Manhattan or Paris or Tokyo.
But, alas, NY to LA is not only a very, very, very long digging distance with a staggering cost, it doesn't happen to NEED 80 lanes-of-freeway of capacity to connect the two.
plus orbits finding me about 107 flights per day from NY to LA, maybe 25,000 passengers if all full...when it hit me this is silly.
A system with huge capital-construction costs will only get started, as I said above, in places close together with high traffic volumes. Since that's where the first one will go in, post on your site the travel time up and down the "Boswash Corridor", not the big jump from NY to LA. (450 miles, 40 million people), not the time for one of the last runs you'll put in.
I've forgotten the title and author of the Analog story about "ITTS loops" but I remember they were loops because VERY deep digging allowed the train to not have a high acceleration sensation as they went up to vast speeds, because they were also dropping into the earth...basically, they FELL up to speed, leveled off, then decelerated again as they rose up the other end of the tunnel.
If you can't dig that deep (and you can't), but can get people to put up with 0.25G acceleration, you can top out at 1000MPH and go 100 miles in 10 minutes - two thirds of it spent accelerating and decelerating. The full Boston-Washington trip would be just over 25 minutes. That would get you all the car, jet, and cargo traffic business for 40 million people, with one-fifth of the NY->LA dig.
I just got annoyed by NY-LA being the chosen time example at the web site.
Excuse me.
Everybody's on about keeping the vacuum, but even before you get to that large problem, you have to dig a lot of tunnel.
A little googling from the monorail story ($124 million/mile) found a subway that ran $224 million per mile. The Channel Tunnel, surprisingly, was no more: I think it must be the savings of scale from digging a lot of it, with no stations for 30 miles, compared to urban subways with a stop every mile or two. (Total: $21B US, for 3 parallel tunnels each 30 miles long).
The proposal requires 2 parallel tunnels, so it would run about $450 million per mile.
It's obviously only of real use for distance travel: you couldn't accelerate to 350 MPH while moving about an urban centre in short hops. Connecting two cities, say 50 miles apart would run you 22.5 billion dollars. That "NY to LA" they go on about would be:
2420 miles X $450 million/mile = $1.1 trillion.
Competing system:
2 new airports the cost of Denver's: $10 billion
Fleet of 44 Boeing 777's @ $230M: $10 billion
That allows departures each way every 15 minutes.
Yes, yes, it takes 4.5 hours, costs way more in energy, etc. But the 98% reduction in capital cost means the trip would be much, much cheaper.
It's a reductio ad absurdum(sp?) argument: what I've just described (an airport at each end for just ONE destination) and a dedicated fleet, is VERY, VERY expensive flying. So if the tube is so much worse, how pricey would it be?
If the tube had no operational or maintainance costs, creating the vacuum were free, and lasted forever, you'd still have to pay off the 5% interest on 1.1 trillion per year. That's 55 billion dollars. If you could manage 55 million passengers per year, (that's a couple of round trips for every inhabitant of NY and LA), it would still be $1000 per trip. Each way.
Who's going to pay that for the time reduction from 4.5 hours to 45 minutes? I guess the same crowd that fly the Concorde. It's not a large crowd. They keep a few planes flying, but they aren't going to make 55 million trips per year.
Sorry.
THEN you can start toting up the cost of putting it into a vacuum...
I'm all for it, as a concept. But we need to invent a way of making tunnels that's cheaper by at least an order of magnitude.
Or Two.
Well, certainly, the HURD is a low-priority now, >10 years after the Linux kernel was done.
...and Torvald's project showed that one keener could have knocked off "rough consensus and running code" in a man-year.
But, 12 years ago, why did they work only on support programs of various sorts and so slowly on the central One Ring to Rule them All? It's as if a group of talented mechanics puttered away on a car project and perfected the glove compartment and brakes and reclining seats while putting in no time on the *engine*...
So, OK, the better-designed and over-designed HURD would presumably have taken 3 man-years, but I don't get why there weren't two guys working on it from 1988 to 1990 and done before Mr. Torvalds.
I think everybody understands that this is a bit like the Panama Canal -- virtually all 6 million cubic yards of earth excavated by the French; then the Americans came in when the project was in collapse and finished up the last 10% of the work, and reaped nearly 100% of the reward.
Can anybody explain in words of one syllable why the crucial closing of the loop, the final cherry on top that made the whole thing work, the kernel itself, eluded the GNU programmers? If they were such amazingly great coders, why was Mr. Torvalds able to do what they didn't? You'd think they'd have done that *first*...
I've volunteered for my local UUG for some years, and work at a *large* employer that predictably went all-Microsoft (except on big servers) and would distill my advice down to this:
1) Advocate Linux servers at work, stressing, when challenged, that, if nothing else, their mere presence will have a good effect on the cost of the MS contract. In the old days, they spoke of the "million dollar Amdahl mug". That is, drinking from it while negotiating your IBM mainframe contract let the IBM guys know you were shopping around for clone mainframes, that you weren't locked in. Hopefully, we can speak of the "million dollar Tux doll" some day...
