This reminds me of Seldon's statistical sociology work in the Foundation series (most explicitly in the first book) -- where he expounds that, as a large group, people are statistically very predictible, and reasonably controlable as a result.
Other large scale societal modeling took place with The Club of Rome's Limits to Growth -- It used the SIMULA simulation language to investigate such questions as population growth, resource usage, environmental degradation and capital investment as co-related variables. They came to some very interesting (and even disturbing) conclusions.
Well, it's one way to ensure you'll never get slashdotted.
If you're forced to use MS Basic for client side pages, why not just try using static pages... They still work, you know... and they provide better functionality for people like me...
If nothing else, they claim to be sponsored by NASA. Now, if anybody could arrange an FAA exception, I'm sure that NASA has the connections/rights/knowledge to arrange it.
Anybody here remember the band "Led Zepplin" (think: Stairway to Heaven)? Guess where they got their name from?
In any case, I remember the University of Alberta Civil Engineers used to participate in annual concrete tobbogan races. (the page also makes references to earlier concrete canoe races). The hard part, of course, was dragging the thing up the hill for a second run.:-)
From the article: And yes, this problem has existed for a considerable period of time, with Microsoft automatic updates (starting
with ME and continuing in XP) being the most widespread possibility. But this is the first time we have had a company with such willful ignorance of security (based on their business plan) distributing an autoupdating piece
of code.
Er, uhm. Is he talking about Microsoft here, or the Kaza people??
For those who don't know about "that lame Canadian April Fool's joke.":
A Montreal on-air personality who has a history of doing such things, managed to arrange a phone call with Bill Gates on April 1. He arranged it under the name "Jean Cretien" -- The Canadian prime minister.
After arranging a later meeting with Bill Gates, he made as if he was putting the appointment in his computer....
(paraquote, because I don't have the actual quote)
Jokster:One moment while I put the meeting in my daytimer program....
Ah. This damn computer! it died again. What a stupid computer! What a stupid operating system! It's always breaking!
Who invented this Windows 95, anyways?
Bill Gates: (guilty silence)...
The on-air jokster fessed up to the April Fools joke later on in the conversation.
First, as far as I understand, ML is switching their Unix systems to Linux. Not their MS Windows systems. ....
Frankly, who cares about the desktop. Linux won't 'conquer' the desktop without third party support.....
In the comments in the article, they mentioned that their software would be able to scale everywhere -- including the desktop. This leaves me with the feeling that they're expecting to have Linux on the desktop as well (eventually, if not immediately)
For a whole lot of uses, Linux really is ready for the desktop. It may not be up to snuff for the really esotheric high-end desktop publishing, but I can do most of what I really need with Star Office and Gimp, etc.
Abi Word and Koffice may not be quite there, yet (I'm one or two revs behind the current 'stable' releases, there), but they're getting close.
One thing that I'm definitely missing in Abi and K is the ability to properly scale included images.
Please remember that "they can't do it" and "they do it differently" are two different statements. If Linux can easily do something -- but just in a different manner than Windows, then it's not a show stopper. It may even turn out that, once te users get used to doing something the Linux way, that they'll be more happy and productive.
Different is not worse. It's just different.
Far better than going after the Patent Office would be going after the patent requestor for the costs associated with filing for a patent via fraudlent (or at least negligent) representation.
"The patent application failed to disclose a single encryption product. Unreal for 1998," Avritch said
What NASA is doing is somewhat along the lines of insurance. Who here actually expects to have their house burned down, or to get squashed in a plane crash??? Nontheless we pay a small pittance in the hopes that if something like that doeshappen, we'll get money to cover the extraordinary expenses.
In this case, the money spent on this project is rather small, in a NASA budget expense -- but even with a 2% chance of partial success, the amortized savings as a result of even a pointer in the right direction are enough to make the fool's rush more than worth it.
As was vaguely aluded to in the article, the possible PR cost to NASA's credibility was probably more of an impediment to funding this venture than the financial cost.
Think what would have happened if people had refused to fund semiconductor research? I mean, really! Electronics on silicon??? That stuf is almost an insulator!!!
Well, saying "RISC chip without any" registers would have been patently untrue.
