The implementation of a protocol must be robust. Each implementation must expect to interoperate with others created by different individuals. While the goal of this specification is to be explicit about the protocol there is the possibility of differing interpretations. In general, an implementation must be conservative in its sending behavior, and liberal in its receiving behavior. That is, it must be careful to send well-formed datagrams, but must accept any datagram that it can interpret (e.g., not object to technical errors where the meaning is still clear).
In the USA at least, commercial airline travel is much much safer than riding in a conventional automobile.
And yet people don't care. They think air travel is dangerous but thinking nothing of their cars that kill 30,000 per year and injure millions per year. In terms of human life, there's a WTC catastrophe *every month* on the highways.
So it's not about safety. It doesn't matter whether an automatic system is safer than a human-controlled system or not. People want contro and don't actually care about safety.
Here is what happened: professional traders like to have keys that execute transactions really damn fast, with no confirmation. To put it in Slashdot terms, they are low-ping bastards. When you're firing a BFG or a rocket launcher, you don't want an "are you sure?" dialog popping up. Same deal with some of these traders.
So this guy at Solomon Brothers had a new trading station (only 10 days old), and the new trading station had a feature where the F12 key repeats the last order, and the guy was looking at one of his other screens, and leaned on the F12 key... auto-repeat... and the next thing you know, Salomon Brothers is unwittingly offering $880 million of French bond futures.
About 1/4 of the orders got cancelled before they were executed. Salomon disputed the rest of the trades, but the French exchange officials denied them and ruled that the trades stood. Salomon turned off the new trading systems that same day and went back to using telephones for a while.
No word on how much money Salomon lost buying back those $880 million face value of French bond futures. The article does say that Salomon's inadvertant trades depressed the value of the French bond futures by 1.4%, so Salomon probably lost about $10 million buying them back at the normal price (1.4% of $880 million).
(1) Cost of catastrophe: $1,000,000. (2) Chance of catastrophe: 5% (3) Cost of setting up parallel system, including hardware, software licenses, system administration: $250,000.
If (1) times (2) is less than (3), then it's actually better not to spend the money on (3).
Of course you can argue with the actual numbers in (1) (2) (3). (1) is the Tribune's own estimate. (2) is estimable by looking at the history of past projects, I'm just guessing 5%. And I just pulled (3) out of the air.
That said, I bet they do have degraded capacity, and that they used it to print half their papers on Monday and all their papers on Tuesday.
I believe the 1,000 highest-paid engineers in the United States make more than the 1,000 highest-paid athletes.
At his peak, Michael Jordan made about $30 million per year.
David Filo owns $1.3 billion of Yahoo stock. That's 40 years of Michael Jordan's peak earnings. Jerry Yang owns $2 billion of Yahoo stock. That's 60 years of Michael Jordan's peak earnings. And that's at todays prices, not the bubble prics.
And there's a lot more Filos and Yangs out there than there are Jordans.
I think the mistake here is to compare "rich famous athlete" with "average engineer". Try comparing "rich athletes" with "rich engineers".
"... who contribute NOTHING to society..."
Oh, are you talking about John Carmack?
It's not your place to decide whether other people are getting value for their money when they CHOOSE to watch Michael Jordan or buy stuff with his name and picture on it.
I don't want better pictures. I want to pick up my camera, point it at something, press the clicky button. Then I want to plug the camera into my computer and have all the pictures pop up on my screen. Then I want to email Mom and attach a few pictures.
There's actually a profession of people to figure out how many customers are like you ("want better pictures") and how many are like me ("want simplicity of operation"). It's called "marketing".
1 year == (365.25 days)(24 hours/day) == 8766 hours. 1 megawatt == 1000 kilowatts, of course. 10 megawatt-years == 87,660,000 kilowatt-hours. The article says that the turbines are slack for 6 hours out of every 24. My Con-Ed bill says that kilowatt-hours cost $0.16 delivered, retail price. The NYT article says that electricity costs $0.05 per kilowatt-hour, wholesale.
So that's $3,300,000 per year of tasty electricity. About 0.1% of what NYC uses, according to the article.