2) Push Linux desktops in places where every cent counts: friends, family, non-profits you volunteer at, your church, and your school. (Oh, Lord, just give us the schools and you can keep your Fortune 500.) And you can best do this (or, really, ONLY do this) by *volunteering* to install and admin the machines yourself.
That last one takes commitment, but that's what the man's asking for.
Pushing the Linux desktop at work is wasted effort; but if we can build some exposure and mindshare in the non-business world, it's a foot in the door. The guy who wrote about "word of mouth" was dead on. Every single Linux desktop that you get into play will cause 10 people to see it and hear about it.
If this thing can record HDTV, it'll have a market. Indeed, if it can just record the now-existing digital cable and satellite signals (several million customers now, I think) at their 450-ish lines of resolution, I'd buy one as soon as it drops below about $900.
I got a "Super-VHS" recorder, had an awful time with it, finally chalked it to experience and got the best one on the market. It's pretty decent, but still doesn't actually get the whole 450 lines, and tends to add some noise and mutate the colours a touch.
D-VHS sounds like it would be able to reproduce the satellite/dig-cable signal perfectly, since it will actually do over 4X that bandwidth...and as HDTV signals become more common, you'll be able to time-shift and save movies, unless they carry out the threat to embed signals the VCR will obey, to not copy.
Which they'll eventually find will screw both this tape format, and HDTV itself.
People WANT to record. Period.
What a bunch of touchingly honest people!
.DOC to the whole corp.
.DOC and find it was plain text with no formatting anyway?
If your goal is to have some effect, you'll go farther by not worrying about the exact truth and say whatever has the most punch.
RMS screed is *bad* because it runs on and on and on - the more words, the more dilute their effect. Pick ONE point.
AT HOME: "Sorry, I can't afford Word at home, it's $300, please 'save as text' or 'save as HTML', or just cut & paste the text.
AT WORK, MESSAGE FROM OUTSIDE: "Sorry, Corporate Policy prohibits opening Word files since we got hit by a virus from one." (plus the 'save as' stuff). This one's great, they'll never send another
AT WORK, MESSAGE INSIDE: (ok, here you're nailed. They KNOW you have Word, and that there's no corp. policy)
"Why are you attaching a file? Just put it on the shared disk, and save storage space".
(OK, that doesn't stop Word use, but that battle's lost at most corps anyway. It does stop the stupid practice of attaching at all, when there's a shared disk.)
PS: Don't you especially hate it when you open the
This gives a second meaning to the term "damage control".
My rough guess is that non-commercial organizations - academia, informal hacker clubs, private individuals, etc - would have released this at least a week or two ago. Really, isn't a month with the hole wide open and nobody even aware of the vulnerability more than long enough?
But by holding back until Microsoft was prepared to handle the media, preferably writing the press release itself - a commercial firm will be owed a big favour by Microsoft.
It is GOOD to be owed a big favour by Microsoft when you have a profit-making IT business to run...
...then they can use the stick.
Pass some laws making it an offense to be egregiously insecure, on the grounds that you have made yourself part of the problem, a menace to others on the public network.
If you're wide open to becoming a siteful of zombies to be used in DDOS, it's like leaving a gun unsecured - on your front lawn.
Far from costing budget money, the fines levied will be a revenue source. And the fear of the fines and the shame of the criminal charge will spur pointy-haired bosses into Getting Serious about security in a way that some tiny tax break never will.
Hey, the /. crowd that knows Metcalfe's Law by heart is missing another simple square law in the comments so far: square footage.
Perhaps as many as 5% of downtown commuters today come by walking or biking, because they actually live in downtown apartments.
If this thing triples the radius outward from downtown that is "walkable", then it increases the number that can come without a car by NINE fold, no? That's a simplistic computation, but the LONG term effect will be to make more apartment buildings go up further away from downtown because you'll be able to "walk to work".
Oh, and the other "square footage" issue only a few have alluded to is that these will probably be allowed in public transit (in my Calgary hometown, bikes are only allowed on the trains, and only off-peak hours), thus increasing public transit usage, and will surely be allowed into your cubicle/office, (doubles as your coatrack), so the "stolen bike" issue will be reduced.
It'll certainly succeed in "medical need" and niche roles - my Dad who needs a hip replacement will get one for sure - he doesn't need a wheelchair but any walk over 200 yards is torture. And we'll see what else it does after that!
You aren't really specific about whether you most want to help in places in need (in which, yes, you'll surely have to pick some other contribution than I.T.) or just go abroad with the talents you've got.
Many commentators say the best bang for the 1st world's development bucks will come from helping nations that are hurting but are also developing - to help tip the balance towards upward progress.
In which case, I think you don't even have to travel far. Learn Spanish and do some travelling south of the Rio Grande and I think you'll find Mexico and other latin american countries are in endless need of teachers and NGO office volunteers that know their way around a computer.