For a nice instruction & register set for an 8 bit processor, the 6809 is still my runaway favorite.
The 68000 seemed like a cross between the 360 and the VAX -- another nice, clean processor with a good set of addressing modes.
Accordingo to a friend of a friend, Motorola lost the processor war when they instituted mandatory drug testing among their employees... By the time the recognized the stupidity of that move, they'd lost a number of really good processor designers.
The processors are being sold as things that they are NOT.
Intel is (for better or worse) the benchmark for CPU speeds these days. Athlon is not selling the 2200 as a 2200mz processor, they're selling it as the equivalent to a 2200Mz P4. In terms of informing a customer of how (relatively) fast their CPU will run quake, this is accurate.
Anybody who knows enough to build and install a wall-mount CPU clock meter that actually measures the clock speed is likely to know that the AMD really is equivalent to the 2.2Gz Intel. For the rest of us, the AMD rating is both more informative for the average customer, and less un-flattering to AMD.
For an equivalent to this argument: Imagine if people bragged about what RPM their wheels span at rather then the speed that their car drove at.
If you wanted to really brag, you'd get a 1/4" wheel and run it at 2200RPM (a whopping 1.6 miles/hour). One could argue that this is not unlike what Intel has been doing with the P4 vs the P3/athlon.
Think about it -- they're trying to sell a 1GZ P4 an an entry leve system about a year after the P3/800 was out -- but the year-old P3 (which would have normally been the entry level system by now) would have been faster than the P4 if intel hadn't 'de-emphasized' the P3.
This is why people came out with the dhrystone, whetstone and other benchmarks back in the '80s -- to get comparisons of the relative cpu power across various CPU architectures for which one-for-one CPU clock speeds were entirely inappropriate (e.g. a 4Mz Z80 was about the same speed as a 1Mz 6502 -- mostly becasuse the Z80 took 4 clock cycles to grab a byte of memory while a 6502 only took one).
A friend of mine was supporting a group of a few hundred Wintendos boxes, and he ran into a problem where Excel was corrupting files on a semi-regular basis. When he took this to his assigned MS support rep, he was repeatedly told (over a number of months) "It must be something that you're doing wrong because I haven't been able to find anybody else with the same problem.
One day he was talking to this rep when my friend mentioned that he was talking to person X at company Y.
"Oh, yeah, he's one of my asignees,' interrupted the rep. "I talk to him all the time."
"Oh," replied my friend rather acusingly, "then you know about the problem that they've been having".
(They had been having the same problem for monthes and had been fed the same line by their [this same] MS rep.).
[guilty silence]
Busted!
And for this 'service' we paid thousands of dollars a year on top of the license fees.
One of the best examples of clock speed vs cpu speed was back in the 8 bit days, with two entirely different memory access schemes.. A z80 took 4 clock cycles to access a byte of memory, while a 6809 or a 6502 would take 1.
This meant that a 4MZ Z80 was about the same as a 1MZ 6502 (actually, it'd often be slower because the 6502 was pretty much a RISC chip without the registers). Nonetheless I'd run into people who were absolutely sure that a (1Mz) Apple was much slower than a souped up (4mz) TRS-80.... rong.
You need the sync so that you can get the individual frames as nicely as possible. Once you do that it's gonna take some fun software to convert to 30/25FPSFPS. The first step, though, is getting each frame.
I realize that separately digitizing each frame is the better way to do it, but DV is DV. I think that the real limit for DV is more likely to be the monitor than the camera -- especially for people who haven't yet gone to DV capable televisions.
As for the cost: even for someone who's only looking at selling a few thousand copies of each movie, that's still over a hundred thousand in gross income from a single movie. It's enough to pay for decent quality -- in both time and equipment. -- of course, this only talks about what's technically feasible. When you're dealing with scam artists, you're just as likely to end up with hi-8->cdrom quality as a reasonably well done film->dv transfer. (I mean, why do the work if 75% of your audience isn't going to notice the difference anyways -- it's still going to be better than VHS).
I once had two cats in an area that didn't seem to have many of them. It was rather interesting. They'd bring in a mouse or a bird almost every other day. They'd eat the birds (leaving behind what I cam to call 'bird bits' -- beaks, feathers, etc., but they'd leave the mice for me as 'gifts'.