Some of Heinlein's predictions sneak in as part of the detail.
Heinlein wrote about US troops in southeast asia in 1963 (Glory Road). And in Citizen of the Galaxy, one of the characters takes a tube transport under the bay. Hey, I've taken BART under the Bay... but BART opened in 1971/1972, and Citizen of the Galaxy was published in 1957!
The book I always wanted Heinlein to write: a completely non-SF book about Ira Johnson's adventures in his youth.
... and I just finished Steppenwolf by Herman Hesse, published 1927. Hesse won the Nobel Prize for literature.
The book cost me $1 or less (that's what I'll pay for a book in NYC, generally). I don't remember if I got it from a library book sale, or a used book store remainder cart, or some apartment's book-swap shelf in the basement next to the laundry, or the giant institutional rummage sale across the street from me several times a year (I picked up 250 books there once for $80, mostly vintage 60's and 70's science fiction), or from one of my friends, or...
New York City is full of cheap books. They're everywhere. I got my copy of Cryptonomicon for 75 cents at a thrift shop. After I read it, I gave it to my friend, and before he could read it, his roommate glommed onto it. Three readers, one Cryptonomicon, $0.25 each, beat that!
I don't have a lot of sympathy if you need to read the hot novel of the month. If you need to read a current technical or non-fiction book, and you're not comfortable hanging out in Barnes and Noble (why do you think they provide chairs?), that's one thing. But really, you have hundreds of years of great books to choose from, focussing on the past three months is myopic.
Turn the question around. Instead of asking "why buy $1000 3.4 GHz chip instead of $500 3.2 GHz chip", ask this: how come Intel doesn't label that second chip "3.4 GHz" and sell it to you for $1000?
That would be $500 extra revenue for Intel. How come they don't do it?
Perhaps it's something to do with increased failure rates, warranty returns, and a negative hit on their reputation.
Intel grades their chips. They mark each chip with the speed that they feel comfortable selling with Intel's name and warranty.
If you want to overclock your chip, it's your chip; you bought it, you didn't license it with a stupid EULA! But the problem comes when another company (not Intel) buys a $500 chip from Intel, overclocks it, and then marks up the price and sells it in a system as if it were a $1000 chip. Intel gets nothing all the trouble, the cheater company gets the markup, and the end user gets the shaft.
I'm not an expert in this field, but it seems to me that there are two ways to get six years of uptime.
(1) A single highly-engineered machine (yeah I know VAX/VMS has clusters, whatever they are). (2) Redundant cluster of many interchangeable parts.
Google has figured out how to do (2) successfully.
I bet that (2) is harder than it looks. How do you protect against a common mode failure in your system software? Do you run a variety of genetically independent OS's and databases's, or do you run identical software on each machine, leaving you open to monoculture failures?
Digression: It's beautiful how eukaryotic organisms solve this problem by having two independent copies of each gene. But if a gene is broken, it generally does nothing rather than produce a lethal result. And the robustness of individual eukaryotes is not enough for the requirements of computers.
Notice that all the tag lines that people quote and use for parody are from the first movie?
"No one can tell you what the XXX is. You have to see it for yourself." "There is no XXX". "No lieutenant, your XXX are already dead." "Whoa." (okay, of course that is from earlier Keanu Reeves movies, but it's only funny in the first Matrix movie). "Humanity is a virus... I am the cure." "Misssteeer Annnndeerrrrsonnn...." "Ever have the feeling..." "It's called LSD. It's the only way to fly." "Guns. Lots of guns." The red pill / blue pill speech. "How did they know what chicken tastes like?"
And on and on.
I don't ever see anybody quoting lines from the second or third movies. "Choice is an illusion between those with power and those with not." -- too pretentious. Smith almost had something going with his "purpose" speech, but then the scene morphed into that crappy video game clip. One good line is "You never really know someone until you fight with them". Another line I like is: "Would that include a bullet from this gun.", but by then the crappy dialogue had gone on so long that the momentum was lost.
My favorite line: when Trinity says: "Neo, I had to" (after he catches her). That line actually ties into the "free will" theme in a thought-provoking way.