With Linux, customers "end up being in the operating systems business," managing software updates and security patches while making sure the multitude of software packages don't conflict with each other," Miller said. "That's the job of a software vendor like Microsoft."
==============
In short, the REAL headline for this story is:
"Microsoft admits is was THEIR job, not customers' job, to apply all those IIS security patches, and thus admits responsibility for all damage from the Code Red and Nimda worms."
Microsoft's compensation cheques are presumably coming out next week.
On the other hand, if you admit your listening tends to be mostly casual (background while you work or party) and it's quite rare that you sit still in an otherwise quiet room, intently listening with your eyes closed to great speakers and equipment, then for God's sake, go with "popular". Nothing's more embarrassing than somebody carefully buying the best caviar, then eating it with mustard on Wonderbread.
I have 3000 MP3 files ripped at 128 from my own collection, the library, etc, used for background music, working at the computer, (noisy) parties, and my portable MP3 player that would only get a half hour if I used a high bit-rate. They're fine for those purposes.
If I want a better listening experience, it has to come from a quiet living room and my full stereo - in which case I have or can get the original CD!
Wasn't there a whole other variable, as well? The material of the coin?
I recall that guineas weren't just an extra shilling compared to a pound, they were GOLD - and professionals & gentlemen took payment in gold, by preference.
And aren't you missing a unit? Sherlock Holmes stories keep mentioning a "sovereign" and I think that was a gold coin, too...but how many shillings was it worth?
...is the price of their tools.
The Romance of the Slide Rule is very much alive today, the love affair is just transferred to the modern version of the tool.
Slashdotters and Tom's Hardware afficionados who explore every capacity and foible of their machines are following the same urge as slipstick experts who practiced use of the LL/0 scale (which did an e^(-x) operation in one move) just so they were prepared should it come up one day.
I must be almost the same age as "timothy" who started this article - it was slide rules for us up to about grade 12, when the richer kids were all getting calculators that did more than 4 functions...then in University, calculators were suddenly essential (but I carried the Rule to exams in case of battery failure).
What I remember is that most geeky kids were old enough to *learn* to use a slide rule at the start of their teens...right around the time you get coordinated & generally grown-up enough to be allowed to touch precision tools. And slide rules, with their fine machinining, high cost, and smooth movement (metal sucked; bamboo rules...mine still work perfectly after 25 years in the drawer) were clearly *TOOLS*, not toys.
A slide rule was a grown-up possession, a minor Rite of Passage.
I don't think slide rules will ever be forgotten as long as dad's, grandad's and great-granddad's get found in attics. But what worries me is my Dad also taught me dozens of MENTAL calculation tricks that nobody needs any more.
The easiest one was squaring number that ends in five. Chop off the five. Multiply the rest by one greater than itself; tack 25 on the end. (i.e. 35 squared: 3*4=12, the answer is 1225.) The hardest was memorizing ten anti-logs to three places; you can do remarkable estimation-level calcs with that one. (I've long forgotten them.)
Anybody ever done a web page of those tricks?
GIS just involves a lot of data - and a lot of problems cross-referencing separate but related datasets. In our case, its the (say) water, sewer, parks, roads and legal maps (and 22 others) that are all drafted separately but must register on top of one another correctly to create synergy.
I've gone from one elegant-but-buggy GIS product (VISION*) to just doing "GIS" by connecting my CAD maps (Microstation) to databases, to a second run at GIS with the Microsoft of the GIS world (ESRI). And they were right; the technology was never my big problem. Staying on top of the ever-mutating dataset is 98% of the work.
I see GIS having big effects in the government and industrial world but almost none at the consumer level, because you have to be a big organization to manage enough *data* to need a GIS to understand it all.
There will be consumer products and services - your PDA will show you the direction to the nearest Thai restaurant - but the only technology you'll need to get this from some remote GIS engine will be a browser.
Oh, and PS, that bodes ill for Open Source GIS software outside the academic world, because big organizations have a positive fondness for the "Microsofts" of any industry they buy from, and an aversion to "unsupported" products. Hate to be the messenger, but its just a corporate-culture fact of life.
I don't have the Japan-knowledge creds of others saying this, but even the most cursory knowledge of the island argues against it.
Skip that they are a "feudal" culture of loyalty to The Company, with a near-permanent ruling party, etc ... that stuff can change. They've made bigger changes. (Playing catch-up to make them...hardly "clicks down the road".)
No, let's look at deeply-held cultural traits that aren't about to go away. They are genetically and culturally homogenous. Differences from the mainstream are not well-liked. Foreigners are not popular.
By the way, in this, they are NOT like England, which is an often-conquered mish-mash of natives, North Europeans, Norman French, and a Roman gene or twelve. (And the English language reflects it.)
And "The Future" is surely one of increasing diversity and inter-mingling. The plummeting cost of travel and the advantages of trade demand it.
If Japan even joins that future, much less makes it, they'll have to radically change deeply-held attitudes to do so.