Then one day, when I was sick, I got up to go to the bathroom and found that they'd left me a bird. I was touched by the gesture... I thanked them heartfully -- and burried the bird.
_____
Then, of course, there was the day that my larger cat brought in a seagull... completely freaked my roommate out.
Or when their mother (they were born to the cat of an earlier incarnation of roommate) brought in a whole pot roast for her kittens (with the string still on). I have no idea where she got a pot roast from, but I'm sure that somebody's barbecue was inexplicably short that day.
They were eating peices off of that pot roast for the better part of a week.
I have a cat door. It allows my cats to go in and out of the house at will. One cat even learned to use a second floor cat door by jumping across from the neighbour's roof.
I'm not up to the idea of waking up at 4AM to let my cat out for 35 minuts before he decides it's too wet out and he want's back in. Better to let him implement his indoor/outdoor policy.
A good "wholly inferior product" is going to be made by digitizing from a copy of the actual film reel. -- I mean why not.... during the 8-12 hours that a movie theatre is closed, hook up a nice film digitizer and let the bastard run to your heart's content until it's time for the morning showing... At the very least, display it on the full screen and use a nice, high-quality DV camera to digitize each frame (synched, appropriately, to the projector). with the sound input jacked direct into the theatre sound system. (and nobody else in the theatre while you're doing it). For the purposes of image quality, this should be indistinguisable from a 'studio' DVD (minus the commentary track, etc.).
In any case.. once the studio comes out with a good 'added features' DVD, there's nothing to stop the pirates from bit-coppying the DVD.
I'm pretty sure that the existance of commercial DVD pirates sans-DECSS should be usable in those cases, though... Shows how CSS does bugger-all to prevent commercial pirating, while DECSS is mostly intended to simply allow people to legally view their DVD on an otherwise non(or ill-) supported OS.
There is, actually, a cost. If I grab image 'X', and it's served by proxy Y, and then I go to check the age of image 'X' later on (to see if I should reload it,) and that query goes to proxy Z, proxy Z is probably going to go grab image 'X' a second time (instead of just telling me that 'X' hasn't changed). If it's a bad implementation (which, it sounds like, is often the case), I'm gonna wait for Z to download the second copy of X before it even bothers to tell me that it's the same age as the copy that I've got.
Also: unless the proxy is doing a packet-by-packet forward of a new image, I'm going to have to wait for the entire image to download to the server before I get it copied to me.... If it's a large image/file, this could be quite a noticable increase in latency.
I actually tried playing with it today, and it appears that it would bd a bit of a problem... If you want to use cachefs, you have two choices:
put in the cache config on EVERY machine you install (hard if you have some true diskless boxes) this is pretty much the only way to define cachefs in the auto.home maps.
configure the cache and modify the automount.master file on machines that have the cache sonfigured. This means having many machines with (somewhat) non-standard configurations..
The problems with the latter are OK where you can do a custom kickstart install (and I've NEVER liked their default disk partitioning, anyways). -- but if you have mount options on the auto.home mapping, this would probably be ignored.
What would be nicest would be if they could define cachefs such that, where the cache is unavailable/corrupt, it gracefully degrades to simply mounting the backing filesystem directly (and possibly complains loudly about it). At least that way, you'd have continuing functionality and backwards compatibility.
Any network or server issues, and everybody's in trouble. One amusing day, network traffic slowed to a crawl...... everybody found their editor stopping for about five minutes every time they did something that loaded a module.
Can't you get around a lot of this by using autoclients and cachefs? I haven't tried using cachefs with automounter, but I'd expect that it'd work. That way most of the users would have a local cache of their often-used data, program and modules, but an update would be automagically propagated across the universe.
(of course, propagation of such an update might have been what caused the massive network slowdown that you reported....)
In many cases, the great scientific breakthrough will point people into a general direction that produces something better than the original breakthrough...
(think of basic research as a bunch of blind men trying to hit a bullseye... The breakthrough is hitting the board. Once people know where the board is, someone else is likely to actually hit the bullseye -- so you hear about the person who first hits the board, and the person who hits the bullseye, but it's rare that the connection between the two events make it through the "15 seconds of fame" filter of media editing.
then one day, you find a house where the door is unlocked, and the house is armed.