As for Matrix Revolutions, urgh. One promising but forgettable speech about love in the subway. And the whole hyper-brawl was a setup for Neo to deliver a stunning epiphany to sum up the whole trilogy, and he says something really lame. He might as well have stuck with "woah".
The traffic is moving 15 mph or less. The delta between you and the traffic is 15 mph or less. You don't cross the white line (you are sharing the lane).
That's what it was the last time I checked the California Vehicle Code. Yes, I used to ride motorcycles in California. Yum!
It's been a while though, so the laws may have changed.
For instance, Brooks talks about the importance of having a dedicated system administrator. The reviewer makes fun of this, pointing out that he has his own dedicated computer, and he doesn't need any administration.
Sure, fine, the desktop computer doesn't need someone to mount tapes on it. But someone's still gotta take care of the spam filter, the firewall, the on-site backups, the off-site backups. There's still plenty of admin work to do.
Similarly for dedicated systems. "Ha, ha! The sort team gets a dedicated 4-to-6-hour slot! How quaint!" Well, the scarce resource is no longer a machine with a compiler/linker/debugger. But how about the embedded target machines? Does every developer have one, or are there two of them in the lab, and one of them is broken, and the other effectively belongs to the alpha geek who's camping it?
Brook's comments about adding developers and increased communication costs also foresages something that happens on many open source projects: random patches arrive faster than anyone can deal with them, leading to huge backlogs for patch review.
I could go on. But enough from me; go read your Brooks again!
It is astonishing that the judge seems to suggest that by not reading a contract, one is not bound to its terms!
I read the handy PDF link, and I don't think the judge is suggesting that at all.
The judge is saying that the plaintiffs failed to claim that they had read Northwest's privacy policy; therefore, the plaintiffs could not have relied on the privacy policy. To me, it looks more like a deficiency in the plaintiff's pleading than anything else. It would be like, say, suing a store for cheating you on an advertised price from a newspaper, but failing to claim that you had read the newspaper ad. It doesn't mean that privacy policies (or newspaper ads) are not binding; it just means that if a plaintiff wants to sue somebody for violating their policy/ad, the plaintiff better get their ducks in order and state that they read the policy/ad and then relied on it for their purchase decision.
Later on in the memo, this judge does slag the privacy policy on other grounds, though.
I have to agree with you, from reading the judge's memo, it looks like the plaintiffs left several critical elements out of their complaint. I'm saying this without actually seeing the complaint though -- just the judge's reaction to it.
Yeah, I think SCO-Novell suit is still up in the air, too.
On SCO's side, I think they will argue that the whole point of APA-2 is to transfer copyrights. It's not like OldSCO said "hey, let's do a deal where we give you a bag of money, and you give us hot air that means nothing. That's the deal we intend to make."
On Novell's side, I think they will argue that APA-2 is just a preliminary agreement and does not actually transfer any copyrights until, say, SCO makes a demand for a specific copyright that they need to own.
My view is that the original lawyers who wrote APA-2 screwed up. APA-2 should be more explicit about which copyrights transfer and when. I'm just an amateur about this, but I wonder if the judge is P.O.'ed that SCO and Novell have a vague contract and now they are relying on a court to tell them both what the agreement is.
That's mostly similar to my view, but there's a subtle and important difference.
It's not that SCO "failed to prove" special damages. That has the connotation that SCO came to court and presented its evidence and the court didn't believe their side. Instead, the situation is: SCO didn't explicitly list the special damages, so the judge told them they had to file a new complaint with an explicit list of special damages. When the court gets the new filing, THEN it will proceed to hear motions and evidence about SCO's special damages.
SCO just has to come back with a new filing where they claim exactly how they were damaged, and the suit proceeds from there.
To draw an analogy... say my car gets wrecked and I file a claim on my insurance company. The insurance company comes back and says "you need to attach a police report and some photographs before we decide whether to pay your claim". That's different from "we don't believe your car suffered any damage, go pound sand".
I can't get to the e-week article but I have read the 29 pages of the judge's order.
I know it's unpopular to say anything good about TSG, and I hate TSG as much as the next slashdotter. More, probably. But to me, this opinion does look positive for TSG.