Er, um.. that should have been "and the house (door) is alarmed. This actually happened to me once, when I was trying to find my way out of a place where I was doing some late night admin work.. I tried a door that turned out to be an entrance to a neighbor's space. The door was unlocked, but had a chain on it and an alarm (which was set up).
It was kinda half-amusing the conversation I had with the police when they arrived...
Are you authorized to be in there?
yes.
then can you come out and talk to us for a while?
no.
why not?
I don't have a key.
are you sure you're authorized to be there?
Yes. I've called the owner, he's on his way. . . . .
So if I never saw it, and didn't know that I had received it, how does it count as being served? They might as well have sent it to "General Delivery" in a large city that I once visited.
One of the other options is to simply post it in the clerk's area of the courthouse. How often do you visit the courthouse of the last city where you lived?
For alternative service to be accepted, (I hope that) they have to show that it has a reasonable chance of succeeding (better, at least, than posting it in the courthouse). In this case, it was (presumably) emailed to an address that the defendants openly used, and was associated with their alleged activities.
It's not so much a problem that they launched an investigation. Investigating an action that brought your server down for a day is probably a good thing.
The problem was that their 'investigation' was a bit on the "shoot first and question the cadaver later" nature. Their first step should have been to spend some time figuring out what happened and the nature of the apparent 'attack'. Had they done that, I think that they would have realized that the 'attack' was quite possibly a mistake and/or the result of a problem with the configuration of their box.
In this case, it appears that one of the first things they did was to issue a search warrant.
As far as I'm concerned, a search warrant should not be issued lightly. I think that both the police department and the judge who signed the warrant should get a (virtual) public flogginh over their actions under these conditions.
As for analogies, I think that a closer one of someone going around the neighbourhood checking for unlocked doors and informing the owner of the insecure box...
then one day, you find a house where the door is unlocked, and the house is armed.
So why didn't you send this information to the local newspaper? Seems to me the voters would love to see what a foul-mouth guy this "Jeff Darga" allegedly is.
Slashdot Mirror
Other large scale societal modeling took place with The Club of Rome's Limits to Growth -- It used the SIMULA simulation language to investigate such questions as population growth, resource usage, environmental degradation and capital investment as co-related variables. They came to some very interesting (and even disturbing) conclusions.
If you're forced to use MS Basic for client side pages, why not just try using static pages... They still work, you know... and they provide better functionality for people like me...
If nothing else, they claim to be sponsored by NASA. Now, if anybody could arrange an FAA exception, I'm sure that NASA has the connections/rights/knowledge to arrange it.
Anybody here remember the band "Led Zepplin" (think: Stairway to Heaven)? Guess where they got their name from?
In any case, I remember the University of Alberta Civil Engineers used to participate in annual concrete tobbogan races. (the page also makes references to earlier concrete canoe races). The hard part, of course, was dragging the thing up the hill for a second run. :-)
Er, uhm. Is he talking about Microsoft here, or the Kaza people??
A Montreal on-air personality who has a history of doing such things, managed to arrange a phone call with Bill Gates on April 1. He arranged it under the name "Jean Cretien" -- The Canadian prime minister.
After arranging a later meeting with Bill Gates, he made as if he was putting the appointment in his computer....
(paraquote, because I don't have the actual quote)
The on-air jokster fessed up to the April Fools joke later on in the conversation.Frankly, who cares about the desktop. Linux won't 'conquer' the desktop without third party support.....
In the comments in the article, they mentioned that their software would be able to scale everywhere -- including the desktop. This leaves me with the feeling that they're expecting to have Linux on the desktop as well (eventually, if not immediately)
For a whole lot of uses, Linux really is ready for the desktop. It may not be up to snuff for the really esotheric high-end desktop publishing, but I can do most of what I really need with Star Office and Gimp, etc.
Abi Word and Koffice may not be quite there, yet (I'm one or two revs behind the current 'stable' releases, there), but they're getting close.
One thing that I'm definitely missing in Abi and K is the ability to properly scale included images.
Please remember that "they can't do it" and "they do it differently" are two different statements. If Linux can easily do something -- but just in a different manner than Windows, then it's not a show stopper. It may even turn out that, once te users get used to doing something the Linux way, that they'll be more happy and productive.
Different is not worse. It's just different.
In this case, the money spent on this project is rather small, in a NASA budget expense -- but even with a 2% chance of partial success, the amortized savings as a result of even a pointer in the right direction are enough to make the fool's rush more than worth it.
As was vaguely aluded to in the article, the possible PR cost to NASA's credibility was probably more of an impediment to funding this venture than the financial cost.
Think what would have happened if people had refused to fund semiconductor research? I mean, really! Electronics on silicon??? That stuf is almost an insulator!!!
For a nice instruction & register set for an 8 bit processor, the 6809 is still my runaway favorite.
The 68000 seemed like a cross between the 360 and the VAX -- another nice, clean processor with a good set of addressing modes.
Accordingo to a friend of a friend, Motorola lost the processor war when they instituted mandatory drug testing among their employees... By the time the recognized the stupidity of that move, they'd lost a number of really good processor designers.
Intel is (for better or worse) the benchmark for CPU speeds these days. Athlon is not selling the 2200 as a 2200mz processor, they're selling it as the equivalent to a 2200Mz P4. In terms of informing a customer of how (relatively) fast their CPU will run quake, this is accurate.
Anybody who knows enough to build and install a wall-mount CPU clock meter that actually measures the clock speed is likely to know that the AMD really is equivalent to the 2.2Gz Intel. For the rest of us, the AMD rating is both more informative for the average customer, and less un-flattering to AMD.
For an equivalent to this argument: Imagine if people bragged about what RPM their wheels span at rather then the speed that their car drove at. If you wanted to really brag, you'd get a 1/4" wheel and run it at 2200RPM (a whopping 1.6 miles/hour). One could argue that this is not unlike what Intel has been doing with the P4 vs the P3/athlon.
Think about it -- they're trying to sell a 1GZ P4 an an entry leve system about a year after the P3/800 was out -- but the year-old P3 (which would have normally been the entry level system by now) would have been faster than the P4 if intel hadn't 'de-emphasized' the P3.
This is why people came out with the dhrystone, whetstone and other benchmarks back in the '80s -- to get comparisons of the relative cpu power across various CPU architectures for which one-for-one CPU clock speeds were entirely inappropriate (e.g. a 4Mz Z80 was about the same speed as a 1Mz 6502 -- mostly becasuse the Z80 took 4 clock cycles to grab a byte of memory while a 6502 only took one).
A friend of mine was supporting a group of a few hundred Wintendos boxes, and he ran into a problem where Excel was corrupting files on a semi-regular basis. When he took this to his assigned MS support rep, he was repeatedly told (over a number of months) "It must be something that you're doing wrong because I haven't been able to find anybody else with the same problem.
One day he was talking to this rep when my friend mentioned that he was talking to person X at company Y.
"Oh, yeah, he's one of my asignees,' interrupted the rep. "I talk to him all the time."
"Oh," replied my friend rather acusingly, "then you know about the problem that they've been having".
(They had been having the same problem for monthes and had been fed the same line by their [this same] MS rep.).
[guilty silence]
Busted!
And for this 'service' we paid thousands of dollars a year on top of the license fees.
This meant that a 4MZ Z80 was about the same as a 1MZ 6502 (actually, it'd often be slower because the 6502 was pretty much a RISC chip without the registers). Nonetheless I'd run into people who were absolutely sure that a (1Mz) Apple was much slower than a souped up (4mz) TRS-80.... rong.
I realize that separately digitizing each frame is the better way to do it, but DV is DV. I think that the real limit for DV is more likely to be the monitor than the camera -- especially for people who haven't yet gone to DV capable televisions.
As for the cost: even for someone who's only looking at selling a few thousand copies of each movie, that's still over a hundred thousand in gross income from a single movie. It's enough to pay for decent quality -- in both time and equipment. -- of course, this only talks about what's technically feasible. When you're dealing with scam artists, you're just as likely to end up with hi-8->cdrom quality as a reasonably well done film->dv transfer. (I mean, why do the work if 75% of your audience isn't going to notice the difference anyways -- it's still going to be better than VHS).