The court denied Novell's motion to dismiss the case, and then granted TSG leave to re-file their case with specific information on special damages.
Damn, this was some heavy reading! Here's my armchair legal analysis.
First, Novell sold a lot of Unix(tm) intellectual property rights to TSG. Novell and TSG signed a contract for this, the Asset Purchase Agreement (APA). Later on, Novell and TSG signed an amedement, APA Amendment 2 (APA-2). Dunno whatever happened to APA Amendment 1.
The original APA says that no copyrights are transferred as part of the sale. APA-2 says that the sale does include some copyrights -- whatever copyrights that TSG needs to enforce its other rights for the property that they paid good money for.
Fast-forward to 2003. TSG starts its campaign: "we own the Unix copyrights. Pay us $$$$$$$$$." Novell puts out its own press releases: "actually, we (Novell) still own the actual copyrights. You don't have to listen to TSG".
TSG gets pissed off about this, says that Novell is lying about TSG's Unix copyrights and that Novell is fucking with TSG's business of selling licenses to those copyrights. TSG sues Novell about this.
TSG: "We paid for those copyrights, see APA-2"
Novell: "No, actually, APA-2 says that we promise to give you whatever copyrights you need later, APA-2 doesn't actually transfer specific copyrights."
The case ends up in Federal court, Judge Kimball. TSG wants the case to be in State court. Novell wants the case to stay in Federal court.
Kimball says: "this case is about whether APA-2 actually transferred the copyrights or not. That's a federal issue. So it stays here in federal court."
Next, Novell says: "it's so CLEARLY OBVIOUS that APA-2 does not transfer copyrights that it's okay for us to state publicly that TSG doesn't own the copyrights. Please tell TSG to stuff their lawsuit and go screw."
Kimball says: "not so fast, Novell. It's not OBVIOUS at all. Maybe APA-2 actually transferred the copyrights but MAYBE NOT. We're going to need a trial to figure that out. Since we're going to need a trial, I'm not going to dismiss the lawsuit on those grounds at this stage."
Next, Novell says: "and oh yeah, TSG's lawsuit is deficient because they weren't specific about how they were damaged -- just because they are trying to license this Unix(tm) property, and we issue press releases pissing all over the idea that Unix(tm) is THEIR property after they paid $100 million for it, that's not enough. See FRCP mumble."
Judge Kimball says: "Novell's got a point. SCO, your legalese has syntax errors and fails to validate. I'm not saying your content is bogus, I'm saying your syntax is wrong. You have 30 days to re-format your lawsuit so that it's valid FRCP".
So, the deal is:
TSG can continue to sue Novell for dumping on TSG's claims that TSG owns the Unix copyrights. TSG must pursue this lawsuit in federal court, because it's a federal issue whether those copyrights actually transferred or not.
Novell's claim that "Novell still owns the copyrights" might be legally potent or it might not. It will take a trial to determine this. But Novell can't make TSG's lawsuit go away at this stage just by claiming this. It might be true, but it's not OBVIOUS that it's true.
TSG has the right to sue Novell for slandering TSG's title to Unix(tm). But their current complaint is defective. They have 30 days for their lawyers to submit a new, more specific complaint. Then the case will proceed, in federal court, and the court can actually do some more work on the question of whether APA-2 transferred the copyrights or not.
It's not a level comparison: "Darl" versus "average computer geek".
For a more level comparison, try "Darl" versus "average computer company executive with similar managerial talent and resources". Say, Darl versus Bob Young.
To be sure, Darl's made some megabucks, but I think that CEO's at companies which make stuff that customers like to buy have been doing much better than Darl.
Slashdot Mirror
That goes for people protocols, too.
In the USA at least, commercial airline travel is much much safer than riding in a conventional automobile.
And yet people don't care. They think air travel is dangerous but thinking nothing of their cars that kill 30,000 per year and injure millions per year. In terms of human life, there's a WTC catastrophe *every month* on the highways.
So it's not about safety. It doesn't matter whether an automatic system is safer than a human-controlled system or not. People want contro and don't actually care about safety.
... is good for the country!
Factoid on relative sizes of the two organizations:
SCO revenues, last 12 months: $61 million.
Daimler-Chrysler revenues, last 12 hours: $230 million.
That's not a computer error. Nick Leeson executed the trades he intended to execute. They just didn't work out, though.
... auto-repeat ... and the next thing you know, Salomon Brothers is unwittingly offering $880 million of French bond futures.
This is a computer error.
Here is what happened: professional traders like to have keys that execute transactions really damn fast, with no confirmation. To put it in Slashdot terms, they are low-ping bastards. When you're firing a BFG or a rocket launcher, you don't want an "are you sure?" dialog popping up. Same deal with some of these traders.
So this guy at Solomon Brothers had a new trading station (only 10 days old), and the new trading station had a feature where the F12 key repeats the last order, and the guy was looking at one of his other screens, and leaned on the F12 key
About 1/4 of the orders got cancelled before they were executed. Salomon disputed the rest of the trades, but the French exchange officials denied them and ruled that the trades stood. Salomon turned off the new trading systems that same day and went back to using telephones for a while.
No word on how much money Salomon lost buying back those $880 million face value of French bond futures. The article does say that Salomon's inadvertant trades depressed the value of the French bond futures by 1.4%, so Salomon probably lost about $10 million buying them back at the normal price (1.4% of $880 million).
Where was the phased or parallel deployment?
Probably in the hands of someone who decided:
(1) Cost of catastrophe: $1,000,000.
(2) Chance of catastrophe: 5%
(3) Cost of setting up parallel system, including hardware, software licenses, system administration: $250,000.
If (1) times (2) is less than (3), then it's actually better not to spend the money on (3).
Of course you can argue with the actual numbers in (1) (2) (3). (1) is the Tribune's own estimate. (2) is estimable by looking at the history of past projects, I'm just guessing 5%. And I just pulled (3) out of the air.
That said, I bet they do have degraded capacity, and that they used it to print half their papers on Monday and all their papers on Tuesday.
I believe the 1,000 highest-paid engineers in the United States make more than the 1,000 highest-paid athletes.
..."
At his peak, Michael Jordan made about $30 million per year.
David Filo owns $1.3 billion of Yahoo stock. That's 40 years of Michael Jordan's peak earnings. Jerry Yang owns $2 billion of Yahoo stock. That's 60 years of Michael Jordan's peak earnings. And that's at todays prices, not the bubble prics.
And there's a lot more Filos and Yangs out there than there are Jordans.
I think the mistake here is to compare "rich famous athlete" with "average engineer". Try comparing "rich athletes" with "rich engineers".
"... who contribute NOTHING to society
Oh, are you talking about John Carmack?
It's not your place to decide whether other people are getting value for their money when they CHOOSE to watch Michael Jordan or buy stuff with his name and picture on it.
"The fact is if you want better pictures ..."
Stop right there.
I don't want better pictures. I want to pick up my camera, point it at something, press the clicky button.
Then I want to plug the camera into my computer and have all the pictures pop up on my screen.
Then I want to email Mom and attach a few pictures.
There's actually a profession of people to figure out how many customers are like you ("want better pictures") and how many are like me ("want simplicity of operation"). It's called "marketing".
1 year == (365.25 days)(24 hours/day) == 8766 hours.
1 megawatt == 1000 kilowatts, of course.
10 megawatt-years == 87,660,000 kilowatt-hours.
The article says that the turbines are slack for 6 hours out of every 24.
My Con-Ed bill says that kilowatt-hours cost $0.16 delivered, retail price. The NYT article says that electricity costs $0.05 per kilowatt-hour, wholesale.
So that's $3,300,000 per year of tasty electricity.
About 0.1% of what NYC uses, according to the article.
Some of Heinlein's predictions sneak in as part of the detail.
... but BART opened in 1971/1972, and Citizen of the Galaxy was published in 1957!
Heinlein wrote about US troops in southeast asia in 1963 (Glory Road). And in Citizen of the Galaxy, one of the characters takes a tube transport under the bay. Hey, I've taken BART under the Bay
The book I always wanted Heinlein to write: a completely non-SF book about Ira Johnson's adventures in his youth.