Then one day, when I was sick, I got up to go to the bathroom and found that they'd left me a bird. I was touched by the gesture... I thanked them heartfully -- and burried the bird.
_____
Then, of course, there was the day that my larger cat brought in a seagull... completely freaked my roommate out.
Or when their mother (they were born to the cat of an earlier incarnation of roommate) brought in a whole pot roast for her kittens (with the string still on). I have no idea where she got a pot roast from, but I'm sure that somebody's barbecue was inexplicably short that day.
They were eating peices off of that pot roast for the better part of a week.
I'm not up to the idea of waking up at 4AM to let my cat out for 35 minuts before he decides it's too wet out and he want's back in. Better to let him implement his indoor/outdoor policy.
In any case.. once the studio comes out with a good 'added features' DVD, there's nothing to stop the pirates from bit-coppying the DVD.
I'm pretty sure that the existance of commercial DVD pirates sans-DECSS should be usable in those cases, though... Shows how CSS does bugger-all to prevent commercial pirating, while DECSS is mostly intended to simply allow people to legally view their DVD on an otherwise non(or ill-) supported OS.
Also: unless the proxy is doing a packet-by-packet forward of a new image, I'm going to have to wait for the entire image to download to the server before I get it copied to me.... If it's a large image/file, this could be quite a noticable increase in latency.
- put in the cache config on EVERY machine you install (hard if you have some true diskless boxes) this is pretty much the only way to define cachefs in the auto.home maps.
- configure the cache and modify the automount.master file on machines that have the cache sonfigured. This means having many machines with (somewhat) non-standard configurations..
The problems with the latter are OK where you can do a custom kickstart install (and I've NEVER liked their default disk partitioning, anyways). -- but if you have mount options on the auto.home mapping, this would probably be ignored.What would be nicest would be if they could define cachefs such that, where the cache is unavailable/corrupt, it gracefully degrades to simply mounting the backing filesystem directly (and possibly complains loudly about it). At least that way, you'd have continuing functionality and backwards compatibility.
Can't you get around a lot of this by using autoclients and cachefs? I haven't tried using cachefs with automounter, but I'd expect that it'd work. That way most of the users would have a local cache of their often-used data, program and modules, but an update would be automagically propagated across the universe.
(of course, propagation of such an update might have been what caused the massive network slowdown that you reported....)
(think of basic research as a bunch of blind men trying to hit a bullseye... The breakthrough is hitting the board. Once people know where the board is, someone else is likely to actually hit the bullseye -- so you hear about the person who first hits the board, and the person who hits the bullseye, but it's rare that the connection between the two events make it through the "15 seconds of fame" filter of media editing.
Er, um.. that should have been "and the house (door) is alarmed. This actually happened to me once, when I was trying to find my way out of a place where I was doing some late night admin work.. I tried a door that turned out to be an entrance to a neighbor's space. The door was unlocked, but had a chain on it and an alarm (which was set up).
It was kinda half-amusing the conversation I had with the police when they arrived...
One of the other options is to simply post it in the clerk's area of the courthouse. How often do you visit the courthouse of the last city where you lived?
For alternative service to be accepted, (I hope that) they have to show that it has a reasonable chance of succeeding (better, at least, than posting it in the courthouse). In this case, it was (presumably) emailed to an address that the defendants openly used, and was associated with their alleged activities.
The problem was that their 'investigation' was a bit on the "shoot first and question the cadaver later" nature. Their first step should have been to spend some time figuring out what happened and the nature of the apparent 'attack'. Had they done that, I think that they would have realized that the 'attack' was quite possibly a mistake and/or the result of a problem with the configuration of their box.
In this case, it appears that one of the first things they did was to issue a search warrant. As far as I'm concerned, a search warrant should not be issued lightly. I think that both the police department and the judge who signed the warrant should get a (virtual) public flogginh over their actions under these conditions.
As for analogies, I think that a closer one of someone going around the neighbourhood checking for unlocked doors and informing the owner of the insecure box...
then one day, you find a house where the door is unlocked, and the house is armed.
verbal moderation: +1 interesting.