... and I just finished Steppenwolf by Herman Hesse, published 1927. Hesse won the Nobel Prize for literature.
...
The book cost me $1 or less (that's what I'll pay for a book in NYC, generally). I don't remember if I got it from a library book sale, or a used book store remainder cart, or some apartment's book-swap shelf in the basement next to the laundry, or the giant institutional rummage sale across the street from me several times a year (I picked up 250 books there once for $80, mostly vintage 60's and 70's science fiction), or from one of my friends, or
New York City is full of cheap books. They're everywhere. I got my copy of Cryptonomicon for 75 cents at a thrift shop. After I read it, I gave it to my friend, and before he could read it, his roommate glommed onto it. Three readers, one Cryptonomicon, $0.25 each, beat that!
I don't have a lot of sympathy if you need to read the hot novel of the month. If you need to read a current technical or non-fiction book, and you're not comfortable hanging out in Barnes and Noble (why do you think they provide chairs?), that's one thing. But really, you have hundreds of years of great books to choose from, focussing on the past three months is myopic.
Turn the question around. Instead of asking "why buy $1000 3.4 GHz chip instead of $500 3.2 GHz chip", ask this: how come Intel doesn't label that second chip "3.4 GHz" and sell it to you for $1000?
That would be $500 extra revenue for Intel. How come they don't do it?
Perhaps it's something to do with increased failure rates, warranty returns, and a negative hit on their reputation.
Intel grades their chips. They mark each chip with the speed that they feel comfortable selling with Intel's name and warranty.
If you want to overclock your chip, it's your chip; you bought it, you didn't license it with a stupid EULA! But the problem comes when another company (not Intel) buys a $500 chip from Intel, overclocks it, and then marks up the price and sells it in a system as if it were a $1000 chip. Intel gets nothing all the trouble, the cheater company gets the markup, and the end user gets the shaft.
I'm not an expert in this field, but it seems to me that there are two ways to get six years of uptime.
(1) A single highly-engineered machine (yeah I know VAX/VMS has clusters, whatever they are).
(2) Redundant cluster of many interchangeable parts.
Google has figured out how to do (2) successfully.
I bet that (2) is harder than it looks. How do you protect against a common mode failure in your system software? Do you run a variety of genetically independent OS's and databases's, or do you run identical software on each machine, leaving you open to monoculture failures?
Digression: It's beautiful how eukaryotic organisms solve this problem by having two independent copies of each gene. But if a gene is broken, it generally does nothing rather than produce a lethal result. And the robustness of individual eukaryotes is not enough for the requirements of computers.
Notice that all the tag lines that people quote and use for parody are from the first movie?
... I am the cure." ...." ..." "It's called LSD. It's the only way to fly."
"No one can tell you what the XXX is. You have to see it for yourself."
"There is no XXX".
"No lieutenant, your XXX are already dead."
"Whoa." (okay, of course that is from earlier Keanu Reeves movies, but it's only funny in the first Matrix movie).
"Humanity is a virus
"Misssteeer Annnndeerrrrsonnn
"Ever have the feeling
"Guns. Lots of guns."
The red pill / blue pill speech.
"How did they know what chicken tastes like?"
And on and on.
I don't ever see anybody quoting lines from the second or third movies. "Choice is an illusion between those with power and those with not." -- too pretentious. Smith almost had something going with his "purpose" speech, but then the scene morphed into that crappy video game clip. One good line is "You never really know someone until you fight with them". Another line I like is: "Would that include a bullet from this gun.", but by then the crappy dialogue had gone on so long that the momentum was lost.
My favorite line: when Trinity says: "Neo, I had to" (after he catches her). That line actually ties into the "free will" theme in a thought-provoking way.
As for Matrix Revolutions, urgh. One promising but forgettable speech about love in the subway. And the whole hyper-brawl was a setup for Neo to deliver a stunning epiphany to sum up the whole trilogy, and he says something really lame. He might as well have stuck with "woah".
I didn't think it was possible to break Spiderman's web servers!
Can we slashdot his super-strength next?
Jeff Dike started user-mode linux in February 1998
message from jeff
Unified Modelling Language may have existed in early 1998; I first saw it in April 1999. But Unified Modelling Language was a lot smaller back then.
And Google did not exist in February 1998!
These days, when I need to name something, I stick the name in google and check for conflicts.
Sam Waksal, too.
What's legal in California is:
The traffic is moving 15 mph or less.
The delta between you and the traffic is 15 mph or less.
You don't cross the white line (you are sharing the lane).
That's what it was the last time I checked the California Vehicle Code. Yes, I used to ride motorcycles in California. Yum!
It's been a while though, so the laws may have changed.
Also, human beings like to express years in two digits. Class of '42; '57 chevy; summer of '66; and so on.
At least, the human beings near me (American) do. I wonder about people who use other calendars with different year-0 points.
Heck, even the parent poster did this! He's talking about "code in the 70s", not "code in the 1970s".
I got the same impression from the review.
For instance, Brooks talks about the importance of having a dedicated system administrator. The reviewer makes fun of this, pointing out that he has his own dedicated computer, and he doesn't need any administration.
Sure, fine, the desktop computer doesn't need someone to mount tapes on it. But someone's still gotta take care of the spam filter, the firewall, the on-site backups, the off-site backups. There's still plenty of admin work to do.
Similarly for dedicated systems. "Ha, ha! The sort team gets a dedicated 4-to-6-hour slot! How quaint!" Well, the scarce resource is no longer a machine with a compiler/linker/debugger. But how about the embedded target machines? Does every developer have one, or are there two of them in the lab, and one of them is broken, and the other effectively belongs to the alpha geek who's camping it?
Brook's comments about adding developers and increased communication costs also foresages something that happens on many open source projects: random patches arrive faster than anyone can deal with them, leading to huge backlogs for patch review.
I could go on. But enough from me; go read your Brooks again!
It is astonishing that the judge seems to suggest that by not reading a contract, one is not bound to its terms!
I read the handy PDF link, and I don't think the judge is suggesting that at all.
The judge is saying that the plaintiffs failed to claim that they had read Northwest's privacy policy; therefore, the plaintiffs could not have relied on the privacy policy. To me, it looks more like a deficiency in the plaintiff's pleading than anything else. It would be like, say, suing a store for cheating you on an advertised price from a newspaper, but failing to claim that you had read the newspaper ad. It doesn't mean that privacy policies (or newspaper ads) are not binding; it just means that if a plaintiff wants to sue somebody for violating their policy/ad, the plaintiff better get their ducks in order and state that they read the policy/ad and then relied on it for their purchase decision.
Later on in the memo, this judge does slag the privacy policy on other grounds, though.
I have to agree with you, from reading the judge's memo, it looks like the plaintiffs left several critical elements out of their complaint. I'm saying this without actually seeing the complaint though -- just the judge's reaction to it.
(And thanks for teaching me about dicta).
Yeah, I think SCO-Novell suit is still up in the air, too.
On SCO's side, I think they will argue that the whole point of APA-2 is to transfer copyrights. It's not like OldSCO said "hey, let's do a deal where we give you a bag of money, and you give us hot air that means nothing. That's the deal we intend to make."
On Novell's side, I think they will argue that APA-2 is just a preliminary agreement and does not actually transfer any copyrights until, say, SCO makes a demand for a specific copyright that they need to own.
My view is that the original lawyers who wrote APA-2 screwed up. APA-2 should be more explicit about which copyrights transfer and when. I'm just an amateur about this, but I wonder if the judge is P.O.'ed that SCO and Novell have a vague contract and now they are relying on a court to tell them both what the agreement is.
That's mostly similar to my view, but there's a subtle and important difference.
... say my car gets wrecked and I file a claim on my insurance company. The insurance company comes back and says "you need to attach a police report and some photographs before we decide whether to pay your claim". That's different from "we don't believe your car suffered any damage, go pound sand".
It's not that SCO "failed to prove" special damages. That has the connotation that SCO came to court and presented its evidence and the court didn't believe their side. Instead, the situation is: SCO didn't explicitly list the special damages, so the judge told them they had to file a new complaint with an explicit list of special damages. When the court gets the new filing, THEN it will proceed to hear motions and evidence about SCO's special damages.
SCO just has to come back with a new filing where they claim exactly how they were damaged, and the suit proceeds from there.
To draw an analogy
I can't get to the e-week article but I have read the 29 pages of the judge's order.
I know it's unpopular to say anything good about TSG, and I hate TSG as much as the next slashdotter. More, probably. But to me, this opinion does look positive for TSG.
The court denied Novell's motion to dismiss the case, and then granted TSG leave to re-file their case with specific information on special damages.
Damn, this was some heavy reading! Here's my armchair legal analysis.
First, Novell sold a lot of Unix(tm) intellectual property rights to TSG. Novell and TSG signed a contract for this, the Asset Purchase Agreement (APA). Later on, Novell and TSG signed an amedement, APA Amendment 2 (APA-2). Dunno whatever happened to APA Amendment 1.
The original APA says that no copyrights are transferred as part of the sale. APA-2 says that the sale does include some copyrights -- whatever copyrights that TSG needs to enforce its other rights for the property that they paid good money for.
Fast-forward to 2003. TSG starts its campaign: "we own the Unix copyrights. Pay us $$$$$$$$$." Novell puts out its own press releases: "actually, we (Novell) still own the actual copyrights. You don't have to listen to TSG".
TSG gets pissed off about this, says that Novell is lying about TSG's Unix copyrights and that Novell is fucking with TSG's business of selling licenses to those copyrights. TSG sues Novell about this.
TSG: "We paid for those copyrights, see APA-2"
Novell: "No, actually, APA-2 says that we promise to give you whatever copyrights you need later, APA-2 doesn't actually transfer specific copyrights."
The case ends up in Federal court, Judge Kimball. TSG wants the case to be in State court. Novell wants the case to stay in Federal court.
Kimball says: "this case is about whether APA-2 actually transferred the copyrights or not. That's a federal issue. So it stays here in federal court."
Next, Novell says: "it's so CLEARLY OBVIOUS that APA-2 does not transfer copyrights that it's okay for us to state publicly that TSG doesn't own the copyrights. Please tell TSG to stuff their lawsuit and go screw."
Kimball says: "not so fast, Novell. It's not OBVIOUS at all. Maybe APA-2 actually transferred the copyrights but MAYBE NOT. We're going to need a trial to figure that out. Since we're going to need a trial, I'm not going to dismiss the lawsuit on those grounds at this stage."
Next, Novell says: "and oh yeah, TSG's lawsuit is deficient because they weren't specific about how they were damaged -- just because they are trying to license this Unix(tm) property, and we issue press releases pissing all over the idea that Unix(tm) is THEIR property after they paid $100 million for it, that's not enough. See FRCP mumble."
Judge Kimball says: "Novell's got a point. SCO, your legalese has syntax errors and fails to validate. I'm not saying your content is bogus, I'm saying your syntax is wrong. You have 30 days to re-format your lawsuit so that it's valid FRCP".
So, the deal is:
TSG can continue to sue Novell for dumping on TSG's claims that TSG owns the Unix copyrights. TSG must pursue this lawsuit in federal court, because it's a federal issue whether those copyrights actually transferred or not.
Novell's claim that "Novell still owns the copyrights" might be legally potent or it might not. It will take a trial to determine this. But Novell can't make TSG's lawsuit go away at this stage just by claiming this. It might be true, but it's not OBVIOUS that it's true.
TSG has the right to sue Novell for slandering TSG's title to Unix(tm). But their current complaint is defective. They have 30 days for their lawyers to submit a new, more specific complaint. Then the case will proceed, in federal court, and the court can actually do some more work on the question of whether APA-2 transferred the copyrights or not.
It's not a level comparison: "Darl" versus "average computer geek".
For a more level comparison, try "Darl" versus "average computer company executive with similar managerial talent and resources". Say, Darl versus Bob Young.
To be sure, Darl's made some megabucks, but I think that CEO's at companies which make stuff that customers like to buy have been doing much better than Darl.